Philosophy of Language in the Brentano School: Reassessing the Brentanian Legacy [1 ed.] 9783030522100, 9783030522117

Citation preview

HISTORY OF ANALYTIC PHILOSOPHY

Philosophy of Language in the Brentano School Reassessing the Brentanian Legacy Edited by Arnaud Dewalque Charlotte Gauvry Sébastien Richard

History of Analytic Philosophy

Series Editor Michael Beaney Humboldt University Berlin King’s College London Berlin, Germany

Series Editor: Michael Beaney, Professor für Geschichte der analytischen Philosophie, Institut für Philosophie, Humboldt-Universität zu Berlin, Germany, and Regius Professor of Logic, School of Divinity, History and Philosophy, University of Aberdeen, Scotland. Editorial Board Claudio de Almeida, Pontifical Catholic University at Porto Alegre, Brazil Maria Baghramian, University College Dublin, Ireland Thomas Baldwin, University of York, England Stewart Candlish, University of Western Australia Chen Bo, Peking University, China Jonathan Dancy, University of Reading, England José Ferreirós, University of Seville, Spain Michael Friedman, Stanford University, USA Gottfried Gabriel, University of Jena, Germany Juliet Floyd, Boston University, USA Hanjo Glock, University of Zurich, Switzerland Nicholas Griffin, McMaster University, Canada Leila Haaparanta, University of Tampere, Finland Peter Hylton, University of Illinois, USA Jiang Yi, Beijing Normal University, China Javier Legris, National Academy of Sciences of Buenos Aires, Argentina Cheryl Misak, University of Toronto, Canada Nenad Miscevic, University of Maribor, Slovenia, and Central European University, Budapest Volker Peckhaus, University of Paderborn, Germany Eva Picardi, University of Bologna, Italy Erich Reck, University of California at Riverside, USA Peter Simons, Trinity College, Dublin Thomas Uebel, University of Manchester, England More information about this series at http://www.palgrave.com/gp/series/14867

Arnaud Dewalque Charlotte Gauvry  •  Sébastien Richard Editors

Philosophy of Language in the Brentano School Reassessing the Brentanian Legacy

Editors Arnaud Dewalque Department of Philosophy University of Liège Liège, Belgium

Charlotte Gauvry Department of Philosophy University of Liège Liège, Belgium

Sébastien Richard Department of Philosophy Free University of Brussels (ULB) Brussels, Belgium

ISSN 2634-5994     ISSN 2634-6001 (electronic) History of Analytic Philosophy ISBN 978-3-030-52210-0    ISBN 978-3-030-52211-7 (eBook) https://doi.org/10.1007/978-3-030-52211-7 © The Editor(s) (if applicable) and The Author(s), under exclusive licence to Springer Nature Switzerland AG 2021 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Cover illustration: AA World Travel Library / Alamy Stock Photo This Palgrave Macmillan imprint is published by the registered company Springer Nature Switzerland AG. The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Series Editor’s Foreword

During the first half of the twentieth century, analytic philosophy gradually established itself as the dominant tradition in the English-speaking world, and over the last few decades, it has taken firm root in many other parts of the world. There has been increasing debate over just what ‘analytic philosophy’ means, as the movement has ramified into the complex tradition that we know today, but the influence of the concerns, ideas and methods of early analytic philosophy on contemporary thought is indisputable. All this has led to greater self-consciousness among analytic philosophers about the nature and origins of their tradition, and scholarly interest in its historical development and philosophical foundations has blossomed in recent years, with the result that history of analytic philosophy is now recognized as a major field of philosophy in its own right. The main aim of the series in which the present book appears, the first series of its kind, is to create a venue for work on the history of analytic philosophy, consolidating the area as a major field of philosophy and promoting further research and debate. The ‘history of analytic philosophy’ is understood broadly, as covering the period from the last three decades of the nineteenth century to the start of the twenty-first century, beginning with the work of Frege, Russell, Moore, and Wittgenstein, who are generally regarded as its main founders, and the influences upon them, and going right up to the most recent developments. In allowing the ‘history’ to extend to the present, the aim is to encourage engagement v

vi 

Series Editor’s Foreword

with contemporary debates in philosophy, for example, in showing how the concerns of early analytic philosophy relate to current concerns. In focusing on analytic philosophy, the aim is not to exclude comparisons with other—earlier or contemporary—traditions, or consideration of figures or themes that some might regard as marginal to the analytic tradition but which also throw light on analytic philosophy. Indeed, a further aim of the series is to deepen our understanding of the broader context in which analytic philosophy developed, by looking, for example, at the roots of analytic philosophy in neo-Kantianism or British idealism, or the connections between analytic philosophy and phenomenology, or discussing the work of philosophers who were important in the development of analytic philosophy but who are now often forgotten. The present volume, edited by Arnaud Dewalque, Charlotte Gauvry and Sébastien Richard, illustrates the value of exploring the wider context in which analytic philosophy arose. Franz Brentano (1838–1917) has generally been seen as one of the main influences on the phenomenological tradition, whose main founder was Edmund Husserl (1859–1938), but his influence on some of the early analytic philosophers, too, has been increasingly recognized, especially as mediated by the work of G. F. Stout (see the book published in this series in 2013 by Maria van der Schaar, G. F. Stout and the Psychological Origins of Analytic Philosophy). The present collection of papers aims to elucidate the broader context in which Brentano and members of both the Brentano School and the Brentanian tradition, as they are called, worked on topics that are not only relevant to our understanding of analytic philosophy but may also be regarded as anticipating some of the ideas and concerns of analytic philosophy. What is called the Brentano School comprises Brentano and his students, most notably, Anton Marty (1847–1914), Carl Stumpf (1848–1936), Alexius Meinong (1853–1920), Edmund Husserl and Kazimierz Twardowski (1866–1938). The Brentanian tradition includes members of the various schools that Brentano’s students in turn established, not only phenomenology but also such schools as the Prague School (founded by Marty) and the Lvov–Warsaw School (founded by Twardowski). (On Marty, see Anton Marty and Contemporary Philosophy (2019), edited by Giuliano Bacigalupo and Hélène Leblanc; and on the

  Series Editor’s Foreword 

vii

Lvov–Warsaw School, see The History and Philosophy of Polish Logic (2014), edited by Kevin Mulligan, Katarzyna Kijania-Placek and Tomasz Placek, and Interdisciplinary Investigations into the Lvov-Warsaw School (2019), edited by Anna Drabarek, Jan Wolenski and Mateusz M. Radzki. For discussion on some of the connections between early analytic philosophy and phenomenology, see Judgement and Truth in Early Analytic Philosophy and Phenomenology (2013) edited by Mark Textor. All four volumes have also been published in the present series.) The focus of the present volume is on the philosophy of language. Brentano is not normally seen as having a philosophy of language as such, but Marty certainly did (see the book on Marty just mentioned), and as Part I of this volume shows, Brentano did indeed have views on language and meaning, which he formulated as part of his general concern with mental phenomena. Central to this was his recognition of the way that the grammar of the linguistic expressions we use can mislead us as to the structure of the relevant mental phenomena or thoughts, an idea that anticipates the distinction between grammatical form and logical form that we now take as a key doctrine of early analytic philosophers such as Frege and Russell, and the general project of a ‘critique of language’ especially associated with Wittgenstein. Early analytic philosophers and Brentano may not have agreed about logic, but the comparison between their projects is nevertheless revealing. When we look at the development of Brentano’s thinking in the work of his students and grandstudents, we find a range of differing views, all of which can be profitably investigated in relation to ideas that are now seen as characteristic of certain analytic philosophers. Part I of this volume explores interesting connections between, for example, Brentano and Frege on context-dependence, and Brentano and Ryle on misleading expressions (where Dewalque argues that Ryle is more ‘Brentanian’ than many of the Brentanians themselves). There is also discussion of Marty’s philosophy of language. Part II addresses debates in the Brentanian tradition concerning the objectivity of meaning, and especially the contributions by Roman Ingarden, Twardowski and Meinong. Part III looks at Brentano’s wider legacy, considering some of the most fruitful transformations of his ideas in the work of Husserl, Karl Bühler and Wittgenstein. The result is a wonderfully wide-ranging

viii 

Series Editor’s Foreword

account of the richness of philosophical thinking about language in the Brentanian tradition, broadly conceived, which not only sheds light on the broader context in which analytic philosophy arose but is also fascinating and philosophically instructive in itself. August 2020

Michael Beaney

Contents

1 Introduction: Mind, Meaning and Reality  1 Arnaud Dewalque, Charlotte Gauvry, and Sébastien Richard Part I Brentano and Philosophy of Language  33 2 The Context Principle in Austro-German Philosophy 35 Guillaume Fréchette 3 A Context Principle in Brentano? 57 Charlotte Gauvry 4 Brentano and Mauthner on Grammatical Illusions 77 Denis Seron 5 Misleading Expressions: The Brentano-­Ryle Connection 95 Arnaud Dewalque 6 Sign and Language in Anton Marty: Before and after Brentano119 Hélène Leblanc ix

x Contents

Part II The Brentano School: Act, Meaning and Object 141 7 De Significatione: The Brentano-­Ingarden Axis143 Sébastien Richard 8 Meaning(s) in Roman Ingarden’s Philosophy of Language169 Olivier Malherbe 9 Overcoming Psychologism: Twardowski on Actions and Products189 Denis Fisette 10 Is the Content-Object Distinction Universally Valid? Meaning and Reference in Twardowski and Meinong207 Bruno Leclercq 11 Extensionality/Intensionality in Polish Philosophy of Language: From Twardowski to Ajdukiewicz227 Jan Woleński Part III Brentano’s Wider Legacy 243 12 Modifying Terms and Modification in Husserl and the Brentano School245 Maria van der Schaar 13 The Early Husserl on Typicality263 Hamid Taieb 14 Wundt and Bühler on Gestural Expression: From  Psycho-Physical Mirroring to the Diacrisis279 Basil Vassilicos

 Contents 

xi

15 On Being Guided, Signals and Rules: From Bühler to Wittgenstein299 Kevin Mulligan Index317

Notes on Contributors

Arnaud  Dewalque is professor of contemporary philosophy at the University of Liège, Belgium, and adjunct professor at the University of Luxembourg. He works mainly on Austro-German philosophy and the intersection between phenomenology and analytic philosophy of mind (consciousness, intentionality, judgement). Denis Fisette  is professor in the department of philosophy at Université du Québec à Montréal and he teaches philosophy of mind, phenomenology and the history of philosophy in Austria and Germany. He studied philosophy in Montreal (Université de Montréal), in Germany (Freie Universität Berlin, 1982–1985) and in USA (postdoctoral research at Stanford University, 1987–1989). In recent years, his historical research has focused more specifically on the philosophical programme of Franz Brentano and his successors (including Husserl’s early phenomenology) while his research in the field of philosophy of mind currently bares on emotions and emotional awareness in general. Guillaume  Fréchette  is collaborateur scientifique at the University of Geneva. He works mainly on 19th and early 20th century Austrian and German philosophy.

xiii

xiv 

Notes on Contributors

Charlotte  Gauvry  is a temporary lecturer at the University of Liège, Belgium, and at the University of Lille, France. Her current research program is devoted to the linguistic analyses of the Brentano School. She published several articles and books at the crossroad of analytic philosophy (in particular on Wittgenstein) and early phenomenology. Hélène Leblanc  is a Postdoctoral Research Fellow in the ARC Schol’Art at the GEMCA of the UCLouvain, Belgium. Previously, she worked on the SNF/ANR project SÊMAINÔ at the University of Geneva, Switzerland. She is also a member of Inbegriff—Geneva Seminar for Austro-German Philosophy. She earned her Ph.D. in philosophy (Lille/ Lecce) with a thesis on semiotic theories in Early Modern Philosophy and has published several papers on this topic. Recently, she edited the thematic issue Beauty and Ugliness in the Austro-German Tradition for the Italian journal Paradigmi, and, with Giuliano Bacigalupo, Anton Marty and Contemporary Philosophy, Palgrave Macmillan, 2019. Bruno Leclercq  is professor of logic and philosophy of language at the University of Liège, Belgium. His work in formal semantics uses tools from several intensional logics (possible worlds, centered worlds, Meinongian objects, …) in order to deal with the relation between meaning and reference. Olivier  Malherbe is a scientific collaborator and a PhD from the Université Libre de Bruxelles. His PhD dissertation “L’homme face à ses oeuvres: création et créativité dans la pensée de Roman Ingarden” (“Man and his works: creation and creativity in Roman Ingarden’s thought”) was dedicated to the aesthetical, ontological and anthropological questions related to the topic of creation and creativity in Ingarden’s work. He published several papers on various Ingardenian topics and co-edited in 2016 with Sebastien Richard a collection of articles about the ontology of Roman Ingarden: Form(s) and Modes of Being. The Ontology of Roman Ingarden (P.I.E. Peter Lang). He is also an attorney at the Brussel’s Bar. Kevin  Mulligan  is Honorary Professor of analytic philosophy at the University of Geneva, Ordinary Professor of philosophy at the University of Italian Switzerland and Director of Research at the Institute of Philosophical Studies, Faculty of Theology, Lugano. He writes about

  Notes on Contributors 

xv

ontology, the philosophy of mind and the history of Austrian thought from Bolzano to Wittgenstein and Musil. He is a member of the Swedish and European Academies. Sébastien Richard  teaches philosophy of science at the Free University of Brussels (ULB). He works mainly on metaphysics, logic and philosophy of language in the Brentanian tradition and in analytic philosophy. He has published several papers and books on these topics, among which De la forme à l’être. Sur la genèse historique et philosophique du projet husserlien d’ontologie formelle (Ithaque, 2014), La conception sémantique de la vérité. D’Alfred Tarski à Jaakko Hintikka (Academia Bruylant, 2008) and Objects and Pseudo-Objects. Ontological deserts and Jungles from Brentano to Carnap (edited with Bruno Leclercq and Denis Seron, De Gruyter, 2015). Denis Seron  is lecturer and senior research associate at the University of Liège, Belgium. His research is centered on empiricism, the (recent and older) philosophy of mind, and Austro-German phenomenology, including Brentano and Husserl. Hamid Taieb  is an Alexander-von-Humboldt postdoctoral researcher at the University of Hamburg. His work focuses on the Austro-German and the Aristotelian traditions. He held teaching and research positions at the Universities of Geneva, Lausanne, and Salzburg, and visiting positions at the ENS Ulm, Humboldt University, King’s College London, and the University of Gothenburg. He published a monograph on Brentano’s theory of intentionality and its Aristotelian sources. Moreover, he wrote several papers on themes in the philosophy of mind, epistemology, and philosophy of language, among others, in the Austro-German and/or Aristotelian traditions. Maria Van der Schaar  is Senior University Lecturer at the Institute of Philosophy, Leiden University. Basil Vassilicos  is Lecturer in Philosophy at Mary Immaculate College, University of Limerick, Ireland. His recent research topics have included the philosophy of self-conscious emotions and the role of gestures in speech acts. His current interests lie in enactive theories of consent, the

xvi 

Notes on Contributors

impact of noise on [long-term] communication, and the dialogue between linguistics and biostatistical biology. Jan Woleński  is professor emeritus of Jagiellonian University, professor at the University of Information, Technology and Management in Rzeszów, member of Polish Academy of Sciences, Polish Academy of Arts and Sciences, International Institute of Philosophy, Academia Europea. He is working in the history of analytic philosophy, applications of logic to philosophy, and legal philosophy. He published 35 books, including 10  in English (including, Logic and in Philosophy in the Lvov-­Warsaw School (1989), Semantics and Truth (2019), French and Spanish, and 900 papers (400 in foreign languages).

List of Figures

Fig. 5.1 Fig. 5.2 Fig. 6.1 Fig. 7.1 Fig. 7.2 Fig. 7.3 Fig. 7.4 Fig. 7.5 Fig. 7.6 Fig. 7.7

Brentano’s analysis of a name Brentano’s analysis of a speech Marty’s analysis of being a sign Simplified Brentanian theory of meaning Complete Brentanian theory of meaning Meinong’s first theory of meaning Meinong’s second theory of meaning Husserl’s theory of meaning Twardowski’s second theory of meaning Ingarden’s theory of meaning

109 110 131 146 147 148 151 155 157 162

xvii

1 Introduction: Mind, Meaning and Reality Arnaud Dewalque, Charlotte Gauvry, and Sébastien Richard

This collection of fourteen original essays is devoted to Franz Brentano’s pathbreaking thoughts on language. Its main objectives are to reconstruct the latter, explore their dissemination in the Brentano School and the Brentanian tradition at large, and throw light on how they anticipated some important advances in the analytic philosophy of twentieth and early twenty first centuries. One of the chief things we do in this introductory chapter, besides clarifying those three objectives, is to spell out the basic assumption, arguably shared by Brentano and all the Brentanians, that a philosophical analysis of meaning is inseparable from considerations about what goes on in the mind and what there is in the world.

A. Dewalque (*) • C. Gauvry Department of Philosophy, University of Liège, Liège, Belgium e-mail: [email protected]; [email protected] S. Richard Department of Philosophy, Free University of Brussels (ULB), Brussels, Belgium e-mail: [email protected] © The Author(s) 2021 A. Dewalque et al. (eds.), Philosophy of Language in the Brentano School, History of Analytic Philosophy, https://doi.org/10.1007/978-3-030-52211-7_1

1

2 

A. Dewalque et al.

What this means should become clear in the course of this introduction. Our plan is as follows. Section 1 offers some preliminary thoughts about the ‘invisibility’ of Brentano’s philosophy of language in contemporary literature, while Sects. 2, 3, 4 present the three aforementioned objectives in somewhat greater detail, thereby setting the framework for the rest of the book. Eventually, Section 5 gives a brief overview of the content of the chapters.

1

 he Invisibility of Brentano’s Philosophy T of Language

Franz Brentano taught philosophy for almost seven years at the University of Würzburg (1866–1873) and for twenty years at the University of Vienna (1874–1894). He is usually considered the leading figure of Austro-German philosophy. His philosophical views were motivated by his large-scale project of renewing philosophy on the basis of thoroughgoing investigations into mental phenomena. They have known a spectacular dissemination through several generations of students in Austria and beyond. Given the breadth of Brentano’s intellectual progeny, the so-called Brentano School may be seen as the starting point of a more encompassing Brentanian tradition. The former comprises Brentano and his first outstanding students, among whom were Carl Stumpf, Anton Marty, Alexius Meinong, Kazimierz Twardowski and the early Edmund Husserl (see Albertazzi et al. 1996; Kriegel 2017a). The question of whether they all shared a fairly unified conception of what philosophy is, or should rather be seen as a heterogeneous group of scholars working on similar topics in a similar way, is still a subject of debate (see Dewalque 2017; Huemer 2019). The situation is all the more complicated that most of them founded in turn their own school: Husserl’s Logical Investigations gave rise to the realistic phenomenological movement (Ingarden, Pfänder, Stein, Scheler, Reinach, Von Hildebrand), Meinong founded the Graz School of object theory (Ameseder, Benussi, Höfler, Mally, Vitasek, Martinak), Twardowski laid the foundations of the Lvov-Warsaw School of Polish

1  Introduction: Mind, Meaning and Reality 

3

philosophy (Łukasiewicz, Leśniewski, Kotarbiński, Ajdukiewicz, Dąmbska, Tarski), and Marty is rightly considered the father of the Prague School of Brentanian orthodoxy (Kraus, Kastil, Katkov). However, it is probably not unfair to say that all those schools exhibit some family likeness and to this extent may be regarded as part and parcel of the Brentanian tradition.1 Representatives of the Brentano School and the Brentanian tradition are best known for having offered significant and enduring contributions to three major research areas, namely philosophy of mind, metaphysics and value theory. By contrast, comparatively little is known about their contribution to philosophy of language, which is rarely, if ever, mentioned in studies dedicated to Brentano’s legacy (see, however, Benoist 2003; Schuhmann 2004, 289–293). For example, Mark Textor, in his introduction to The Austrian Contribution to Analytic Philosophy, makes a quick list of Brentanian themes developed by Brentano’s pupils and mentions only the ontology of parts, the epistemology of perception and memory, and the theory of values (Textor 2006, 6–10). Similarly, the excellent collection of essays gathered in (Fisette and Fréchette 2013) does not contain a single chapter devoted to Brentano’s analysis of language, no more than (Kriegel 2017a). Generally speaking, the Brentanian philosophy of language has remained something of a blind spot in the Brentano studies. As we shall see momentarily, this situation is not surprising. In fact, there is a good reason why Brentano’s philosophy of language remained largely ‘invisible’ so far in the literature.2 Before saying more on that, one notable exception must nonetheless be mentioned, namely the theory of language of Anton Marty, who is usually described as Brentano’s ‘Minister for Linguistic Affairs’ and whose outstanding contribution in this field is now well recognised (see Mulligan 1990; Fréchette and Taieb 2017, part 3; Leblanc and Bacigalupo 2019, part 1). Importantly, Marty’s influential Investigations into the Foundation of General Grammar and the Philosophy of Language (Marty 1908) paved the way to Karl Bühler’s Theory of Language (Bühler 1934, 2011; see also 2012; Cesalli and Friedrich 2014). And yet, the sheer importance of Marty’s and Bühler’s contributions should not conceal the fact that further representatives of the Brentano School also offered original analyses of language. This is certainly true of

4 

A. Dewalque et al.

Marty’s master, Brentano himself, whose own thoughts about language are scattered across various unpublished course notes and have been sadly neglected so far (with only a few exceptions, see Srzednicki 1966; Albertazzi 1989; Rollinger 2009, 2014). But a similar observation also holds for further representatives of the Brentano School, like Meinong and Twardowski, or of the wider Brentanian tradition, like Martinak (1901), Mauthner (1901–1902, 1906) or—later—Ingarden (1965). The essays gathered in this volume aim at filling this gap, at least in part, by providing the reader with a more thorough understanding of the Brentanian approach to language, sign and meaning. At this point an objector might protest, with good reason, that there is something puzzling in talking about ‘philosophy of language’ in the Brentano School and the Brentanian tradition. If indeed philosophy of language is understood as a specific area of philosophical inquiry distinct from philosophy of mind and metaphysics—a view which arguably became customary at some point in the post Frege-Russell analytic tradition—then certainly Brentano and his followers did not elaborate a fully fledged philosophy of language. As we said, Brentano’s big project was to renew philosophy on the basis of psychological investigations. Yet, by ‘philosophy’ he only meant metaphysics supplemented with (what he took to be) the three ‘practical’ branches of philosophy, namely logic, ethics and aesthetics. Obviously, there is no room in this metaphilosophical picture for philosophy of language understood as an autonomous area of investigation. This, alongside the fact that Brentano’s logic courses remained unpublished to date, explains why Brentano’s thoughts on language attracted so little attention in the literature. However, as noted by (Cesalli and Mulligan 2017, 257), this does not mean that considerations on language were utterly absent from Brentano’s writings nor that the Brentanians did not have philosophically interesting views about meaning. In fact, what makes their thoughts about language interesting arguably is their shared conviction that a philosophical analysis of language—and, more pointedly, of what it is for signs and sounds to be endowed with meaning—cannot possibly be disconnected from a philosophical analysis of mind and reality, of what goes on in the mind and what there is in the world. Differently put, it is probably a tacit assumption made by Brentano and his followers that any serious

1  Introduction: Mind, Meaning and Reality 

5

philosophical investigation into meaning (linguistic or other) is to be seen as an integral part of a broader research programme in which mind, meaning and reality are to be integrated into a single coherent picture— or so we shall argue in this introductory chapter.

2

Brentano’s Approach to Language

The first objective of this book is to reconstruct Brentano’s thoughts on language. In this section we offer a preliminary overview of what makes his approach original (more is said about that in Part I of this book).

2.1

Behind the Words: Mental Phenomena

To grasp the distinctiveness of Brentano’s approach to language, it is helpful to make a very brief detour into the history of analytic philosophy. According to a widespread narrative, philosophy of language is born with the idea that investigating language for itself is the most promising way of understanding meaning and intentionality. This idea is traditionally supported by the following assumptions (see e.g. Fennell 2019, 25). First, linguistic signs and sounds mirror the structure of meaning in some important way. Next, unlike mental episodes, which are private—i.e. directly ‘accessible’ to their owner only—linguistic signs and utterances provide the philosopher with a public datum open to intersubjective investigations. Thus understood, the rise of philosophy of language as a separate branch of philosophical inquiry has often been connected to Frege’s claim that meaning is objective in some sense, hence can be studied independently of mental goings-on. In Frege’s well-known words (which probably are reminiscent of Lotze’s interpretation of Plato), it is a ‘third realm’ distinct both from the mental and the physical realms (Frege 1990, 353) and, for that reason, akin to Plato’s realm of Ideas. For the present purpose, there is no need to inquire whether this traditional narrative is entirely accurate or not. Suffice it to note that this Platonistic view of meaning was subject to lively debates in the subsequent analytic tradition and has been attacked on various fronts. Analytic

6 

A. Dewalque et al.

philosophers as different as Schlick, Quine or Ryle, for example, most decidedly distanced themselves from any kind of Platonistic hypostasing of linguistic meanings. Yet, at the same time, the thought that philosophy of language is an autonomous area of inquiry that could and should be insulated from philosophical psychology, arguably remained overarching for a while in twentieth century analytic philosophy. Interestingly, it is Roderick Chisholm—a Brentano scholar—who was one of the first to challenge the idea that signs have a meaning independently of what is going on in the mind of the speaker. On Chisholm’s view, signs are endowed with meaning only insofar as they are meaningfully employed by minded creatures with intentional states. In other words, the intentional characteristics of language are to be explained by the mind’s intentionality and not the other way around (the locus classicus in this debate is Chisholm and Sellars 1972). Later on, a similar view was advocated by John Searle in his influential book on intentionality (1983). In fact, Searle states head-on that ‘many of us working in the philosophy of language see many of the questions of language as special cases of questions about the mind’ (Searle 2004, 7). Returning to Brentano, we believe it is illuminating to introduce his pioneering thoughts on language against the background of those developments, which belong to the subsequent history of analytic philosophy. The best source for reconstructing his view of language is the manuscript EL 80, Logic (online edited by Robin Rollinger), which contains a set of notes Brentano used for his logic courses in Würzburg (1869–1870, 1870–1871) and Vienna (1875, 1877).3 In this manuscript, he already advocates a strongly anti-Platonistic view of language and meaning. One of the main tenets of this view is that linguistic analysis cannot be exhausted by grammatical analysis but requires instead a description of the ‘inner’ mental structures associated with linguistic utterances. Language, Brentano writes, is ‘essentially the sign of thinking’ (Brentano Ms. EL 80, 12.978[9]). Whenever you utter a word like, say, ‘Socrates’, you indicate to some addressee that you have a mental phenomenon— namely, a presentation (Vorstellung) of Socrates—and you purport to influence the addressee’s mental life in a specific way—namely, in making him/her think of Socrates. Similarly, whenever you utter a full sentence like, say, ‘Socrates was Athenian’, you indicate to the addressee that you

1  Introduction: Mind, Meaning and Reality 

7

have another mental phenomenon—namely, a judgement. Thus, ‘it is the overall goal of language to capture or express our mental phenomena, [that is,] what is presented, judged about, wished, loved as such’ (Brentano Ms. EL 80, 13.008[2]–[3]). As a natural consequence of this view, Brentano is led to emphasise the communicative function of linguistic utterances. Very roughly, the idea is that words foremost serve to bridge the gap between the minds of various individuals. Admittedly, this idea is not particularly novel and Brentano did not go as far as maintaining that the meaning of a linguistic utterance depends on what language users intend to do when speaking—a claim which presumably first shows up in Marty’s analyses (Cesalli and Mulligan 2017, 262). Another, probably more interesting implication of Brentano’s approach is that it is impossible to define the concepts of meaning, name, speech and the like without referring to mental phenomena, that is, without looking beneath the surface grammar of linguistic expressions and relating them to what is going on at the psychological level. For the sake of illustration, let us consider the case of categorematic expressions, i.e. expressions which are meaningful in themselves. The following definitions can be extracted from Brentano’s logic course for such expressions E. • E is a name =Def E expresses an act of presentation (Vorstellung). • E is a speech (Rede) =Def E expresses an act of judgement (acceptation/ rejection) or an act of interest (attraction/repulsion). • Etc. As we can see, for Brentano, categorematic expressions have a meaning in themselves, because they are self-sufficient expressions of certain mental phenomena. Furthermore, linguistic categories seem to be connected to certain classes of mental phenomena. These classes of mental phenomena are those that Brentano identified in his Psychology from an Empirical Standpoint. Naturally, whether or not a word is endowed with meaning for you certainly is a matter of contingency. If you do not understand German, the word Eichhörnchen will probably have no meaning for you in the sense that it does not trigger any presentation into your mind. Conversely, knowing that Eichhörnchen is the German word for ‘squirrel’ is

8 

A. Dewalque et al.

tantamount to having some (conceptual or imaginative) presentation of a squirrel whenever you encounter the word Eichhörnchen. Yet, putting aside such contingencies, the word Eichhörnchen may be said to be meaningful in itself insofar as, even if it is taken in isolation (i.e. when it is not part of a speech), it may be said to express a mental phenomenon, namely an act of presentation of something, be it a chipmunk, a squirrel, a rodent or even an I-don’t-know-what. Not all linguistic expressions are able to trigger a mental phenomenon by themselves. Indeed, some words like ‘and’, ‘or’, ‘because’, ‘therefore’, ‘is’, etc., are not self-sufficient expressions of mental phenomena: when you read them or hear them in isolation, it is impossible for you to have a presentation of something which corresponds to one of those words. They are not categorematic expressions, but syncategorematic ones. Unlike the Eichhörnchen case, these expressions only acquire a meaning when they are used in combination. For example, the word ‘and’ cannot possibly express (or arouse) an act of presentation, although the phrase ‘Laurel and Hardy’ can. While ‘and’ is a syncategorematic (or, as Marty puts it, synsemantical) expression, ‘Laurel and Hardy’ is a categorematic (or autosemantical) one. What these analyses show is that for Brentano linguistic expressions are best understood within a psychological context, i.e. by being related to the mental acts they indicate or to a complex that expresses such acts (more on the idea of psychological context in Chap. 2 and Chap. 3). Therefore, linguistic analyses should never be made in isolation. On the contrary, philosophy of language should be based on descriptive psychology, understood as a phenomenal psychology or a theory of mental phenomena.

2.2

Grammatical Illusions

On Brentano’s view, words get their meaning in virtue of the underlying mental phenomena they express (or arouse). Yet, equally central to his view is the claim that there is no perfect match—no one-to-one correspondence—between the words we use and the mental phenomena we purport to express with them. As a matter of fact, there are many

1  Introduction: Mind, Meaning and Reality 

9

structural mismatches, in the sense that some linguistic expressions have a complex structure while the corresponding mental phenomenon is simpler or vice versa. For example, in the Psychology from an Empirical Standpoint, Brentano identifies the predicative structure of declarative sentences as the linguistic source of error regarding the nature of judgement: what prevented savvy philosophers to see that the phenomenon of judging is a simple act of accepting-as-true or rejecting-as-false is the complex predicative structure (S is p) by means of which this simple phenomenon is captured in ordinary language (see Brentano 1925, 74–75; 1995, 228). Conversely, in Sprechen und Denken, Brentano argues that linguistic expressions are ‘by far less diverse’ than mental phenomena, to the effect that an apparently simple word like the word ‘state’ actually captures a very complex thought. Moreover, further examples of mismatch are offered by the well-known cases of suppositio materialis (e.g. ‘“tree” is a four-letter word’), equivocation (i.e. when different thoughts are expressed by means of the same words) and synonymy (when the same thoughts are expressed by means of different words). One specific class of mismatch cases is that of modifying adjectives like ‘fake’, ‘counterfeit’, etc. Such adjectives are said to be modifying inasmuch as, instead of attributing some property to an object, they change the nature of what is presented: when you talk of a ‘fake gun’, you are not thinking of a real gun which would have the property of being a fake, and when you talk of ‘counterfeit money’, you are not thinking of real money which would have the property of being counterfeit. Naturally, how to best describe such modifications is an open question (see Chap. 12). Twardowski, for example, proposed in 1921–1925 a more sophisticated analysis of modifying adjectives than the Brentanian account he had first endorsed in his Habilitation thesis: while determining is a simple function, he argues, modifying is a complex one which involves partially deleting the content of a presentation and replacing it with other characteristics (see Twardowski 1977, 1999). Be that as it may, there is here a specific kind of mismatch between the linguistic expression, which has all the appearances of a simple determination, and the underlying mental phenomenon, which does not consist in adding a determination to something.

10 

A. Dewalque et al.

An important consequence of all that is that you cannot just infer the structure of thought from the structure of language. It is crucial to recognise that there are grammatical illusions or, to put it differently, that some expressions are misleading in a systematic way (see Chap. 4 and Chap. 5). This thought is at the core of Brentano’s programme of ‘critique of language’ (Sprachkritik), which as we shall see anticipated important views held by Mauthner, Ryle and Wittgenstein—among others. One way of understanding this programme is as a systematic translation of misleading expressions into non-misleading ones. The idea is not to restore a perfect match between words and mental phenomena, since given the intrinsic limitations of language, establishing a one-to-one correspondence would be hopeless. Rather, it is to rephrase some sentences in such a way that the addressee will not be tempted to posit fictional entities.

3

 he Transformation of the Brentanian T Research Programme

The second objective of this volume is to explore the reception of Brentano’s thoughts on language among his students and his student’s students. This reception is inseparable from a certain evolution and transformation of Brentano’s seminal ideas, with novel concepts and questions coming to the fore. This section gives a first pass at the evolution of the Brentanian research programme, which is addressed in greater details in Part II of this book.

3.1

A Research Programme

It is sometimes suggested that, for Brentano, descriptive psychology must assume the role of first philosophy and that this methodological primacy of the philosophy of mind is ‘a unifying theme of the Brentano School’ (Kriegel 2017b, 30). In philosophy of language, it means, as we have seen, that an analysis of meaning and language must first be grounded in a philosophical analysis of the mental acts expressed by linguistic expressions. For Brentano this analysis of mental acts has to be a descriptive

1  Introduction: Mind, Meaning and Reality 

11

psychological one. The hypothesis of the primacy of the mental in the philosophical analysis of linguistic phenomena belongs to the core of what we have called the Brentanian research programme in philosophy of language. This programme was probably never made fully explicit, but it is a natural consequence of the architectonic position that Brentano gave to descriptive psychology among the philosophical disciplines. Twardowski’s analysis of the word ‘nothing’ can be used to illustrate this methodological position. In the sentence ‘nothing is eternal’, the word ‘nothing’ seems to be used as a substantive. It should thus refer to something, since if it was not the case there would be nothing about which we could say that it has the property of being eternal. Apparently, this problem is a semantic one: does the word ‘nothing’ refer to something? However, Twardowski does not solve this problem directly at a semantic level. The answer to the semantic question depends first on the answer to a psychological question: does the word ‘nothing’ express a presentation? According to Twardowski, this is not the case: the word ‘nothing’ in the sentence ‘nothing is eternal’ does not express a presentation by itself—it is not a categorematic term, but a syncategorematic one. ‘Nothing’ is an abbreviation for ‘non-something’ (or ‘that, which is not something’). Now, Twardowski argues, in order for a phrase of the form ‘non-F’ to be categorematic, there must be a more general notion G which is likely to be divided into ‘the Gs which are F’ and ‘the Gs which are not F’. For example, ‘non-smoker’ alludes to the division of human beings into those who smoke and those who don’t, and to that extent only can be said to be a categorematic expression. But ‘non-something’ does not meet this condition, for a simple reason: there is no higher notion G which is more general than ‘something’ and could be divided into ‘the Gs which are something’ and ‘the Gs which are not’. Therefore, the phrase ‘non-something’, unlike ‘non-smoker’, cannot possibly arouse any presentation into your mind and is not a categorematic expression. Rather, to understand how the word ‘nothing’ works in the sentence ‘nothing is eternal’, we must understand the judgement expressed by this sentence, i.e. the judgement that there is nothing eternal. It can then be concluded that the word ‘nothing’ is a part of a sentence that marks the rejection of the idea that something is eternal.

12 

A. Dewalque et al.

This analysis of the word ‘nothing’ bears striking similarities with Carnap’s famous analysis of the same word (1931). It is not an isolated case. After his reistic turn, Brentano treated entia irrealia as mere ‘entia linguae, as fictions that are due to a misunderstood multiplication of linguistic expressions’ (Brentano 1966, 174). He suggested that sentences containing subjects that do not refer to anything can be paraphrased into sentences that express the same thought but contain only subjects that name real things (Brentano 1995, 287). The first sentences are only ‘abbreviations’ of the second ones. This use of paraphrases to get rid of the illusions created by our ordinary use of language is similar to the philosophical methodology defended by Russell from 1905 onwards: philosophical problems can be solved by rephrasing them in order to exhibit their deep logical structure, i.e. the authentic propositional constituents that are concealed under the surface grammar of sentences. The Brentanian tradition and the emerging analytic philosophy both think that the deep structure of language is a logical one and that the disclosure of this structure can solve some of the problems raised by language. However, they do not agree on a fundamental point: the nature of logic itself. For the first analytic philosophers, the logic to be used was the formal logic invented by Frege, whereas for the Brentanians it was the logic reformed by descriptive psychology (see Simons 1987). If ordinary language can be misleading, it is because it does not adequately express the logic of our thoughts. While for Frege, Russell or Carnap logic is an autonomous theoretical discipline, for Brentano and the Brentanians it is a practical one: it is the ‘art of judgement’, and more specifically the art of ‘correct judgement’ (Brentano Ms. EL 80, 12.956; Twardowski 2016, 12). In this respect, logic is grounded on descriptive psychology, since it is the latter which provides the logician with a descriptive (i.e. conceptual) theory of presentations and judgements.

3.2

The Objective Side of Meaning

The first analytic philosophers would certainly not have accepted the Brentanian subordination of logic to descriptive psychology. In the emerging analytic philosophy, the mental correlation with linguistic

1  Introduction: Mind, Meaning and Reality 

13

expressions was forcefully rejected to preserve the objective aspect that meaning seems to exhibit, especially in mathematics and logic. It is philosophy of language that has to play the role of first philosophy. Does that mean that the Brentanian tradition embraces a psychologistic conception of meaning? Some Brentanians did indeed embrace a form of ‘semantic subjectivism’. However, psychologism was also heavily debated among them. If all the Brentanians accept the view that a psychological method has to be employed to solve philosophical problems, this does not mean that they all share the view that meaning is a subjective entity, and even those who accept this view do not necessarily consider that it implies psychologism with respect to logic. The case of Twardowski is again interesting in this respect. In his Habilitation thesis, he first identifies the meaning of a name with the ‘content’ of the presentation expressed (or aroused) by this name (Twardowski 1977, 9). Yet, under the influence of Husserl’s and Łukasiewicz’s criticism of psychologism, he developed an objectivist conception of meaning by carefully distinguishing the mental acts from their products (see Chap. 12). Still, this switch does not mean that he gave up the view that descriptive psychology has to be the main tool for philosophical analysis. Twardowski thus seems to have abandoned the view that descriptive psychology must play the role of first philosophy in favour of a more subtle methodological position in which different philosophical disciplines have to play a complementary role according to the philosophical problem that is studied. Even if psychology is still the main philosophical tool for solving philosophical problems, it is not the only one. Philosophical problems must keep their specificity and philosophers must use psychological, logical, grammatical and metaphysical analyses to solve them.4 This intertwining of philosophical disciplines is not specific to Twardowski, but is a ‘methodological constant’ of the Brentanian tradition (Fisette and Fréchette 2007, 144). Another core thesis of the Brentanian research programme is that the analysis of linguistic expression cannot be disconnected from the analysis of the objects to which these expressions refer. The reason for this is that, for Brentanian philosophers, mental acts are intentional, i.e. are directed toward objects. For Brentano, these objects can be only real objects, while for others they can be real or ideal (Husserl, Twardowski) and for yet

14 

A. Dewalque et al.

others they can be real, ideal or even purely intentional (Ingarden), and even non-existent (Meinong) (see Chap. 7). It can be suggested that the investigations of the objects intended by mental acts gradually moved away some of the Brentanians from descriptive psychology. For instance, if the ‘theory of the object in general’—the theory of objects—was first ‘built up on the basis of a descriptive psychological analysis of the different kinds of mental acts’ (Smith 1994, 159) by Twardowski and Meinong, it soon became in the hands of the latter an autonomous discipline.5 As for Husserl, he partly developed his own phenomenology in order to clarify the intentional relation postulated by Brentano as the mark of mental phenomena (see Fréchette 2019, 428–430), and he insisted that phenomenology is best conceived of as distinct from descriptive psychology (whereas Brentano identified the two). In consequence, Husserl’s first students, such as Reinach or Ingarden, did not use descriptive psychology as a foundation for linguistic analysis anymore, but rather phenomenological analysis. This does not mean that the Brentanian research programme in philosophy of language disappeared in the different philosophical schools developed by Brentano’s students. It simply evolved: the philosophical analysis had still to be correlated with an analysis of mental acts, but this analysis  did not necessarily have to be a descriptive psychological  one. The persistence of this programme can still be discerned in Meinong’s and Ingarden’s theories of meaning (see Chap. 7). The accounts of meaning and language of Brentano’s heirs that we have just mentioned also departed from the sort of nominalism defended by their master. Most of them multiplied the entities that could be intended by mental acts (states of affaires, non-existents objects, ideal qualities, essences, purely intentional objects, and so on). From Brentano’s point of view that multiplication was unnecessary and only suggested by misleading linguistic pictures (see Chap. 5). Some students of Brentano remained, however, faithful to his ontological austerity and descriptive-­ psychological method. This is in particular the case of Marty (see Chap. 6), Kastil and Krauss. If most members of the Lvov-Warsaw School also maintained a strong nominalist account of language and meaning, their situation is somewhat different from Brentano’s closest students. Students of Twardowski

1  Introduction: Mind, Meaning and Reality 

15

thought that anti-psychologism could be reached by the new formal logic or ‘logistic’. Therefore, they devoted themselves to its study and became less interested in descriptive psychology (Kotarbiński 1976, 195). Their analyses in philosophy of language came closer to those pursued at the same time in the analytic tradition, even if they never accepted the linguistic turn and maintained the Twardowskian multidisciplinary approach to philosophical problems. Brentano’s influence on them was over all of a metaphilosophical nature, in so far as they were convinced that philosophy should not be speculative but adopt the exact method of the natural sciences.6

4

Brentano and Analytic Philosophy

As mentioned at the outset, a third objective of this volume is to explore the wider legacy of the Brentanian analyses on language, including significant connections with analytic philosophy. It is common knowledge that Brentano exerted an underground but in-depth influence on early analytic philosophy. During his lifetime, his views were discussed by George Stout, who helped to give his well-known students, George Edward Moore and Bertrand Russell, their seminal intuitions (see e.g. Schaar 2013). Besides, there are numerous, direct and indirect connections between representatives of the Brentanian tradition and (early) analytic philosophers: Twardowski’s pathbreaking distinction between ‘content’ and ‘object’ has often been compared to Frege’s distinction between Sinn and Bedeutung, Wittgenstein’s views have been described as continuous with the Brentanian tradition, and Marty’s thoughts on language have been compared to (neo-)Gricean theories of communication (see e.g. Longworth 2017). However, the many connections between the Brentanian analyses and prominent views in analytic philosophy have not been subject to a systematic exploration yet. Such connections are highlighted in several chapters throughout this book. For example, Brentano’s views are compared with Frege’s in Chap. 2 and Chap. 3, and with Ryle’s in Chap. 5. Part III of this volume offers a closer look at Brentano’s legacy in Wittgenstein, Bühler and Polish philosophy. In this closing section of our introduction, we want to suggest

16 

A. Dewalque et al.

that Brentano’s linguistic analyses have anticipated four historical stages of the analytic tradition, for four different reasons.

4.1

Philosophy as a Critique of Language

We already mentioned that Brentano’s ‘critique of language’ presents some striking similarities with early analytic positions like those endorsed by Russell, Ryle or Wittgenstein. Although protean, most of early twentieth century Austrian contributions to the philosophy of language can be characterised in terms of ‘critiques of language’ (Sprachkritiken). While Fritz Mauthner (1849–1923) is probably one of the first, if not the first, to use this concept, the views advocated by Ernst Mach, Karl Kraus, Oskar Kraus, and the German Hans Vaihinger, among others, are indeed different versions of a general ‘critical’ attitude toward language. To be more specific, according to Mauthner (1901–1902), such an attitude tends to define philosophy as a whole as a ‘critique of language’ whose purpose is to dissolve the grammatical illusions that contaminate our ordinary language. As a result, it aims at showing that most of the so-­ called philosophical or ontological problems—typically, problems tied to existential generalisations—are pseudo-problems that need to be dissolved by linguistic analysis. In this respect, one is entitled to consider the Brentano School as pertaining to this critical tradition. Marty is probably the first Brentanian to have developed a systematic approach on Sprachkritik (see e.g. Mulligan 2019, 198), but several commentators (e.g. Kraus 1929, 16; Weiler 1970; Haller 1979; Mulligan 2019) have convincingly shown that Brentano himself, at least during his reistic period, was already pursuing a similar research programme. Already in his 1874 Psychology, he considered ordinary language as deceptive (albeit useful) and consequently called for a systematic rephrasing in the light of descriptive-psychological analysis (Brentano 1995, 45). As mentioned before, predicative judgements should be rephrased in terms of existential ones. Those rephrasing requirements are even more manifest after Brentano’s reistic turn. He considers then that there are linguistic traps due to the existence of misleading expressions, i.e. expressions which seem to refer to real things although

1  Introduction: Mind, Meaning and Reality 

17

they do not (as for instance ‘Pegasus’, ‘the centaur’, ‘the golden mountain’, ‘abstract terms’, ‘universals’, etc.). This is what Brentano calls ‘linguistic fictions’ (see e.g. Brentano’s letter to Marty dated March 1901, in Brentano 2009, 43). As a consequence, the purpose of his critique of language is to rephrase fallacious grammatical associations by introducing sharp distinctions between names in the logical sense (‘categorematic’ expressions) and ‘names’ in a merely grammatical sense (‘centaur’, ‘golden mountain’, etc.), the latter being considered as ‘syncategorematic’ expressions (see Brentano’s correspondence with Marty, in particular Brentano 2009, 49). Brentano even uses the term ‘translation’ in a 1907 dictation: the purpose of Sprachkritik, he writes, is to ‘translate [ordinary language] back into concrete terms’ (Brentano 2009, 65). It is sometimes suggested that another ‘Austrian’ philosopher, namely Ludwig Wittgenstein, is a representative member of this critical tradition (see e.g. Mulligan 2012, 39–42; Cesalli and Mulligan 2017, 257). Although very attractive, this claim arguably has to be nuanced. It is true that Wittgenstein himself acknowledges his allegiance to thinkers such as Karl Kraus or Fritz Mauthner. Moreover, as is well known, he asserted in his Tractatus that ‘all philosophy is a “critique of language”’ (Wittgenstein 1922, 4.0031). However, he immediately added: ‘but not at all in Mauthner’s sense’ (Wittgenstein 1922, 4.0031, our emphasis). According to Wittgenstein, Russell’s own practice would be closer to what he means by ‘critique of language’: ‘Russell’s merit is to have shown that the apparent logical form of the proposition need not be its real form’ (Wittgenstein 1922, 4.0031). As a result, the comparison suggested here deserves some caution. It makes sense to compare the critical method of Sprachkritik with the therapeutic approach of the later Wittgenstein, who also considers philosophical mistakes as purely linguistic and as produced by incorrect uses of language. As he puts it: ‘Philosophy is a battle against the bewitchment of our intelligence by means of language’ (Wittgenstein 1953, I, § 109, 40). Yet, Wittgenstein’s criterion of correction has clearly not to be provided by the revelations of psychological analysis (as is the case in Brentano’s view) but by the diktat of ordinary uses. In addition, Wittgenstein would never assume that ordinary language has to be ‘rephrased’. Therefore, although both Brentano’s and Wittgenstein’s programmes can be rightly

18 

A. Dewalque et al.

considered as ‘critiques of language’, their respective targets have to be distinguished: while the former considers ordinary language to be illusory, the latter insists that it is philosophical language which is illusory.

4.2

Ordinary Language Philosophy

Despite huge differences, the Brentanian tradition also presents some non-superficial connections with the philosophy of ordinary language that the second Wittgenstein and John L. Austin independently brought to the fore in the 1930s–1960s. Given what precedes, one could say that the Brentanian tradition and ordinary language philosophy differ in virtue of how critical they are toward ordinary language. However, such conclusion probably needs to be qualified. Brentano certainly considers that ordinary language, which is pervasive, should not be taken ‘at face value’, so to speak. Yet, he also assumes that ordinary language is the unique tool available when it comes to pursuing philosophical investigations. Hence, ordinary language, however imperfect, is priceless. That is the reason why, despite his rephrasing attempts, Brentano never intended to substitute ordinary language with some ideal language, contrary to some of his notable contemporaries, such as Frege (see for instance Brentano Ms. EL 72, I, 16). Furthermore, Brentano paid positive attention to some specific aspects of ordinary language, in particular to its practical dimension. In his correspondence with Marty, he even asserts that a logical name, although it is a categorematic expression, becomes plainly meaningful only when used in a ‘practical’ expression, i.e. ‘as a part of the function of assertion’ (Marty, May 28, 1873; see also Kraus 1934, 39) or when used in ‘a complete expression’ (Ms. EL 80, 13.001). Based on such telling textual evidences, some (see e.g. Kraus 1934; Mayer-Hillebrand 1977, xxiii; Mulligan 1990, 17–18; and Chap. 2 in this volume) even count Brentano and a few Brentanians among the ‘contextualists’. Some of their intuitions would have anticipated Frege’s well-known context principle: ‘Never to ask for the meaning of a word in isolation, but only in the context of a proposition [im Satzzusammenhange]’ (Frege 1884, x). Even if there are indeed some similarities between those

1  Introduction: Mind, Meaning and Reality 

19

different compositional claims, it is at least controversial whether they all speak of the same ‘context’, in particular because the Fregean context is propositional and not psychological at all (see e.g. Chap. 3). Therefore, it is probably less controversial to stick to a more modest correlative claim: Brentano’s and Marty’s attention to the context-sensitive ‘vagueness’ of common names anticipated Wittgenstein’s theory of ‘family resemblance’ (see Mulligan 1990, 2012), although the former accounts for the ‘vagueness’ of common names by resorting to ‘typicality’ and ‘mental content’ (see e.g. Taieb in press; and Chap. 13).7 Accordingly, we believe that doctrinal adjectives such as ‘pragmatist’ and ‘contextualist’ should be applied with caution to representatives of the Brentanian tradition, even though the latter showed a surprisingly deep sensitiveness to the practical dimension of ordinary language and to composition principles.

4.3

Communication Theory

It is commonly claimed that the Brentano School paved the way for the (neo-)Gricean theories of meaning that are still very influential nowadays. More precisely, Marty is paradigmatically considered as having anticipated Paul Grice’s position (see e.g. Liedtke 1990; Cesalli 2013; Longworth 2017; Recanati 2019). The Brentanian and Gricean traditions indeed share the idea that a distinction must be made between natural and non-natural meaning (on this controversial distinction, see again Recanati 2019) and to consider non-natural meaning as intentional. According to both traditions, an utterance has a meaning only when it is used with a communicative purpose, that is, with the intention of manifesting one’s own mental life and influencing the mental life of the addressee (see Marty 1908 and Grice 1957). More pointedly, there are essentially two ‘linguistic functions’ according to Marty. First, the function of expression or intimation (Kundgabe, Aüßerung): according to this function, an expression is meaningful only if it expresses or intimates the speaker’s own mental state to the hearer. Second, the function of communication: one speaks with the intention to steer or to control someone else’s mental states, or at least with the intention to influence his mental life. According to Marty, the second function

20 

A. Dewalque et al.

clearly is the primary one. In this respect, it makes sense to consider that he anticipated the later intentional theories of communication. While Marty can probably be considered the most faithful heir of Brentano, it remains controversial whether Marty’s theory of communication was already anticipated in the analyses of his mentor. The first definition of the intentional function obviously comes from Brentano. However, it is not clear whether Brentano was also sensitive to the steering dimension of the second linguistic function. Be that as it may, it makes sense to consider the intentionalist theory of meaning as a clear outcome of the Brentano School which consequently anticipated one of the most influential views of the later analytic philosophy of language.

4.4

An Integrative View

As we have seen, central to the Brentanian approach to language is the claim that linguistic, psychological and ontological investigations are to be integrated into a single picture. Now it may be argued that this ‘integrative view’ anticipates some of the most contemporary psycholinguistic theories of language. Various recent theories indeed have it that philosophy of mind, metaphysics and philosophy of language are closely connected to one another. For example, most of today’s philosophers of mind used to be philosophers of language in the 1960s–1970s (think, among many others, of Putnam, Dretske, Searle, Burge, McDowell, etc.). Besides, many contemporary philosophers of language are happy to introduce psychological and ontological considerations in their linguistic analyses, making room for modal and intentional considerations in their theories (see e.g. Recanati 2012; Soames 2013; Boghossian 1994). The Brentanian prefiguration of those ‘integrative’ theories has been partly documented in (Moltmann and Textor 2017), which rightly underlined the ‘Brentanian’ origin of the ‘act-based conceptions of propositional contents’. For all those reasons, it is probably not incorrect to say that the Brentanian research programme anticipated significant developments in the analytic tradition.

1  Introduction: Mind, Meaning and Reality 

5

21

Content of the Volume

This volume is divided into three parts: Brentano and philosophy of language (Part I), The Brentano School: act, meaning and object (Part II), and Brentano’s wider legacy (Part III). The first part deals with Brentano’s seminal contribution to philosophy of language, highlighting its relations to the broader Austro-German tradition and to analytic philosophy. It contains five chapters. Chapters 2 and 3 address the claim that meaning is context-sensitive and raise the question of whether this claim (or some version thereof ) can be traced to Brentano. In Chap. 2, Guillaume Fréchette compares Brentano’s approach with Frege’s well-known context principle, which is usually regarded as the definitional feature par excellence of analytic philosophy. This view was championed by Michael Dummett in the 1990s. Before him, philosophers such as Peter Hacker and Willard van Orman Quine were also sympathetic to this view. There are, however, many proposals regarding how to understand Frege’s context principle as paving the way to analytic philosophy, some of them more radical than others. Fréchette argues that the less radical interpretations, which offer a larger applicability of the principle while being closer to Frege’s motivation, are already to be found in Austro-German philosophy. In Chap. 3, Charlotte Gauvry connects the distinction between categorematic and syncategorematic expressions to the question of whether Brentano’s analyses are context-sensitive. On the one hand, she insists that, for Brentano, even categorematic expressions are somehow context-­ sensitive. On the other hand, she goes on, Brentano’s conception of context is mainly internal-psychological. Contrary to the later notion of context employed by Wittgenstein (among others), it does not integrate social and normative components. This, she submits, is a major difference between Brentano and so-called ordinary language philosophers. Chapters 4 and 5 are devoted to Brentano’s account of grammatical illusions. Denis Seron explores in Chap. 4 some differences and similarities between Brentano’s and Fritz Mauthner’s critique of language. He argues that the very starting point of both is one and the same fact: the failure of existential generalisation. Brentano’s critique of grammatical

22 

A. Dewalque et al.

illusions, he concludes, may be seen as part and parcel of a broader tradition that includes prominent figures of Austrian and German philosophy such as Mach, Vaihinger, and Wittgenstein. In Chap. 5, Arnaud Dewalque argues that Ryle’s account of misleading expressions, which is usually considered a milestone in the history of analytic philosophy, is continuous with Brentano’s critique of language. Not only do Ryle and Brentano share an implicit understanding of misleading expressions in terms of a gap between surface grammar and truth conditions, but their analyses are driven by a form of ontological parsimony which sharply contrasts with rival views in the Brentano School, like those of Meinong and Husserl. Ryle, Dewalque concludes, is more ‘Brentanian’ than many representatives of the Brentano School. Part I ends with a paper devoted to the reception of Brentano’s linguistic research programme in Anton Marty, who certainly was its most orthodox defender. Hélène Leblanc explores in this Chap. 6 the scholastic background of Marty’s early theory of signs and meaning, showing how it combined with Brentano’s legacy to shape Marty’s mature ‘functionalist’ or ‘instrumentalist’ theory of language. Part II of the volume reconstructs the multifaceted debates on the objectivity of meaning in the Brentano School and its aftermath. It brings into the picture the views held by Meinong, Husserl, Ingarden, Twardowski and the members of the Lvov-Warsaw School. Chapters 7 and 8 deal with Ingarden. To be more specific, in Chap. 7, Sébastien Richard’s aim is to cast some light on the influence of the Brentanian tradition on Ingarden’s theory of meaning. He first sets out the most common theory of meaning that can be found in the Brentanian tradition, a theory that finds its origin in the work of the founding father of this tradition. This theory conceives meaning essentially as a subjective entity. Richard then shows that under the influence of the critique of psychologism some Brentanians developed alternative theories that try to highlight the objective aspect of meaning. He explains the attempts to do so made by Husserl, Meinong and Twardowski and finally shows how some traces of these can be found in Ingarden’s own theory of meaning. In Chap. 8, Olivier Malherbe offers a transversal review of Ingarden’s theory of meaning. Although meaning finds a first complete elucidation in Ingarden’s first major work, The Literary Work of Art, Ingarden’s

1  Introduction: Mind, Meaning and Reality 

23

subsequent analyses, particularly in aesthetics and axiology, lead him to speak of ‘meaning’ in a second sense, irreducible to the first. Herein, they considerably enrich this first approach and offer interesting ways to account for the irrational, emotional and contextual elements of meaning. The essay particularly focuses on the complementarity of these two senses of ‘meaning’ and shows that the second one is of an axiological nature. To sustain this thesis, Ingarden’s concepts of metaphysical quality and value are analysed. Some concrete examples of value-shaped languages according to Ingarden (i.e. artistic, scientific and ordinary language) are then given. Chapters 9 and 10 are devoted to theories of meaning developed by prominent representatives of the Brentano School who broke with Brentano’s ontological parsimony, namely Twardowski and (most notably) Meinong. They address various issues connected to the theory of intentional acts and the objectivity of meaning. Thus, in Chap. 9, Denis Fisette reconstructs and briefly assesses Twardowski’s strategy to avoid psychologism. The core of this strategy lies in the product-act distinction and, more pointedly, in the claim that products are not reducible to acts. For instance, Twardowski claims that the product of mental acts (the thought) has to be studied by logic and the action of thinking (the thinking) has to be studied by psychology. Psychologism rises when philosophers do not carefully maintain this distinction. Bruno Leclercq explores some consequences of Meinong’s Gegenstandstheorie for the theory of meaning and reference in Chap. 10. He shows that after Twardowski, Meinong makes place for inexistent objects and their properties and claims that this view provides a very interesting counter-model to the standard account of meaning and reference which is linked to extensional semantics. However, Leclercq argues, by equating inexistent objects with the sets of their descriptive features, Meinongian formal systems tend to jeopardise the content-object distinction. The universal validity of such a distinction may be assessed, he suggests, by looking at different kinds of “inexistent objects”. Jan Woleński presents in Chap. 11 the extensionality/intensionality distinction in the Lvov-Warsaw School and discusses several attempts made by Polish philosophers to eliminate intensionality. Although the extensional/intensional distinction goes back (at least) to medieval

24 

A. Dewalque et al.

logicians, its contemporary version occurs in Frege, Russell, Wittgenstein and Carnap. Twardowski introduced the distinction between content and object of presentations and applied it to linguistic expressions. He directly influenced many Polish philosophers and logicians, including Łukasiewicz, Leśniewski, Ajdukiewicz and, indirectly, Tarski. All mentioned philosophers and logicians shared the extensionality thesis in a very strong sense. Roughly speaking, a reasonable logic should be extensional—intensionality is a logical defect. In Leśniewski’s system, the thesis of extensionality is a theorem. Łukasiewicz’s modal logic, based on three-valued logic, is extensional—these feature of Ł-modal logic make it differ from the Lewis systems. Tarski’s truth theory is extensional—the predicate ‘is true’ refers to the set of true sentences. Finally, Ajdukiewicz in his works tried to extensionalise intensional contexts. Part III of this volume contains four essays on Brentano’s wider legacy in the field of philosophy of language. Chapters 12 and 13 are devoted to some aspects of Husserl’s phenomenology. Maria van der Schaar shows in Chap. 12 that conceptual analysis in the Brentano school is not based on a form of conceptual atomism, and is in that sense different from more standard accounts of conceptual analysis. A concept like that of ‘illusion’ is understood by Husserl in terms of perception and an operation of modification; in a similar way, one can understand the idea of merely thinking as a modification of the act of judgement. The Husserlian idea of modification has its origin within the Brentano School in an interest in the logical-linguistic phenomenon of expressions containing modifying terms, like ‘toy-duck’ or ‘false gold’. Van der Schaar’s chapter shows the philosophical significance of the idea of modification in Husserl’s Logical Investigations. It ends with a first proposal for a phenomenological account of modification. Hamid Taieb presents and evaluates in Chap. 13 the early Husserl’s account of typicality. In the Logical Investigations, Husserl holds that the meaning of ordinary-language (common) names is sensitive to typicality: this meaning depends on typical examples which vary in different contexts and are more or less similar to one another. This seems to entail that meanings, which according to Husserl are concepts, are ‘fluctuating’ (schwankend) and vague. Prima facie, such a claim contravenes his theory

1  Introduction: Mind, Meaning and Reality 

25

of ideal meanings, or concepts, which are ‘fixed’ (fest) and sharp. However, Husserl wants to save this theory. He claims that the fluctuation and vagueness in question are not to be found in the meaning itself, or the concept, but rather derive from the act of meaning. Thus, he apparently manages to make room for typicality in ordinary language while accepting only fixed and sharp meanings. After presenting the Husserlian theory, Taieb evaluates it and asks whether Husserl will still be committed, despite his own claims, to accepting prototype concepts to account for typicality in ordinary language. Chapters 14 and 15 deal with Bühler’s theory of linguistic and non-­ linguistic expressions, as well as with Wittgenstein thoughts on guidance and rule-following. In Chap. 14, Basil Vassilicos explores how Wundt’s and Bühler’s respective conceptions of gestural expression have implications for how each conceives of what, in broad terms, may be understood as a ‘grammar of gestures’: that is, the rules for the formation and performance of gestures with and without speech. Unlike previous scholarship that has looked at the relationship of Wundt and Bühler, the aim in this essay is to give particular attention to the relevance of their respective accounts for current philosophical and linguistic research on gesture. Building on Bühler, Vassilicos can offer an alternative to the psychologistic and solipsistic model of gestural expression that can be found in Wundt and that seems to be resurgent today in works by McNeil and others. Via Bühler’s notions of diacrisis and of the underlying functions that guide human communication, he proposes an understanding of the relationship between gestures and speech that aims to render accurately both the highly contextualised and the highly structured conditions of their mutual employment. Kevin Mulligan has examined in several papers and a book the conceptual relations between the descriptions of mind, language and colours in the philosophies of Brentano’s heirs and the descriptions given later by Wittgenstein. In Chap. 15, he looks at what Bühler and Wittgenstein have to say about the phenomenon of being guided by something and two of their favourite examples—reading and our relations to rules. ***

26 

A. Dewalque et al.

Previous drafts of those chapters have been presented at the conference Philosophy of Language in the Brentanian Tradition, hosted at the University of Liège on September 17–19, 2018, thanks to the financial support of the F.R.S.-FNRS (research project ‘Linguistic Fictions: From Brentano to the Ordinary Language Philosophers’). We are grateful to all the participants there and to authors for their contributions. We are also grateful to Mike Beaney and several anonymous reviewers for their help at various stages of this book project.

Notes 1. As it is constructed here, the Brentanian tradition involves the students of Brentano’s students whose philosophical views still bear a certain family likeness with his. Note well: this characterisation is purposefully vague. Our point simply is that it makes sense to look at the members of the aforementioned schools in the light of Brentano’s legacy, even though they do not strictly speaking belong to the Brentano School. 2. The talk of ‘invisibility’ has first been introduced by Roberto Poli to refer to the oblivion into which Brentano’s thoughts have fallen in twentieth century philosophy. See (Poli 1998). 3. Further interesting sources are Brentano’s notes on Sprechen und Denken (EL 66), which have been published and translated in (Srzednicki 1965, 116–121), and Marty’s unpublished notes on Brentano’s logic course (Deductive and Inductive Logic, Würzburg, Summer Semester 1869–70 and 1870–71, Archives of Anton Marty, Br 7/1, F.  Brentano, Logik Kollegien, Franz-Brentano Archiv, Graz). 4. It can be noted that the subtitle of Twardowski’s 1912 paper on actions and products is “Some Remarks from the Borderline of Psychology, Grammar and Logic”, which recalls the title of a paper by Marty from 1894: “Subjectless Sentences: On the Relation of Grammar to Logic and Psychology” (Marty 1918). 5. On the place of psychology in Meinong’s object theory, see (Dewalque and Raspa 2019). 6. Kotarbiński defended an ontological theory—that he called ‘reism’—akin to the one developed by Brentano at the beginning of the twentieth century. According to this theory, only real objects exist. Even if, like Brentano,

1  Introduction: Mind, Meaning and Reality 

27

Kotarbiński used a translation strategy to show that all sentences containing names purporting to refer to universals, fictions, and so on, are in fact non-referring names, he developed his reistic doctrine independently from Brentano. He based it on Leśniewski’s logic and not on descriptive psychology. Therefore, Kotarbiński’s reism cannot be put forward as an example of the Brentanian research programme in philosophy of language. 7. On a more peripheral note, it is common knowledge that another member of the Brentanian tradition, Adolf Reinach, is often considered a forerunner of the Oxonian ‘speech act theory’ (see e.g. Mulligan 1987; Nehrlich  and Clarke 1996). Whereas this is a good illustration of the fruitfulness of the Brentanian approach to language, Reinach’s proto-­ theory of speech acts is too isolated to be considered representative of the Brentanian tradition in general.

References Albertazzi, Liliana. 1989. Brentano’s and Mauthner’s Sprachkritik. Brentano Studien 2: 145–157. Albertazzi, Liliana, Massimo Libardi, and Roberto Poli, eds. 1996. The School of Franz Brentano. Dordrecht: Springer. Benoist, Jocelyn. 2003. Sprachkritik ou sémantique. Sur le schisme de l’école brentanienne. Les Études Philosophiques 1: 35–52. Boghossian, Paul. 1994. The Transparency of Mental Content. Philosophical Perspectives 8: 33–50. Brentano, Franz. 1925. In Psychologie vom empirischen Standpunkt, ed. Oskar Kraus, vol. 2. Leipzig: Meiner. ———. 1966. Die Abkehr vom Nichtrealen. Hamburg: Felix Meiner. ———. 1995. Psychology from an Empirical Standpoint. Trans. Antos C. Rancurello, D.B. Terrell, and Linda L. McAlister. London: Routledge. ———. 2009. The True and the Evident. Trans. Roderick M.  Chisholm, Ilse Politzer and Kart Fischer. Taylor and Francis e-library. ———. n.d.-a. Ms. EL 72. Logik. Cambridge (Mass.): Houghton Library. ———. n.d.-b. Ms. EL 80. Logik. Unpublished Manuscript. Provisional Online Edition. ed. Robin D. Rollinger. Franz Brentano Archiv Graz. http://gams. uni-graz.at/archive/objects/context:bag/methods/sdef:Context/ get?mode=logik. Bühler, Karl. 1934. Sprachtheorie. Jena: Fischer.

28 

A. Dewalque et al.

———. 2011. Theory of Language. The Representational Function of Language. Trans. Donald Fraser Goodwin. Amsterdam and Philadelphia: John Benjamins. ———. 2012. In Schriften Zur Sprachtheorie, ed. Achim Eschbach. Tübingen: Mohr Siebeck. Carnap, Rudolf. 1931. Überwindung der Metaphysik durch logische Analyse der Sprache. Erkenntnis 2: 219–241. Cesalli, Laurent. 2013. Marty’s Intentionalist Theory of Meaning. In Themes from Brentano, ed. Denis Fisette and Guillaume Fréchette, 139–163. Amsterdam and New York: Rodopi. Cesalli, Laurent, and Janette Friedrich, eds. 2014. Anton Marty and Karl Bühler. Between Mind and Language. Basel: Schwabe. Cesalli, Laurent, and Kevin Mulligan. 2017. Marty and Brentano. In The Routledge Handbook of Brentano and the Brentano School, ed. Uriah Kriegel, 251–263. New York and London: Routledge. Chisholm, Roderick M., and Wilfrid Sellars. 1972. The Chisholm-Sellars Correspondence on Intentionality. In Intentionality, Mind, and Language, ed. Ausonio Marras, 214–248. Urbana, Chicago and London: University Of Illinois Press. Dewalque, Arnaud. 2017. The Unity of the Brentano School. In Handbook of Brentano and the Brentano School, ed. Uriah Kriegel, 236–248. New York and London: Routledge. Dewalque, Arnaud, and Venanzio Raspa, eds. 2019. Psychological Themes in the School of Alexius Meinong (Meinong Studies 10). Berlin: De Gruyter. Fennell, John. 2019. A Critical Introduction to the Philosophy of Language. New York and London: Routledge. Fisette, Denis, and Guillaume Fréchette. 2007. À l’école de Brentano. De Würzbourg à Vienne. Paris: Vrin. ———. 2013. Themes from Brentano. Amsterdam and New York: Rodopi. Fréchette, Guillaume. 2019. The Origins of Phenomenology in Austro-German Philosophy: Brentano and Husserl. In A Companion to Nineteenth Century Philosophy, ed. John Sand, 418–453. Oxford: Wiley Blackwell. Fréchette, Guillaume, and Hamid Taieb, eds. 2017. Mind and Language. On the Philosophy of Anton Marty. Berlin: De Gruyter. Frege, Gottlob. 1884. Grundlagen der Arithmetik. Breslau: Koebner. ———. 1990. Kleine Schriften. 2nd ed. Hildesheim: Olms. Grice, Paul. 1957. Meaning. The Philosophical Review 66 (3): 377–388. Haller, Rudolf. 1979. Studien zur österreichischen Philosophie. Amsterdam: Rodopi.

1  Introduction: Mind, Meaning and Reality 

29

Huemer, Wolfgang. 2019. Is Brentano’s Method a Unifying Element of the Brentano School? Rivista Di Filosofia Neo-Scolastica CXI (4): 897–910. Ingarden, Roman. 1965. Das literarische Kunstwerk. Tübingen: Max Niemeyer. Kotarbiński, Tadeusz. 1976. Franz Brentano as Reist. In The Philosophy of Franz Brentano, ed. Linda L. McAlister, 194–203. London: Duckworth. Kraus, Oskar. 1929. Selbstdarstellung. Leipzig: Meiner. ———. 1934. Wege und Abwege der Philosophie. Prague: Robert Lerche. Kriegel, Uriah, ed. 2017a. The Routledge Handbook of Franz Brentano and the Brentano School. London and New York: Routledge. ———. 2017b. Brentano’s Philosophical Program. In The Routledge Handbook of Franz Brentano and the Brentano School, ed. Uriah Kriegel, 21–32. London: Routledge. Leblanc, Hélène, and Giuliano Bacigalupo, eds. 2019. Anton Marty and Contemporary Philosophy. London: Palgrave Macmillan. Liedtke, Frank. 1990. Meaning and Expression: Marty and Grice on Intentional Semantics. In Mind, Meaning and Metaphysics: The Philosophy and Theory of Language of Anton Marty, ed. Kevin Mulligan, 29–49. Dordrecht: Kluwer. Longworth, Guy. 2017. Grice and Marty on Expression. In Mind and Language. On the Philosophy of Anton Marty, ed. Guillaume Fréchette and Hamid Taieb, 263–284. Berlin: De Gruyter. Martinak, Eduard. 1901. Psychologische Untersuchungen Zur Bedeutungslehre. Leipzig: Barth. Marty, Anton. 1873. Letter from Marty to Brentano. May 23th 1873. Ms. Cambridge, MA: Houghton Library. ———. 1908. Untersuchungen Zur Grundlegung Der Allgemeinen Grammatik Und Sprachphilosophie. Halle: Niemeyer. ———. 1918. Über subjektlose Sätze und das Verhältnis der Grammatik zu Logik und Psychologie IV, V, VI and VII. In Gesammelte Schriften II.1, ed. Josef Eisenmeier, Alfred Kastil, and Oskar Kraus. Halle: Niemeyer. Mauthner, Fritz. 1901–1902. Beiträge zu einer Kritik der Sprache. Vol. 3. Stuttgart: J.G. Cotta. ———. 1906. Die Sprache. Frankfurt am Main: Rütten u. Loening. Mayer-Hillebrand, Franziska. 1977. Einleitung. In Brentano, Franz, 1977. Die Abkehr vom Nichtrealen, ed. Franziska Mayer-Hillebrand, 1–99. Hamburg: Meiner. Moltmann, Friederike, and Mark Textor. 2017. Act-based Conceptions of Propositional Content. Oxford: Oxford University Press.

30 

A. Dewalque et al.

Mulligan, Kevin, ed. 1987. Speech Act and Sachverhalt. Reinach and the Foundations of Realist Phenomenology. Dordrecht: Springer. ———. 1990. Mind, Meaning and Metaphysics: The Philosophy and Theory of Language of Anton Marty. Dordrecht: Springer. ———. 2012. Wittgenstein et la philosophie austro-allemande. Paris: Vrin. ———. 2019. Misleading Pictures, Temptations and Meta-Philosophies: Marty and Wittgenstein. In Anton Marty and Contemporary Philosophy, ed. Giuliano Bacigalupo and Hélène Leblanc, 197–232. London: Palgrave Macmillan. Nehrlich, Brigitte, David Clarke. 1996. Language, Action and Context. The early History of Pragmatics in Europe and America. Amsterdam: John Benjamins Publishing Company. Poli, Roberto. 1998. The Brentano Puzzle: An Introduction. In The Brentano Puzzle, ed. Roberto Poli, 1–13. Aldershot: Ashgate. Recanati, François. 2012. Mental Files. Oxford: Oxford University Press. ———. 2019. Natural Meaning and the Foundations of Human Communication: A Comparison Between Marty and Grice. In Anton Marty and Contemporary Philosophy, ed. Giuliano Bacigalupo and Hélène Leblanc, 197–232. London: Palgrave Macmillan. Rollinger, Robin D. 2009. Brentano’s Logic and Marty’s Early Philosophy of Language. Brentano Studien 12: 77–98. ———. 2014. Brentano and Marty on Logical Names and Linguistic Fictions: A Parting of Ways in the Philosophy of Language. In Anton Marty and Karl Bühler. Between Mind and Language, ed. Laurent Cesalli and Janette Friedrich, 167–200. Basel: Schwabe. Schaar, Maria van der. 2013. G.F. Stout and the Psychological Origins of Analytic Philosophy. London: Palgrave Macmillan. Schuhmann, Karl. 2004. Brentano’s Impact on Twentieth-Century Philosophy. In The Cambridge Companion to Brentano, ed. Dale Jacquette, 277–297. Cambridge: Cambridge University Press. Searle, John. 1983. Intentionality. Cambridge: Cambridge University Press. ———. 2004. Mind. A Brief Introduction. New  York; Oxford: Oxford University Press. Simons, Peter. 1987. Brentano’s Reform of Logic. Topoi 6: 25–38. Smith, Barry. 1994. Austrian Philosophy: The Legacy of Franz Brentano. Chicago: Open Court. Soames, Scott. 2013. Cognitive Propositions. Philosophical Perspectives 27 (1): 479–501. Srzednicki, Jan. 1965. Franz Brentano’s Analysis of Truth. The Hague: Nijhoff.

1  Introduction: Mind, Meaning and Reality 

31

———. 1966. Some Elements of Brentano’s Analysis of Language and their Ramification. Revue Internationale de Philosophie 20 (4): 434–445. Taieb, Hamid. in press. Ordinary Language Semantics: The Contribution of Brentano and Marty. British Journal for the History of Philosophy. Textor, Mark. 2006. Introduction. In The Austrian Contribution to Analytic Philosophy, ed. Mark Textor, 1–19. London and New York: Routledge. Twardowski, Kazimierz. 1977. On the Content and Object of Presentations. A Psychological Investigation. Trans. R. Grossmann. The Hague: Nijhoff. ———. 1999. On the Logic of Adjectives (1923/27). In On Actions, Products, and Other Topics in Philosophy, ed. Johannes Brandl and Jan Woleński and Trans. Arthur Szylewicz, 141–143. Amsterdam: Rodopi. ———. 2016. In Logik: Wiener Logikkolleg 1894/95, ed. Arianna Betti and Venanzio Raspa. Berlin: De Gruyter. Weiler, Gershon. 1970. Mauthner’s Critique of Language. Cambridge: Cambridge University Press. Wittgenstein, Ludwig. 1922. Tractatus Logico-Philosophicus. Trans. Charles Kay Ogden. London: Routledge and Kegan Paul. ———. 1953. Philosophische Untersuchungen/Philosophical Investigations. Trans. Elizabeth Anscombe. Oxford: Blackwell.

Part I Brentano and Philosophy of Language

2 The Context Principle in Austro-German Philosophy Guillaume Fréchette

Frege’s context principle is often seen as the primary definitional feature of analytic philosophy. In the following, I will refer to this view as the big picture thesis. Many prominent analytic philosophers of the twentieth century defended the big picture thesis. This is the case for Michael Dummett, Peter Hacker and Willard van Orman Quine, among others. Exactly what is the motivation behind the big picture thesis? In order to answer this question, I will first return to Frege’s presentation and discussion of the context principle in the Grundlagen der Arithmetik (Frege 1884/1953). Second, I will suggest that those defending the big picture thesis are using the expression ‘context principle’ to refer to different principles or theses. While these principles or theses are clearly related to one another, they do not always imply one another. For some philosophers, Dummett (1993) most prominently, the context principle implies a reductionist view of the analysis of thought. For others, such as Quine (1969), it implies that in order to explain a term, we must only show how to translate the complete sentences in which it is used. Finally, many

G. Fréchette (*) Department of Philosophy, University of Geneva, Geneva, Switzerland © The Author(s) 2021 A. Dewalque et al. (eds.), Philosophy of Language in the Brentano School, History of Analytic Philosophy, https://doi.org/10.1007/978-3-030-52211-7_2

35

36 

G. Fréchette

philosophers influenced by the late Wittgenstein, such as Hacker (1997), prefer to see the context principle as the illustration of the Wittgensteinian view that the sentence is the minimal move in a language game. I will argue that some of these understandings of the context principle were already defended by Austro-German philosophers from Bolzano to Bühler. However, none of these Austro-German philosophers defended a reductionist thesis regarding the analysis of thought. For this reason, I would suggest that if the context principle is a defining feature of analytic philosophy, as suggested by the big picture thesis, this is not the case in virtue of Dummett’s reductionist reading of the principle but rather in virtue of the readings corresponding to some of the theses on language defended by Austro-German philosophers.

1

 he Context Principle in Frege’s T Grundlagen

According to Dummett (1993, 4), one takes the linguistic turn when one recognises [F]irst, that a philosophical account of thought can be attained through a philosophical account of language, and, secondly, that a comprehensive account can only be so attained. (Dummett 1993, 4)

Although Frege did not explicitly make this turn, here, Dummett refers to Frege’s introduction of the context principle in the Grundlagen in 1884 as an introduction of the linguistic turn. Addressing the question of what number words mean, following Dummett’s suggestive translation, Frege invokes [T]he thesis that it is only in the context of a sentence that a word has a meaning: the investigation therefore takes the form of asking how we can fix the senses of sentences containing words for numbers. (Dummett 1993, 5)

Kenny agrees with Dummett: the linguistic turn finds its origins in Frege’s Grundlagen of 1884, that is,

2  The Context Principle in Austro-German Philosophy 

37

[…] when Frege decided that the way to investigate the nature of number was to analyse sentences in which numerals occur. (Kenny 1995, 211)

At this point, it might be useful to return to exactly what Frege says in the Grundlagen: One must ask about the meaning of words in the context of the sentence, not individually.1 (Frege 1884, x, my transl.)

Frege’s context principle makes its first appearance in 1884, in the introduction to the Grundlagen der Arithmetik as the second of three fundamental principles. Frege questions the pertinence of genetic definitions in arithmetic. Such genetic definitions, as explanations for the origins (Entstehung) of our concepts, lead, at best, to Mill’s arithmetic of Pfefferkuchen and Kieselsteine but not to arithmetic as such, and they decisively challenge the knowability of the world: If everything were in continual flux, and nothing maintained itself fixed for all time, there would no longer be any possibility of getting to know anything about the world and everything would be plunged in confusion. (Frege 1953, xix; 1884, vii)

A straightforward way to put this, without the Heraclitean Anspielung, is to say that psychologism leads to idealism. What is the problem with genetic definitions? Some people [e.g. J.S. Mill] suppose, it would seem, that concepts sprout in the individual mind like leaves on a tree, and that they get to know their essence by studying their origins; and they seek to define them psychologically in terms of the nature of the human mind.2 (Frege 1953, xix, modified transl.)

If we wish to avoid psychologism and, with it, idealism, then we must separate the psychological from the logical. This is the first of the three principles introduced in the Grundlagen. To put it briefly, the why of ‘origin’ is not the why of ‘essence’. One shouldn’t confuse these.

38 

G. Fréchette

What should we do? This is where the second principle, the context principle, is introduced, as an antidote to psychologism: ‘ask for the meaning of words in the context of the proposition, not individually’. How is that an antidote to psychologism? In § 60, we find part of the answer: That we can form no idea of the content of a word is therefore no reason for denying all meaning to a word, or for excluding it from our usage. The opposite view seems to originate from the fact that we consider the words individually and ask for their meaning, which leads us to take an idea as the meaning. Accordingly, a word for which we can find no corresponding inner picture appears to have no content.3 (Frege 1953, 71, modified transl.)

In other words, as a psychologistic mathematician, one may deny that an expression such as ‘the square root of minus one’ means anything because one can’t form an idea of this number, as in the case of natural numbers. However, this comes from one’s searching for the meaning of ‘ −1 ’ in isolation. Prima facie, it seems reasonable to understand the context principle, as presented by Frege, as a useful tool to use in avoiding psychologism, and it also seems reasonable to extend the scope of its application to the meaning of words in general. It is also an antidote to idealism since it presupposes the knowability of the world: given that the world is knowable and that there is something eternal and constant in the flow of things, relying on internal images when asking for the meaning of a word is utterly suboptimal. One must consider the proposition (what is firm and eternal in the constant flow of things) in which the word occurs.

2

Some Readings of the CP

Starting with Frege himself and following Dummett’s intuition regarding the context principle (CP) as the defining feature of analytic philosophy, we may isolate three basic interpretations of the CP:

2  The Context Principle in Austro-German Philosophy 

39

CP1: The CP is an antipsychologistic/anti-idealistic antidote: to understand the meaning of ‘ −1 ’ or any other subsentential expression, do not consider the mental images that one forms while using the expression but look at what the sentences including the expression ‘ −1 ’ (or any other subsentential expression) have in common. This is the gist of the idea expressed in Frege (1953, 71). CP2: The CP is the defining feature of analytic philosophy because it states that the analysis of thought is carried out via the analysis of language— because it allows for the linguistic turn. This is the idea expressed by Dummett (1993, 4). CP3: The CP is a metaphysical principle stating the conceptual priority of thoughts over their constituents (Dummett 1991, 184).

CP1 is simply the reformulation of the context principle as presented by Frege in the Grundlagen. In defending CP2 and CP3, Dummett starts with CP1 but adds a reductionist thesis regarding metaphilosophy, namely the notion that the analysis of thought should be carried out via linguistic analysis: CP2. If the meaning of a subsentential expression is determined by the sentential context (CP1) and the analysis of thought is to be carried out via the analysis of language (CP2), then the semantic priority of sentential expressions over subsentential ones must be reflected by the conceptual priority of thoughts over their constituents (CP3). Therefore, CP3 may be understood as the psychological application of the linguistic thesis that sentential expressions are conceptually, or semantically, prior to subsentential ones. However, Dummett has more to say on the context principle, offering further interpretations of the principle in various places. Other interpretations of the principle have been proposed elsewhere.4 For the purpose of this paper, I will focus on three readings of the principle: CP4: The CP lays the ground for contextual definitions in semantics: the primary vehicle of meaning is not the word, but the sentence. In Quine’s words, ‘to explain a term we need only show, by whatever means, how to translate the whole sentences in which the term is to be used’ (Quine 1969, 72). CP5: The CP as a psychological principle: ‘it is possible to grasp the sense of a word only as it occurs in some particular sentence’ (Dummett 1993, 97).

40 

G. Fréchette

CP6: The CP as a thesis on communication (Wittgenstein, Dummett, Hacker): ‘A sentence is, as we have said, the smallest unit of language with which a linguistic act can be accomplished, with which a ‘move can be made in the language-game’: so you cannot do anything with a word—cannot effect any conventional (linguistic) act by uttering it—save by uttering some sentence containing that word’ (Dummett 1973, 194). ‘[Bentham’s version of the context principle] rightly stresses that the sentence is, as Wittgenstein was later to argue, the minimal move in the language game’ (Hacker 1997, 67).

CP1–CP6 show that the CP can be understood in many ways. CP2 is by far the most radical interpretation because it reduces the analysis of thought to the analysis of language. However, one can endorse a version of the CP without endorsing the reductionist claim suggested by Dummett. In fact, it seems that only CP2 and CP3 imply a reductionist view of the analysis of thought. Interestingly, these seem to be the only two readings that are rejected by most Austro-German philosophers, often quite independently of Frege. Before attempting to explain this fact, let us first have a look at the readings and their Austro-German ancestors.

3

Contextual Definitions and Paraphrases

CP4 does not focus on the CP as the announcement of the linguistic turn but rather as the expression of a semantic innovation, namely the idea of contextual definition in semantics, first formulated by Bentham in his theory of fictions. Quine writes in Epistemology Naturalized: Bentham’s [innovation] was the recognition of contextual definition, or what he called paraphrasis. He recognized that to explain a term we do not need to specify an object for it to refer to, nor even specify a synonymous word or phrase; we need only show, by whatever means, how to translate the whole sentences in which the term is to be used. […] This idea of contextual definition, or recognition of the sentence as the primary vehicle of meaning, was indispensable to the ensuing developments in the f­ oundations of mathematics. It was explicit in Frege, and it attained its full expression in Russell’s doctrine of descriptions as incomplete symbols. […]

2  The Context Principle in Austro-German Philosophy 

41

Contextual definitions precipitated a revolution in semantics. […] The primary vehicle of meaning is seen no longer as the word, but as the sentence. Terms, like grammatical particles, mean by contributing to the meaning of the sentences that contain them. […] It was the recognition of this semantic primacy of sentences that gave us contextual definitions, and vice versa. I attribute this to Bentham. Generations later we find Frege celebrating the semantic primacy of sentences, and Russell giving contextual definitions its fullest exploitation in technical logic. (Quine 1969, 72)

The focus bestowed upon Frege’s context principle among analytic philosophers does not merely go back to Dummett and Quine. Quine, at least, recognised that he learned from Bentham in the 1932 edition of the Theory of Fictions by Ogden (Bentham and Ogden 1932). More importantly, Ogden himself borrowed from Wisdom’s book on Bentham from 1931 (Wisdom 1931), in which Wisdom first coined the phrase ‘analytic philosophers’ as a description of those philosophers who, like Bentham, use contextual definitions (in Bentham, ‘paraphrases’) to conduct eliminativist work.5 What does Bentham include under paraphrase, or contextual definition? Take the word ‘obligation’, which in Bentham’s ontology, is the name of a fictitious entity. One determines the meaning of such fictitious terms from the context in which they occur, i.e. by giving a paraphrase in which the term does not occur, such as in S* as a paraphrase of S: SBentham: ‘You have the obligation to pay the income tax on your salary’; SBentham*: ‘If you don’t pay the income tax on your salary, you will get fined by the Ministry of Finance’.

Therefore, in order to determine the meaning of ‘obligation’, one must only provide the context in which it occurs. There are different ways of giving the context in which a term occurs. In the case of S and S*, the meaning of ‘the obligation for you to F’ is nothing else than the meaning expressed by ‘the consequence that will inevitably occur (i.e. getting fined by the Ministry of Finance) because of your not doing F’.

42 

G. Fréchette

In 1810, Bernard Bolzano had already formulated the gist of ‘Bentham’s innovation’ in his Beyträge: […] how does [the mathematician] begin to reach an understanding with his readers about such simple concepts and the word that he chooses for their designation? This is not a great difficulty. […] To distinguish such explications [Verständigungen] from a real definition we could call them designations or paraphrases [Umschreibungen]. They also belong to the class of conventions, insofar as they are intended only to supply a certain concept with its particular symbol. They would thus be the first thing with which any scientific exposition must begin insofar as it has simple concepts. (Bolzano 2004, 107; 1810, 55–56)

Bolzano’s concept of Verständigung is developed in more detail in the fourth volume of the Wissenschaftslehre. There, the idea of context takes an important place: It is well known that when we encounter a sign which was unknown to us up to that point, in connection with several others whose meanings are known, we often find ourselves in a position to determine with more or less precision what the former represents based only on the assumption that the author did not intend to express anything obviously absurd. In such cases one says that we have recognized the meaning of the sign from its use or from the context.6 (Bolzano 2014, IV, 383–384; 1837, IV, 547)

Here, Bolzano suggests that one may explain the meaning of a term, say of ‘gander’, to someone who does not know the term by using it in different contexts. One could, for instance, explain the meaning of the term ‘gander’ when saying to someone who doesn’t know what a gander is: ‘a gander is to a goose what a drake is to a duck, what a boar is to a pig…’ Although, strictly speaking, they are paraphrases, Bolzano’s Verständigungen have a mainly heuristic function: they are meant as an alternative to conceptual analysis in cases in which the concepts in question are simple or we are unsure about their proper analysis; they are not meant as a tool for ontological parsimony. However, Bolzano also uses paraphrases in the latter sense. In order to eliminate terms from sentences

2  The Context Principle in Austro-German Philosophy 

43

that misleadingly appear to have a referential function, such as the term ‘nothing’ in the sentence ‘nothing is more certain than death’, Bolzano proposes the following paraphrase: SBolzano: ‘nothing is more certain than death’; SBolzano*: ‘the idea of something that would be more certain than death has no object’.7

Reminiscent of Bentham, Brentano also proposed a paraphrase strategy to deal with names of fictions. The point is formulated in Brentano’s letter to John Stuart Mill from 1873, which is partly reproduced in Psychology from an Empirical Standpoint: The proposition, ‘A centaur is a poetic fiction’, does not imply, as you rightly point out, that a centaur exists, rather it implies the opposite. But if it is true, it does imply that something else exists, namely a poetic fiction. […] The truth of the proposition does not require that there be a [centaur], but it does require that there be something else. (Brentano 1995 [1874], 70; 286 f )

For Mill, such propositions indicate that the copula ‘is’ does not always express existence, therefore distinguishing between two uses of ‘is’, merely nominal in the case of centaurs and existential in the case of judgements about existing things. Brentano’s answer to Mill is that this distinction is unnecessary: there is only the existential use of the copula, but the predicate ‘…is a poetic fiction’ that does not function attributively, as predicates usually do, but in a modifying way, as predicates such as ‘…is dead’ or ‘…is painted’ do. The modifying function of the predicate is such that it does not predicate the thing which the sentence seems to be about but something else, in this case the imagined centaur thought of by the poets. Here, Brentano’s paraphrase strategy works in the following way. A true judgement such as: SBrentano: ‘a centaur is a poetic fiction’ must be paraphrased by: SBrentano*: ‘presented-centaurs-in-the-poet’s-mind exist’.

44 

G. Fréchette

Here, the bottom line of Brentano’s answer to Mill seems to be the following: ‘centaur’, just like ‘table’ and ‘chair’, is a name for an intentional object insofar as these name objects of presentations. It is, however, only in the context of a judgement, in which the attributive and modifying functions of attributes come into play, that names receive their existential import; i.e. they may be said to refer to something. Similarly, true negative existential judgements about centaurs are dealt with by Brentano in a way reminiscent of Bolzano’s paraphrase strategy for sentences containing terms with no referential function. For Brentano, S1Brentano: ‘centaurs do not exist’ must be paraphrased by S1Brentano*: ‘whoever judges that centaurs exist judges falsely (incorrectly)’.

Pace Quine, it seems that Bentham’s paraphrase strategy is not really Bentham’s innovation but rather a strategy that was developed in Austro-­ German philosophy even before Bentham. Bolzano used a paraphrastic strategy in 1810 to show that we can grasp the meaning of terms expressing simple concepts even if we cannot define them by producing a paraphrase, or paraphrases, that is, ‘several propositions in which the concept to be introduced appears in different combinations and is designated by its own word’ (Bolzano 2004, 107/1810, 56). The concept of a point is such a simple one in Bolzano’s view: we can understand its meaning by using paraphrases such as ‘the point is the simple in space’, ‘the point is the boundary of a line and yet no part of the line’, etc.8 Therefore, if the recognition of contextual definitions showed the semantic primacy of sentences over words, as Quine suggests, then it is an innovation that we owe to Bolzano. Even if we take as an essential feature of the innovation the fact that we thereby ‘do not need to specify an object for it [the term] to refer to’ (Quine 1969, 72), we also find an Austro-German pendant to the view in the conceptions of the paraphrase developed by Bolzano in (1837) and by Brentano in (1874, 1995). If, as Quine suggests, Frege’s context principle has its source in the idea of contextual definition, Frege’s context principle goes back at least as much to Bolzano and Brentano as it does to Bentham.9

2  The Context Principle in Austro-German Philosophy 

4

45

Grasping and Communicating

Let us now turn to the last two theses. Prima facie, it seems reasonable to consider CP5 and CP6 as two theses that might be defended separately. While CP5 is a psychological thesis about grasping the sense of words, CP6 is a thesis about communication. At least two facts suggest that these readings belong together. First, for someone like Dummett, who defends CP2, it simply follows from CP2 that CP5 and CP6 come together: if the analysis of thought is reducible to the analysis of language, grasping a sense and communicating a sense work on the same basis, that is, by using sentences. Second, one important feature of the Austro-German philosophy of language, in contrast with the prevalent view in German philosophy, was the empiricist view of the origins of language: while philosophers and psychologists such as Steinthal, Wundt and Humboldt championed the view that language is based on a number of onomatopoeic sounds and gestures naturally (innately) triggered by some psychophysical mechanism, Austro-German philosophers defend the view that language developed out of the intention of primitive humans to express their mental phenomena and, analogously, to express information about physical events.10 In short, for Marty and many Austro-Germans, because the essence of language is communication, grasping the sense of a word also means grasping what the person expressing the name ‘means’ (meinen) by it. The Austro-German position fully comports with the analytic readings of CP in terms of CP5 and CP6 but does not imply the reduction of thought to language, as in CP2. This letter from Marty to Brentano from 1873, two years before the publication of his dissertation, sums up the issue at stake here nicely: On the question concerning the meaning of the words, I tried not to go from the names to the statements, but rather the other way around. In this, I was driven by the reflection that only statements are actually complete linguistic expressions. The function and the ‘meaning’ [Bedeuten] that belongs to it are precisely this ‘meaning’ [Bedeuten], in the way we must expect it of a linguistic expression according to the essence and goal of language, that is, the manifestation of something [Kundgabe von etwas]. By contrast, the name does not manifest anything, it signifies [zu erkennen geben] nothing and the way in which it means something is conceivable

46 

G. Fréchette

only as a part of the function of assertion. However, I do not have complete trust in this. It may be simpler and more natural to go the other way around. Indeed, it does not seem right to consider the presentations naturally as prior to the judgement in the same way as one considers names as prior to statements, because the presentation is thinkable without the judgement, while by contrast names do not seem thinkable in language without being a part of a statement. Because the goal of language is communication, indication, and this is not conceivable without judging creatures and through statements. However, because the judged content is like the presented content, it seems more natural and simpler to speak first of the presented and the named.11 (Marty 1873, my transl.)

Meaning (Bedeuten) as the ‘manifestation of something’ (Kundgabe von etwas) is, for Marty and many Austro-Germans, what only a complete linguistic expression—a statement (Aussage)—does. Although names are linguistic expressions of presentation contents and a presentation content is thinkable without a judgement, names only obtain meaning in the context of a judgement. We find a similar distinction in Bolzano between what he calls the meaning of a sign (Bedeutung eines Zeichens), which corresponds roughly to Marty’s presentation content,12 and the sense of a sign (Sinn eines Zeichens), which corresponds to Marty’s Bedeuten: We could call the meaning [Bedeutung] of the sign is that particular idea which it is intended to produce, and succeeds in producing; while we could confine the words sense (Sinn) and understanding [Verstand] to those ideas which are merely intended to be produced by a certain sign in a given case.13 (Bolzano 2014, III, 44, modified transl.)

In other words, a word ‘w’ is usually determined to express the objective idea [w]. However, sometimes, we also use ‘w’ to awake in our interlocutor a different objective idea, say [x], or more importantly, a proposition such as [x has B]. If I hear someone shouting ‘hello’ from a well, I could, of course, think of a greeting and greet him or her back with another ‘hello’, but given the context, I would normally take this ‘hello’ to mean that someone needs help. Bolzano goes even further: in such a situation, even hearing a word yet unknown to us would lead us to the Bedeuten expressed by ‘that one should hurry up to help him/her’.

2  The Context Principle in Austro-German Philosophy 

47

Be the word a man who suddenly finds himself in mortal danger calls out to us ever so unknown, we may still conclude quite confidently that he intends to tell us that we should help him.14 (Bolzano 2014, III, 360; 1837, III, 545)

Concluding that Ivo, by voicing the word ‘Upomoć!’ yet unknown to his monolingual English hearer, means the proposition that this hearer should come to help him, is what Bolzano calls interpretation (Auslegung). This case is meant to illustrate that we often, or perhaps always, grasp the meaning of words by way of interpreting them in the context of a linguistic utterance thanks to the Sinn (Bolzano) or the Bedeuten (Marty) that an interlocutor wants to convey to us.

5

 Compromise between Brentano A and Frege

In the last section, we suggested that Austro-German philosophers believe that CP5 and CP6 belong together because they share the conviction that there is no natural or necessary connection between individual words and what they express. They do not endorse Dummett’s reductionism regarding the analysis of thought. Grasping, for oneself, the sense of a word, works in a similar way as communication. In a recent paper on emotivism, when discussing Oskar Kraus’s reading of Bertrand Russell’s The Analysis of Mind and noting that Kraus ‘expressed great irritation that Wittgenstein and Russell do not acknowledge […] (what he takes to be) Marty’s endorsement of the context principle in 1875 and later’, Mulligan (2017, 161) pointed out the first paternity dispute regarding the context principle between analytic philosophy and Austro-German philosophy. Kraus writes: Wittgenstein for instance holds: ‘only the sentence has a sense, only in the context of a sentence does a name have a meaning’—In Marty’s Origins of Language of 1875, in the Investigations of 1908, one finds the same theory. The name is characterised as a simple ‘theoretical autosemanticon’, only the complete discourse has the value of a practical autosemanticon.15 (Kraus 1934, 39, my transl.)

48 

G. Fréchette

Mulligan is careful in avoiding taking side on this, limiting himself to saying that Kraus takes Marty to be endorsing the context principle. In fact, Marty does endorse the context principle, but because he does not endorse Dummett’s reductionism—rather, he believes, like most of Brentano’s heirs, in the conceptual priority of presentations over judgements, i.e. in the thesis that all mental acts are either presentations or based on presentations—he restricts the use of the principle to language and communication and leaves open the possibility that names, at least in a theoretical sense, are ‘complete expressions of a presentation’.16 This was the concern expressed in his early letter to Brentano quoted above, in which Marty confessed that his intuition about sentences being the basic building blocks of language did not seem compatible, at least prima facie, with Brentano’s thesis about the conceptual priority of presentations over judgements. Marty’s compromise is to accept Brentano’s conceptual priority thesis and, with it, the idea that complete (or ‘autosemantic’) expressions are those that express a mental phenomenon, while circumscribing the cases in which a subsentential expression is an autosemantic expression to merely theoretical ones. This compromise might seem superficial because theoretical autosemantic nominal expressions are, in fact, never used as such in communication. Even Brentano himself, at least in 1905, was aware of this: One could not say that he who utters the word ‘horse’ conveys thereby that he has the idea of a horse. If he did not have it no one would say that he lied. (It is) also (the case) that no conversation is carried by mere calling of names, but much more by uttering (whole) sentences.17 (Brentano 1905; Srzednicki 1965, 117)

Moreover, because communication is the essence of language, according to Marty, theoretical autosemantic nominal expressions simply cannot, essentially, be used for a linguistic purpose. What purpose do they serve, if not simply to accommodate Brentano’s conceptual priority thesis? One plausible explanation for the compromise is that for most of Brentano’s heirs, meaning is something some expressions have in virtue of the fact that they are expressions of mental phenomena. In other words,

2  The Context Principle in Austro-German Philosophy 

49

expressions have meaning because they express the content of a mental phenomenon. This means that meanings, as linguistic entities, are derivative of content as a psychological entity. The distinction between expression and meaning is based, for Brentano and his heirs, on the more fundamental distinction between act and content. However, it also means that because mental phenomena are more fundamental than meanings, there is mental content that is not linguistically expressible. This is the case, for instance, with the content of intuitive presentations.18 This content is not linguistically expressible, because as Marty stresses, we can speak of communication and of meaning in the proper sense only when the same (1908, 433) mental content is produced in both the hearer and the speaker. The only way to achieve this is to use the content of conceptual presentations, which involve a description such as ‘the only primary colour between blue and yellow’, as a description of my experience of some red. Therefore, I can grasp the meaning of a subsentential expression, such as ‘the only primary colour between blue and yellow’, in a context of communication, and in a sentence, that is, when I say ‘red is the only primary colour between blue and yellow’, and this is the way we usually communicate meanings. However, the conceptual presentation voiced by ‘the only primary colour between blue and yellow’ has a meaning, at least as a so-called ‘theoretical’ autosemantic nominal expression, because it can be produced as the same mental content in both the hearer and the speaker in communicative contexts. Here, like Frege, Marty too uses the context principle as a tool to avoid psychologism.

6

F urther Austro-German Applications of the Context Principle

The compromise offered by the Brentanians between the conceptual priority of presentations, on the one hand, and the interpretations CP5 and CP6, on the other hand, shows that they agree with the context principle insofar as the mental processes of grasping a sense (CP5) and communicating (CP6) are involved as belonging together without implying the reductionist reading suggested by CP2. Marty’s compromise, as described

50 

G. Fréchette

in the above section, suggests a position compatible with CP1, namely with Frege’s concern about psychologism as the motivation for the context principle in the Grundlagen. Like Frege, Marty holds that one should not attempt to understand the meaning of a subsentential expression by considering the mental image we form while using the expression but rather by considering what sentences including the expression have in common. It is only in the context of a sentential expression that a subsentential expression may be said to express the same mental content in the hearer and the speaker. By rejecting Dummett’s reductionist thesis and defending the conceptual priority of presentations over expressions, the Austro-German conceptions of the context principle allow for a wider application of the principle, for instance, to non-linguistic contexts. This was already suggested by Brentano in the early Vienna lectures on logic, in which he stresses the pragmatic dimension of speech and thereby the non-linguistic context in which speech acts occur.19 In Logical Investigations, Husserl defends the similar idea that the perceptual context in some cases contributes to the meaning of an expression. This is the case in deictic expressions, which can only be meaningful in the perceptual context in which they are used: I say ‘this’, and now mean the paper lying before me. Perception is responsible for the relation of my word to this object, but the meaning does not lie in perception. When I say ‘this’, I do not merely perceive, but a new act of pointing (of this-meaning) builds itself on my perception, an act directed upon the latter and dependent on it, despite its difference. In this pointing reference, and in it alone, our meaning resides.20 (Husserl 2001, 197–198)

Years later, Bühler was clearly influenced by Husserl’s theory of indexicals in his Sprachtheorie. In fact, with Brentano and Marty, Bühler sees the linguistic context as only one sort of context in the larger realm of context in general (Umfeld), including verbal and non-verbal cases. If the linguistic context is perhaps the most important in a theory of language, it is not the only one:

2  The Context Principle in Austro-German Philosophy 

51

There is no need to prove to an expert that the most important and most interesting surrounding field of a language sign is its context; the individual sign appears in company with its fellows, and the company proves to have an effect as a surrounding field. In addition to this typical case, there are also two others; there are cases in which language signs occur without a context though by no means without a surrounding field [i.e. the sympractical and symphysical fields]. (Bühler 1990, 175–176; 1934, 154 f.)

The larger applicability of the context principle suggested by Austro-­ German philosophers is not the only advantage of their views over Dummett’s more radical understanding suggested in CP2: the Austro-­ German accounts of the principle discussed here stand closer to Frege’s motivation for the context principle, namely the need to have something ‘firm and eternal’ (etwas Festes, Ewiges) to warrant for the knowability of the world and to avoid psychologism. Marty’s concern about the sameness of the mental contents is one example, and Bolzano’s ‘propositions in themselves’ (Sätze an sich) or Husserl’s ‘meanings’ (Bedeutungen) perhaps even better ones. Not only do the Austro-Germans stand closer to Frege’s motivation for the context principle, but they also developed the very same tools as the ones identified by Quine as being the mark of the context principle, namely contextual definitions and paraphrases. If the context principle is in some important sense one of the defining features of analytic philosophy, then one should have a closer look at the Austro-German contribution before taking the linguistic turn.21

Notes 1. German original: ‘[N]ach der Bedeutung der Wörter muss im Satzzusammenhang, nicht in ihrer Vereinzelung gefragt werden’ (Frege 1884, x). Here and in the following quotations from texts originally written in German, I quote the original text only in cases where I modified the published translation or when the translation is my own. 2. German original: ‘Man [e.g. J.S. Mill] denkt sich, wie es scheint, dass die Begriffe in der einzelnen Seele so entstehen, wie die Blätter an den Bäumen und meint ihr Wesen dadurch erkennen zu können, dass man

52 

G. Fréchette

ihrer Entstehung nachforscht und sie aus der Natur der menschlichen Seele psychologisch zu erklären sucht’ (Frege 1884, vii). 3. Compare also (Frege 1953, 73 and 114 f.). German original: ‘Es ist also die Unvorstellbarkeit des Inhaltes eines Wortes kein Grund, ihm jede Bedeutung abzusprechen oder es vom Gebrauche auszuschließen. Der Schein des Gegentheils entsteht wohl dadurch, dass wir die Wörter vereinzelt betrachten und nach ihrer Bedeutung fragen, für welche wir dann eine Vorstellung nehmen. So scheint ein Wort keinen Inhalt zu haben, für welches uns ein entsprechendes inneres Bild fehlt’ (Frege 1884, 71). 4. Let me mention here en passant two other readings of the principle: First, the metasemantic reading derived from Dummett (1973, 495) and defended by Stainton (2006, 159), according to which the sense of an expression less than a complete sentence must consist only in the contribution it makes to determining the content of a sentence in which it may occur. In Stainton’s formulation: ‘the only things that have meaning nonderivatively are sentences, so it must be in virtue of their role within sentences that subsentential expressions have meaning at all’. Second, the reading of the principle favored by Wright (1983) and Hale (1987) and according to which the principle means that only in the context of sentences can we determine the reference of singular terms. 5. See (Wisdom 1931, 13 f.). On Wisdom and analytic philosophy, see (Beaney 2013). 6. German original: ‘Bekanntlich sind wir gar oft, wenn wir ein Zeichen, das uns bisher noch unbekannt war, in einer Verbindung mit mehreren anderen von bekannter Bedeutung antreffen, durch die bloße Voraussetzung, daß der Schreiber nichts offenbar Ungereimtes habe ausdrücken wollen, im Stande, mit mehr oder weniger Genauigkeit zu bestimmen, was er sich unter dem ersteren vorgestellt haben. In solchen Fällen sagt man, wir hätten die Bedeutung des Zeichens aus dem Gebrauche oder Zusammenhange erkannt’ (Bolzano 1837, IV, 547). 7. See (Bolzano 2014, II, 151 f.; 1837, II, 212 f.) 8. The sentences are taken from (Bolzano 2004, 107). 9. It is not clear whether Frege had any appreciation of Bentham, but he was at least aware of the school of Brentano: he wrote the Grundlagen following the advice of Stumpf, a student of Brentano, who thought that he should write a less technical introduction to his Begriffsschrift. See Stumpf ’s letter in (Frege 1976, 256).

2  The Context Principle in Austro-German Philosophy 

53

10. The best example of this opposition of the Austro-German to the German philosophy of language, both in its critical and constructive dimension, is (Marty 1875). 11. German original: ‘In der Frage nach der Bedeutung der Worte habe ich versucht, nicht von den Namen zu den Aussagen, sondern den umgekehrten Weg zu gehen. Dazu bestimmte mich die Überlegung, dass nur die Aussage eigentlich ein vollendeter sprachlicher Ausdruck ist. Die Funktion und das ‘Bedeuten’ das ihm zukommt ist gerade dasjenige “Bedeuten”, wie man es nach dem Wesen und Zweck der Sprache von einem sprachlichen Ausdruck erwarten muss, nämlich Kundgabe von etwas. Dagegen der Name teilt nichts mit, gibt nichts zu erkennen und die Art, wie er etwas bedeutet, begreift sich nur als Teil der Funktion der Aussage’. ‘Doch traue ich der Sache nicht vollkommen. Vielleicht ist es doch einfacher und natürlicher den umgekehrten Weg zu gehen. Zwar scheint es nicht richtig, dass wie man naturgemäss die Vorstellung betrachtet vor dem Urteil, in derselben Weise die Betrachtung des Namens vor der Aussage natürlicher erscheine. Denn die Vorstellung ist denkbar ohne das Urteil; dagegen die Namen in der Sprache scheinen nicht denkbar ohne als Teil der Aussage. Denn der Zweck der Sprache ist Mitteilung, Kundgabe und dies ist nicht denkbar ausser unter urteilenden Wesen und durch Aussagen. Aber da das Beurteilte dasselbe ist wie das Vorgestellte erscheint es doch natürlicher und einfacher, zuerst vom Vorgestellten und Genannten zu sprechen’. 12. One could of course object that Bolzano’s ‘Bedeutung eines Zeichens’ (meaning of a sign) is an idea in itself [Vorstellung an sich], i.e. a non-­ psychological entity, while Marty’s presentation content is a psychical phenomenon in Brentano’s sense. I answer to this objection on the nature of Marty’s presentation content in the next section. 13. German original: ‘[D]ie Bedeutung eines Zeichens [ist] nur diejenige objective Vorstellung […], zu deren Erweckung es bereits bestimmt ist, die es auch in der That zu erwecken pfleget; Sinn oder Verstand desselben aber diejenige, deren Erweckung wir in einem einzelnen Falle damit beabtichtigen’ (Bolzano 1837, III, 44). 14. German original: ‘So sey uns das Wort, welches wir einen, in plötzliche Lebensgefahr gerathenen Menschen uns zurufen hören, noch so unbekannt: wir können doch mit ziemlicher Sicherheit schließen, er wolle uns durch dasselbe bedeuten, daß wir zu seiner Hülfe herbeieilen sollen’ (Bolzano 1837, III, 545).

54 

G. Fréchette

15. German original: ‘Wittgenstein lehrt z.B. “Nur der Satz hat Sinn, nur im Zusammenhange des Satzes hat ein Name Bedeutung”—In Martys Ursprung der Sprache 1875, in den Untersuchungen 1908 findet sich die gleiche Lehre. Der Name wird als bloßes “theoretisches Autosemantikon” bezeichnet, nur die ganze Rede gilt als praktisches Autosemantikon’. 16. Compare with Marty (1908, 476). 17. This passage from (EL 66) has been inserted (with some modifications) in Mayer-Hillebrand’s edition of the logic lectures (Brentano 1956, 37). 18. See for instance (Marty 1908, 420): ‘Anschauungen im strengen Sinne sind nicht durch Sprache und Worte mitteilbar’. Compare also with (Marty 1908, 433). 19. ‘One often brings speaking in opposition to action. However, speaking is in itself an action, an activity through which one wants to awaken certain mental phenomena [dated April 23rd, 1875]’ (Brentano 1875, 12590, my translation). German original: ‘Man bringt oft Reden in Gegensatz zu Handeln. Aber Reden ist selbst ein Handeln, eine Tätigkeit, wodurch man gewisse psychische Phänomene hervorrufen will’. 20. See also (Mulligan 2012, 149) on this passage. 21. Many thanks to Laurent Cesalli, Claudio Majolino, Kevin Mulligan and the other participants of the Liège conference on Brentanian philosophy of language (September 2018) for their stimulating comments on the talk from which this paper originates. The research for this paper was financially supported by the Swiss National Science Foundation through the SNF project n° 100012_182858 on Réalismes : Universaux, relations et états de choses dans les traditions austro-allemande et médiévale.

References Beaney, Michael. 2013. The Historiography of Analytic Philosophy. In The Oxford Handbook of The History of Analytic Philosophy, ed. Michael Beaney, 30–60. Oxford: Oxford University Press. Bentham, Jeremy, and Charles Kay Ogden. 1932. Bentham’s Theory of Fictions. London: Kegan Paul. Bolzano, Bernard. 1810. Beyträge zu einer begründeteren Darstellung der Mathematik. Prague: Widtmann. ———. 1837. Wissenschaftslehre. Vol. 4. Sulzbach: Seidel’sche Buchhandung.

2  The Context Principle in Austro-German Philosophy 

55

———. 2004. Contributions to a Better-Grounded Presentation of Mathematics. In The Mathematical Works of Bernard Bolzano, Ed. and Trans. Steve Russ, 83–137. Oxford: Oxford University Press. ———. 2014. Theory of Science, 4 vols. Trans. Rolf George and Paul Rusnock. Oxford: Oxford University Press. Brentano, Franz. 1874. Psychologie vom empirischen Standpunkte. Leipzig: Duncker und Humblot. ———. 1956. In Die Lehre vom richtigen Urteil, ed. Franziska Mayer-Hillebrand. Bern: Francke Verlag. ———. 1995. Psychology from an Empirical Standpoint. Trans. Linda L. McAlister et al. London: Routledge. Bühler, Karl. 1934. Sprachtheorie. Die Darstellungsfunktion der Sprache. Jena: Fischer. ———. 1990. Theory of Language. The Representational Function of Language. Trans. Donald Fraser Goodwin. London: John Benjamins. Dummett, Michael. 1973. Frege: Philosophy of Language. Cambridge, MA: Harvard University Press. ———. 1991. Frege: Philosophy of Mathematics. Cambridge, MA: Harvard University Press. ———. 1993. Origins of Analytical Philosophy. London: Duckworth. Frege, Gottlob. 1884. Die Grundlagen der Arithmetik. Eine logisch-mathematische Untersuchung über den Begriff der Zahl. Breslau: Wilhelm Koebner. ———. 1953. The Foundations of Arithmetic. A Logico-Mathematical Enquiry into the Concept of Number, 2nd revised edition. Trans. John Langshaw Austin. New York: Harper and Brothers. ———. 1976. Wissenschaftlicher Briefwechsel. Hamburg: Meiner. Hacker, Peter M.S. 1997. The Rise of Twentieth Century Analytic Philosophy. In The Rise of Analytic Philosophy, ed. Hans-Johann Glock, 51–76. Oxford: Blackwell. Hale, Bob. 1987. Abstract Objects. Oxford: Basil Blackwell. Husserl, Edmund. 1901. Logische Untersuchungen. Zweiter Band. Elemente einer phänomenologischen Aufklärung der Erkenntnis. II. Teil. Halle: Niemeyer. ———. 2001. Logical Investigations, vol. 2. Trans. John N. Findlay. London: Routledge. Kenny, Anthony. 1995. Frege: An Introduction to the Founder of Modern Analytic Philosophy. London: Penguin Books. Kraus, Oskar. 1934. Wege und Abwege der Philosophie. Vorträge und Abhandlungen. Prague: Calve.

56 

G. Fréchette

Marty, Anton. 1875. Über den Ursprung der Sprache. Würzburg: A.  Stuber’s Buchhandlung. ———. 1908. Untersuchungen zur allgemeinen Grammatik und Sprachphilosophie. Halle: Max Niemeyer. Mulligan, Kevin. 2012. Wittgenstein et la philosophie austro-allemande. Paris: Vrin. ———. 2017. The Origins of Emotivism, Expressivism and the Error Theory: Marty, Scheler, Russell, Ogden and Richards. In Mind and Language. On the Philosophy of Anton Marty, ed. Guillaume Fréchette and Hamid Taieb, 149–168. Berlin: De Gruyter. Quine, Willard van Orman. 1969. Ontological Relativity and Other Essays. New York: Columbia University Press. Srzednicki, Jan. 1965. Franz Brentano’s Analysis of Truth. The Hague: Martinus Nijhoff. Stainton, Robert. 2006. The Context Principle. In Concise Encyclopedia of Semantics, ed. Keith Allan, 158–164. Amsterdam: Elsevier. Wisdom, John. 1931. Interpretation and Analysis in Relation to Bentham’s Theory of Definition. London: Kegan Paul. Wright, Crispin. 1983. Frege’s Conception of Numbers as Objects. Aberdeen: Aberdeen University Press. Archive Materials

Brentano, Franz. 1875. Logik-Vorlesung (EL 72). Franz Brentano Archives. Houghton Library: Harvard University. ———. 1905. Sprechen und Denken (EL 66). Franz Brentano Archives. Houghton Library: Harvard University. Marty, Anton. 1873. Brief von Marty an Brentano, Schwyz, 28. Mai 1873. Copy from the Franz Brentano Archiv in Graz (Austria).

3 A Context Principle in Brentano? Charlotte Gauvry

According to the standard narrative, the philosophers and psychologists of the Brentano School, Anton Marty excepted, were not interested in language. Their main concerns were rather psychological and above all metaphysical. This is particularly obvious in the case of Kazimierz Twardowski and Alexius Meinong. By contrast, it is currently held (see e.g. Coffa 1991; Dummett 1973; Simons 1992; Sluga 1980) that Gottlob Frege, and possibly Bernard Bolzano—i.e. the outstanding representatives of the semantics tradition—paved the way for philosophy of language as a stand-alone branch of philosophical inquiry. A supporting argument for this claim goes as follows: in order for philosophy of language to become a discipline, language has first to be regarded as a specific and autonomous topic of investigation, in particular with respect to ontological or psychological entities. Such condition would be necessary to consider philosophy of language, and not metaphysics or psychology, as the first philosophy (see e.g. Benoist 2003). This autonomy principle

C. Gauvry (*) Department of Philosophy, University of Liège, Liège, Belgium e-mail: [email protected] © The Author(s) 2021 A. Dewalque et al. (eds.), Philosophy of Language in the Brentano School, History of Analytic Philosophy, https://doi.org/10.1007/978-3-030-52211-7_3

57

58 

C. Gauvry

is indeed at work in Bolzano’s and in Frege’s ‘Platonism’. Yet, it is not operative in the Brentanian analyses of language. Philosophers of a Brentanian persuasion rather consider language as a preparatory tool for psychology (see e.g. Brentano 1995a, 33; see also the introduction to the present volume). As a consequence one could reasonably infer that there is no ‘philosophy of language’ in the Brentano School, at least not in the early analytical sense.1 In that respect, the leading question of this chapter (and possibly of the entire volume) could appear far-fetched: did Brentano somehow pave the way for Frege and Wittgenstein? More precisely, did he already introduce in his manuscripts on language an anticipation of the so-called ‘context principle’—famously defined by Gottlob Frege: ‘Never to ask for the meaning of a word in isolation, but only in the context of a proposition [im Satzszusammenhange]’ (Frege 1980, x)? The (historical) purpose of my chapter is to show that we are facing a paradox. Although considering Brentano as a ‘philosopher of language’ in the Frege-Wittgenstein sense is questionable—for reasons that I will dwell on below –, his dense and little-known manuscripts on language are full of ‘pragmatic’ insights, even fuller than Frege’s, Russell’s or the early Wittgenstein’s own works. Brentano actually anticipates some of the main claims of the later philosophers of ordinary language. The (systematic) purpose of this chapter is then to explore Brentano’s analyses on ordinary language in order to ask whether it makes sense to consider him a ‘contextualist’. To that aim, in Sect. 1, I will first assess the various evidences found in the literature to hold that Brentano is indeed a philosopher of language. Then, I will pay attention to Brentano’s specific linguistic claims, and in particular to his use of the distinction between ‘syncategorematic’ and ‘categorematic’ expressions.2 In Sect. 2, I will focus on ‘syncategorematic expressions’, whose meaning functions are sensitive to some syntactical considerations. In Sect. 3, I will examine further the notion of ‘categorematic expressions’, (also called ‘autosemantic expressions’) which manifests that the naming function of some expressions is a-contextual. Sect. 4 challenges this last conclusion. On the one hand, I will show that, for Brentano, even categorematic expressions require to be practically used, in a complete utterance, in order to have a complete meaning. On the other

3  A Context Principle in Brentano? 

59

hand, I will investigate whether such practical considerations are sufficient to make him a ‘contextualist’. Throughout this article, I will essentially focus on Brentano’s Logic manuscripts, which include the 1869–1871 Würzburg Lesson “Deduktive und Induktive Logik” (Ms. EL 80)—which was also taught in Vienna in 1875 and 1877 –, and the 1878–1885 Vienna Lesson “Die elementare Logik und die in ihr nötigen Reformen” (Ms. EL 72). Although still unpublished, these manuscripts are precious and provide a reliable source on Brentano’s work, contrary to the often quoted Franziska Mayer-­ Hillebrand’s compilation (see Brentano 1956), which includes numerous and not clearly identified fragments which are not from Brentano. Besides, although they cover almost twenty years, these lessons were all taught before Brentano’s so-called ‘reist’ turn and are thus rather homogeneous.

1

Brentano: A Philosopher of Language?

As made obvious in the introduction of the present volume, Brentano clearly endorses strong and original positions regarding language. In particular, he criticises the so-called nativist view that was defended by many of his influential predecessors, notably Wilhelm von Humboldt, Paul Johann von Heyse, Heymann Steinthal, and Wilhelm Wundt. According to the nativist theory, language is a mechanical secretion of thought that has no finality. By contrast, following some intuitions of John S. Mill, Brentano advocates a functionalist position.3 Even if he probably expresses it less clearly than Marty, Brentano considers that language has the function to express one’s own mental act and to influence someone else’s mental life.4 Besides, just as Mill and Alexander Bain before him, Brentano characterises language as an auxiliary of thought (Hilfsmittel des Denkens). Linguistic signs, like mathematical signs, are indeed helpful tools which help to register our thoughts (Ms. EL 80, 12.978[2]), 12.991[3]; Ms. EL 72, I.10) and to ease the work of our memory (Ms. EL 80, 12.991, 13.112[3]; Ms. EL 72, I.9). For instance, to use Brentano’s example, it is easier to remember the ‘Our Father’ prayer than all the dogmas those

60 

C. Gauvry

words rest on, or to use the sign ‘1 000 000’ in an addition rather than to express its specific meaning. Moreover, language signs proceed by distinction (Unterscheidungszeichen), and consequently introduce differences between the concepts (contrary to our thoughts that proceed by associations of ideas (Ideenassoziationen) (Ms. EL 80, 12.990[2]). As a consequence, the use of linguistic signs increases and consolidates our knowledge (Ms. EL 72, I.10). However, Brentano, again as Mill before him, underlines the disruptive function of ordinary language. Although useful, linguistic signs are indeed equivocal and introduce substantial confusions, for instance when the same sign is used to express different thoughts (homonymy case) (Ms. EL 80, 12.995[1]) or when two different signs are used to express the same thought (synonymy case) (Ms. EL 80, 12.996[2]). Ordinary language can even introduce categorical mistakes, for instance when we use the same word, say ‘thought’, to characterise both a ‘presentation’ and a ‘judgement’, which are two different categories of mental phenomena, but not to characterise ‘emotions’, ‘feelings’, ‘desires’ and ‘acts of will’, which compose the third category of mental phenomena (Ms. EL 72, I.10β; Brentano 1995a, 244 [73]). More generally linguistic grammar can be misleading, in particular when it suggests that a term is referential although it is not (universal, abstract terms, linguistic fictions, etc.). However, whether Brentano considers linguistic signs as helpful or disruptive, he defines them in terms of tools that one can use. Linguistic signs are even considered as the ‘main tools’ (Hauptwerkzeuge) of our psychological and philosophical analyses. As a result, Brentano pays great attention to the various uses that we can make of our language. To take the interesting example of linguistic fictions again, he holds that ‘for the sake of brevity’ or for ‘facility’ reason, one can use a linguistic fiction rather than a more complex periphrasis. In his 1901 Letter to Marty, he explicitly specifies that: linguistic fictions ‘are useful as ways of speaking’ (Brentano 2009, 44, my emphasis). As a consequence, Brentano’s conception of ordinary language is fairly nuanced. On the one hand, he clearly considers that ordinary language is delusive since it introduces linguistic and philosophical confusions. On the other hand, he also asserts that it is a helpful, if not the only, tool that the philosopher can use in order to undertake conceptual analyses. The

3  A Context Principle in Brentano? 

61

following quotation of the Psychologie provides a clear summary of his position: We have already observed that ordinary language prepared the way for psychological investigations by means of the general names it assigns to mental phenomena. Naturally ordinary language is not entirely reliable, and it would mislead those who depended upon it too much, just as it would facilitate the discovery of truth for those who utilize its definitions with caution. (Brentano 1995a, 33)

Here appears the paradox that I first mentioned. Brentano is obviously sensitive to the variety of linguistic uses and he does not repudiate ordinary language. He clearly rejects the idea of a Begriffsprache (quoting Boole or Jevons) or of an ideal or purified language (Ms. EL 72 I 16). In that respect, Brentano is even more attached to ordinary language than, for example, Frege and the other semantic philosophers who paved the way for the philosophy of ordinary language. This preliminary conclusion will become less puzzling as soon as one slightly shift the perspective on Brentano’s position. Although Brentano is highly interested in ordinary language, he never considers linguistic or even logical issues for their own sake. That is precisely the reason why language is always considered an (useful, necessary and sometimes misleading) auxiliary of the thought. Logic itself is not regarded as theoretical but as a practical method that helps clarifying the thought (see Brentano 1956, 3). In other words, even if Brentano is indeed interested in ordinary language and provides innovative analyses of its operations, he never considers language as anything else than a practical way to express our thought.5 As a consequence, although it is uncontroversial that Brentano pays deep attention to ordinary language, holding that there is room for ‘philosophy of language’ in his views is more problematic.6 For my part, whether he is a philosopher of language or not, I will claim that it is worth considering his dense linguistic analyses and his conceptual innovations on language. In that respect, I will now pay

62 

C. Gauvry

specific attention to one of his major contributions: the distinction between ‘syncategorematic’ and ‘categorematic expressions’.

2

Syncategorematic Expressions

Even if they are often draft in a cursory manner, Brentano’s manuscripts on logic gather very interesting insights such as the following one: 20. Not all of what is said is in itself a speech. There are rather three kinds of things said [Gesprochene]: 1. Some are meaningless when considered on their own [für sich allein bedeutungslos sein] and able to indicate something only through their connexion with other terms [mit anderen verbunden], e.g. particles, declensions of substantives etc. Objection. Solution. The suppositio materialis is not about a sign but about a designated term: ‘but’ is a conjunction just as the one there (standing for the beef ) is fat. Or a name is for a particle rather than the particle itself (syncategorematic expressions). 2. What is said can already be fully meaningful, but only through its naming [benennend] something without being a complete proper expression [ein eigentlicher fertiger Ausspruch] (a speech [eine Rede]), as, e.g. any name that we utter (categorematic expressions, also names made up several-words). 3. It can be meaningful and a complete expression (a speech). E.g. statement, exclamation, request. (Ms. EL 80, 13.0001)

There are consequently three categories of expressions: 1. The first class includes ‘syncategorematic expressions’, i.e. various terms as syllables, particles, adverbs, etc. All those terms present the similarity that they are not meaningful, and cannot name something on their own but only in connection with other terms. At first sight, the naming and meaning functions7 of those expressions seem to be, if not ‘contextually’, ‘syntactically’ dependant. Answering the ­(controversial) case of ‘suppositio materialis’, which could be presented as an ‘objection’, Brentano specifies that even if some adverbs or par-

3  A Context Principle in Brentano? 

63

ticles seem to be ‘nouns’ in the grammatical sense—for instance when one says: ‘“But” is a conjunction’—they have to be considered as ‘syncategorematic’. 2. The second class gathers ‘categorematic expressions’, i.e. ‘names’ or ‘complexes of names’ such as: ‘house’, ‘horse’, ‘cow of the farmer’, etc. Those expressions, which are also called ‘autosemantic terms’, have a naming function on their own. However, they are not ‘complete proper speeches’ and it is unlikely that they have a ‘meaning’ function on their own. 3. The third class regards the ‘speech’ itself (die Rede). Only this third class of expressions is considered a complete expression (fertiger Ausspruch) that is plainly meaningful. It includes statements, exclamations, requests, etc. According to this classification, in particular with respect to the first and the third classes, Brentano seems to endorse a ‘context’ or at least a ‘syntactical’ principle. Syncategorematic and categorematic expressions have a proper meaning only when considered in a complete utterance. Based on similar textual evidences, some scholars (see e.g. Kraus 1934, 39; Mayer-Hillebrand 1977, xxiii; Fréchette, in this volume) hold that Brentano was a supporter of the so-called ‘context principle’ which takes, to use a preliminary and broad formulation, that a word has a meaning only in the context of a sentence. Consequently, Brentano would advocate a principle similar to the influential principal addressed by Frege in the introduction of The Foundations of Arithmetic (Frege 1980, x). Although interesting, I think that this comparison is dubious, for at least two reasons. First, there are strong requirements for a view to be called ‘contextualist’, at least in the Frege-Wittgenstein sense. Second, Brentano’s position is too complex to be settled by any single label. To clarify these doubts, I will focus more specifically on the classification just outlined. I will start with ‘syncategorematic expressions’. Which terms do they exactly include? It could be tempting to define them as follows: syncategorematic expressions gather every term that is not a ‘noun’ in the grammatical sense. However, when considering more closely Brentano’s text, it seems that the grammatical criterion is not telling enough. Brentano indeed first mentions that some complexes of nouns can

64 

C. Gauvry

also be considered syncategorematic: ‘So not only syllables but also words, and even whole complexes of words […] e.g. particles such as ‘of ’, ‘to’, ‘truly’, ‘not’, ‘only’. Also the words ‘none’ (= not one), ‘some’. Article. Cases: ‘me’, ‘him’, ‘of the bird’ [des Vogels], ‘of the house’, etc.’ (Ms. EL 80, 13.008[7–8], my emphasis) have to be counted as ‘syncategorematic expressions’. Consequently: ‘of the bird’ or ‘of the house’ are considered syncategorematic. According to this first quotation, it is true, however, that the nouns are not syncategorematic on their own (bird, house) but only as part of a complex (of the bird, of the house). More radically, the late Brentano takes non-referential nouns, also called ‘linguistic fictions’, to be syncategorematic expressions (2009, 43, 49; see also 1977a, 101). For instance, he mentions: ‘redness’, ‘Centaur’, ‘truth’, ‘goodness’, etc. (see e.g. Brentano 1911; see also 1977b, 1995b). As a result, although some substantives can be grammatically considered as ‘nouns’ (‘truth’, ‘goodness’, etc.), they should not be considered ‘names in the logical sense’ (see Brentano 2009, 49) but syncategorematic expressions in such cases. The case of ‘suppositio materialis’ also manifests that the grammatical criterion is not decisive for Brentano (and can even be misleading): even if a ‘conjunction’ seems to be grammatically used as a ‘noun’, it nonetheless remains a conjunction. Consequently, the grammatical criterion is not sufficient to discriminate syncategorematic expressions. Another solution would be to use a logical criterion, such as the one used by Twardowski (1894, 22) to prove that ‘nothing’ is a syncategorematic expression. According to him, a term of the form ‘non-x’ is categorematic only if it can be considered as a division of a higher order concept. For instance ‘non-human’ is categorematic because it is a sub-­ division of the concept ‘animal’ (or ‘being’) contrary to ‘nothing (= non-­ something)’, which is not a division of any other concept. But here again the criterion fails to successfully show that ‘of the bird’, ‘of the house’ or ‘goodness’ have to be considered syncategorematic. Hence, the use of this Twardowskian criterion is only relevant for a particular case of categorematic terms: terms of the ‘non-x’ form. As a consequence, Brentano introduces his own criterion (which Twardowski will also endorse) that is neither grammatical nor logical but psychological:

3  A Context Principle in Brentano? 

65

We said that [categorematic expressions] differ from syncategorematic expressions since they mean something on their own [für sich allein]. We said that they differ since [categorematic expressions] are on their own the closed expression of a mental phenomenon. (Ms. EL 80, 13.013)

Only the following ‘psychic’ criterion is hence determining: whether an expression is meaningful enough in order to fully express a presentation. Since ‘and’, ‘to’, ‘of the bird’, ‘of the house’ or ‘goodness’ are not expressing any meaningful presentation, they have to be considered syncategorematic. As a matter of fact, it is misleading to hold that the relevant criterion to determine that an expression is ‘syncategorematic’ is that it is meaningful only when ‘connected to other words’. The grammatical indication is informative since most terms that are syntactically dependant are indeed meaningless on their own, but it is neither a sufficient nor a necessary condition. It can even be delusive. By contrast, the critical criterion is what I call the ‘mental expressivity condition’; whether an expression can announce or express (ausdrucken) a ‘meaningful presentation’. As a preliminary conclusion, it makes sense to consider the meaning function of ‘syncategorematic expressions’ as depending on the syntax of the sentence. Such expressions do not mean anything on their own but only in connection to other terms: they are ‘mitbedeutend’. However, the determining meaning principle is not purely ‘grammatical’, or even ‘linguistic’. The condition to be meaningful; for an expression, is solely to be expressive, i.e. to express a mental presentation (and potentially to influence someone else’s mental life).

3

Categorematic Expressions

As mentioned in my first quotation of the Logic manuscript, ‘categorematic expressions’, also called ‘autosemantic expressions’, are mostly common nouns. However, as previously explained, Brentano is far from blindly trusting grammatical distinctions. He actually introduces a distinction between ‘names’ and ‘names in the logical sense’ (Brentano 2009,

66 

C. Gauvry

49). Clarifying this distinction is instrumental to characterise what a ‘categorematic expression’ is. According to Mill’s 1843 Science of Logic, which exercised a deep influence on Brentano, a name is a linguistic expression (a word or complex of words) that functions either as the subject of a proposition, or as the predicate of an expression, literally as a ‘categorematic’ term. Mill indeed derives the English word ‘categorematic’ from the Greek verb ‘kategoreo’, which means ‘to predicate’ (Mill 1974, 25). Very similar characterisations can be found in Brentano, for instance in the controversial (Brentano 1956, 37). Consequently, the class ‘names’ mostly gathers substantives, pronouns and substantive-phrases. However, here again, the grammatical and logical criteria are not relevant. First, Mill considers that some adjectives can be counted as ‘names’ when they are used in a ‘grammatical ellipsis’. To take Mill’s example, when the adjective ‘white’ is an abbreviation of a ‘white object’, for instance in the sentence: ‘Snow is white’, then it can be considered as a predicate, that is as a ‘categorematic’ term, hence as a ‘name’ (Mill 1974, 24). I am not sure how Brentano would address this specific case. Nevertheless I think that he wouldn’t trust any grammatical analyses to characterise a categorematic expression. Just as for ‘syncategorematic expressions’, the main criteria to determinate what a ‘categorematic expression’ is, is psychological. I quote again (Brentano Ms. EL 80, 13.013): ‘[Categorematic expressions] are on their own the closed expression of a mental phenomenon’. The quotation indeed attests that the distinction between categorematic and syncategorematic expressions is psychological. However, it appears that the first part of the same quotation, i.e. ‘categorematic expressions […] mean something on their own’ is ambiguous because it can support several, and even opposite, interpretations. At first glance, one can indeed use this ‘textual evidence’ in order to support a non-contextualist reading of ‘categorematic expressions’. More precisely, based on a (superficial) reading of the quotation, one could hold that, for Brentano, the meaning function of a categorematic expression does not depend on any syntactical consideration, which would contravene any context principle.

3  A Context Principle in Brentano? 

67

In a way, this is precisely the objection that Gilbert Ryle raised against Mill’s view and which could, based on this interpretation, also be addressed to Brentano. In a 1957 paper, Ryle indeed devotes great attention to Mill’s position—mentioning in passing his influence on Brentano and his students. After having recognised the virtues of Mill’s remarks on names, proper names, denotations and connotations, Ryle criticises a number of ambiguities of Mill’s theory. More precisely, he suggests that under his influence, most of his successors (starting with Russell) endorsed a somehow atomistic and decontextualised conception of meaning. His two main reproaches are the following: 1) ‘Most words are not nouns’ (Ryle 1957, 245) and 2) ‘The meanings of significations of many kinds of expressions are matters not of naming things, but of saying things’ (Ryle 1957, 253). Here is the complete formulation: The meaning of a sub-expression like a word or a phrase, is a functional factor of a range of possible assertions, questions, commands and the rest. It is tributary to sayings […]. This precisely inverts the natural assumption […] that the meanings of words and phrases can be learned, discussed and classified before consideration begins of entire sayings, such as sentences. Word-meanings do not stand to sentence-meanings as atoms to molecules or as letters of the alphabet to the spelling of words, but more nearly as the tennis-racket stands to the strokes which are or may be made with it. (Ryle 1957, 249)

In a word, Ryle argues that Mill (and his successors) gives too much autonomy to the name; especially to its ‘naming’ function. This charge clearly comes from Wittgenstein (e.g. 1953, § 383). In other words, Mill’s main error is nothing less than to have missed the ‘context principle’. According to the first interpretation of the quotation above, Brentano also seems guilty. If categorematic expressions have a naming function and a meaning function ‘on their own’, without being ‘tributary to [the entire sayings]’ (Ryle 1957, 249), those functions have to be considered a-contextual. However, I consider that this reading of the quotation is misleading, which will become obvious if we consider the quotation in its proper background. To do so, one ought to pay a global attention to the logic manuscript and to the fact that, to quote (Ms. EL 80, 13.001[2]) again:

68 

C. Gauvry

2. What is said can already be fully meaningful, but only through its naming [benennend] something without being a complete proper expression [ein eigentlicher fertiger Ausspruch] (a speech [eine Rede]), as, e.g. any name that we utter (categorematic expressions, also names made up of several-words).

According to this longer quotation, to say that a categorematic expression is ‘fully meaningful’ on its own only means that it can name something without being coordinated to other terms. In other words, only its ‘naming function’ is a-contextual. Conversely, it is not explicitly claimed that the meaning function of categorematic expressions is not ‘tributary’ to syntactical considerations. The question whether meaning functions are also dependent to syntactical considerations is critical since the naming function is somehow dependent to the meaning function: Brentano indeed states that ‘one names by means of the meaning!’ (Ms. EL 80, 13.018[5]) That is consequently the question that I will address in the next section.

4

A Contextualist?

As I mentioned previously, Brentano values ordinary language. As a natural consequence, he clearly considers that all linguistic expressions have to be analysed in the ordinary contexts of their utterance. Hence, as far as I understand the Logic manuscript, only the naming functions of categorematic expressions are autonomous. By contrast, a name is meaningful when used in ‘a complete expression (a speech), e.g. statement, exclamation, request’ (Brentano Ms. EL 80, 13.001[2]). That is the reason why, anticipating Ryle’s objection, Brentano does not restrict its analysis to the nouns, focusing instead on ‘what is said’ (das Gesprochene). Some other textual evidences are even clearer in that respect, for instance: ‘Nobody speaks in uttering names, we all speak in uttering sentences [Sätzen], be they statements [Aussagen] or exclamations, questions, orders, requests, etc.’ (Brentano 1977a, 378). Moreover, Brentano even introduces a similar distinction in his correspondence with Marty: between theoretical and practical uses of common names (see e.g. Marty 1873). According to this

3  A Context Principle in Brentano? 

69

additional specification, even if a name is meaningful on its own ‘in theory’, its meaning function is fully determined only when the name is ‘practically’ used in a concrete speech situation. For all those reasons, I am personally convinced that Brentano already considers that meaning is ‘tributary to the entire saying’ (see e.g. Kraus 1934, 33–34; or Mulligan 1990, 17–18). Brentano indeed paid attention to the practical dimension of speech and to some syntactical principles. In this respect his manuscripts obviously anticipate some of the main intuitions of the later philosopher of ordinary language and deserve genuine interest. However, whether Brentano can be considered a contextualist is polemical for reasons that I will now present. First, what is at stake with the notion of ‘context principle’, at least according to its first historical introduction, is the precedence of the whole proposition over its constituents. According to Frege’s paradigmatic formulation, context indeed means the context of the ‘proposition’; der Satzszusammenhang. In direct continuation, Wittgenstein also famously asserts in his Tractatus Logico-Philosophicus: ‘Only in the context of the proposition [im Zusammenhang des Satzes] has a name meaning [Bedeutung]’ (Wittgenstein 1922, 3.3). In those inaugural and very similar formulations, ‘context’ consequently means ‘der Zusammenhang des Satzes’ and the principle can be rephrased this way: a word has a meaning (Bedeutung) only in the context of the proposition. However, is there something as a ‘propositional context’ principle in Brentano? There are certainly some references to the ‘complex’ of the sentence understood as a whole in Brentano. As just mentioned, he indeed speaks of a ‘complete speech’ (Ms. EL 80, 13.001[2]), of words connections (Ms. EL 80, 13.009, 13.011). He even mentions in the controversial Brentano 1956 the ‘Redeszusammenhang’, i.e. the ‘context of the speech’ (Brentano 1956, 36) and ‘ein sinnvolles Redeganze’, i.e. ‘a meaningful whole speech’ (Brentano 1956, 44). However, the exact nature of this ‘Redeganze’ is not explicitly clarified. As said previously, some (in particular Franziska Mayer-Hillebrand 1977, xxiii) do not hesitate to consider that this ‘Redeganze’ refers to the ‘Zusammenhang des Satzes’. I would be reluctant to endorse this reading for at least two reasons. The first one is terminological. To my knowledge, even if Brentano sometimes speaks of ‘Sätze’, he never uses himself the locution: ‘Zusammenhang des Satzes’.

70 

C. Gauvry

Besides, even if there were some unknown occurrences in his manuscripts, it would not mean much since the German notion of ‘Satz’ is particularly ambiguous as it both means ‘proposition’ and ‘sentence’. Second, some of Brentano’s main theses lead me to think that his conception of a ‘Satz’ is not propositional. As a main illustration, contrary to Frege, he does not endorse the semantic view according to which the standard form of the meaning is a proposition which has a truth-value. This is obvious given his use of the notion of ‘content’, which is clearly not semantic but rather psychological.8 Consequently, although Brentano holds that the meaning of an expression is dependant on the ‘whole sentence’ in which it is uttered, I consider that this claim does not in itself allow to conclude that Brentano endorses a ‘context’ principle in the Frege-early Wittgenstein sense. Yet, it is true that the notion of ‘context’ does not always mean ‘propositional context’, even in the ‘contextualist’ tradition. If someone speaks, following the so-called philosophers of ordinary language, of ‘context principle’, she would indeed rather refer to the external ‘circumstances’ in which the speech is uttered. This conceptual and terminological variation is obvious in Wittgenstein’s own work. After the late 1920s, he indeed mostly speaks of the ‘besondere Umstände’ or of the ‘Umgebung’ of the speech and he does not take the proposition to be the context. I just quoted two occurrences from the Philosophical Investigations but I could have mentioned many others: § 583. What is happening now has significance—in these surroundings [Umgebung]. The surroundings [Umgebung] give it is importance. (Wittgenstein 1953, § 583 129–130) § 607. These were not the circumstances [Umstände] of my saying the words.—But then, what were the circumstances [Umstände]?—I was thinking about my breakfast and… (Wittgenstein 1953, § 607, 134)

The ‘circumstances’ which are here mentioned are the social circumstances in which the discourse is uttered. Moreover, as made very clear in § 583: ‘The surroundings [Umgebung] give [the meaning] its importance’. As a consequence, ‘context principle’ also means a linguistic principle according to which the meaning of the words is determined in the social

3  A Context Principle in Brentano? 

71

and conventional circumstances of their utterance. Therefore, those circumstances have to be considered as normative.9 It is a substantial aspect of their definition. Once again, is such a principle at work in Brentano? Although it may be true of some of his direct or indirect inheritors (e.g. Ludwig Landgrebe or Karl Bühler),10 there are no clues showing that Brentano was really interested in the social context and especially in the conventional dimension of such social circumstances. On the contrary, my leading hypothesis is that the determining ‘Zusammenhang’ (or ‘Ganze’) of the entire saying is understood in an internal way by Brentano. To put it in other terms, the so-called ‘context’ in which any expression has to be placed in order to be meaningful is nothing more than the expressive sentence whose function is to express a mental act. That is the reason why the content of this meaningful sentence (which has not necessarily a propositional form and which can instead adopt the form of an ‘exclamation’ or a ‘request’) is nothing else than the mental content of the act expressed by the sentence. What a meaning utterance means is precisely this mental content, which is sometimes called the ‘immanent object’ of the psychic act (Ms. EL 80, 13.018[1]). The interesting case of linguistic fictions makes this very last consequence even more obvious. As already explained in (Brentano 1995a), one of the main purposes of the philosophy, understood as a ‘critique of language’, is to rephrase the terms of our ordinary language so that they refer to real objects. As a result, one of the main challenges of Brentano’s critique of language is to ensure that all expression refers, at least in modo recto, to a psychic content (that is real). As a consequence, as a solution to the problems raised by Universals or by fictions, Brentano holds that terms like ‘virtue’, ‘Pegasus’, or ‘golden mountains’ refer in modo recto to the mental content of the associated psychic acts: ‘I believe that there are virtuous men’, ‘I imagine that Pegasus is a winged horse’, ‘I think that there are no golden mountains, etc.’ In this respect, the ‘syntactical complex’ in which those different expressions have to be placed in order to be meaningful is the periphrasis which aims to express the mental content of the related psychic acts. Personally, I have strong doubts that this ‘periphrasis’ can be correctly described in terms of ‘context’, in the sense used by the first contextualists or even by philosophers of ordinary language.

72 

5

C. Gauvry

Conclusion

On the one hand, it is true that Brentano pays substantial attention to the practical dimension of our different usages of the linguistic terms and even holds that an auto-semantic name has a meaning function only when used in a practical utterance. For that reason, one is entitled to underline the practical dimension of Brentano’s innovative linguistic analyses and to consider him a forerunner of philosopher of ordinary language. On the other hand, to my knowledge, Brentano is neither a propositionalist nor interested in the social and conventional circumstances of our utterances. Therefore, I do not consider that a ‘context principle’, in the strict sense of the term, is at work in his analyses. According to my last hypothesis, there is a determining complex of our linguistic expressions, however. This complex has to be considered mental. It is the complex of the periphrasis that expresses the mental content of the act to which the meaningful expressions necessary refer. Therefore, I would rather speak of a ‘mental complex principle’ or of an ‘expressivity principle’ rather than of a ‘contextual principle’ in Brentano. This last conclusion confirms the inaugural paradox. Although Brentano is more sensitive to the practical dimension and to the richness of ordinary language than the first analytic philosophers, he cannot be considered a ‘philosopher of ordinary language’ and even less a ‘contextualist’.

Notes 1. Although interesting, this view is of course highly controversial. Questioning it is precisely one of the challenges of the current volume. See also Mulligan fruitful attempts to underline the proximity between early semantics and Brentanian theories of meaning (e.g. Mulligan 1990, 2012, 2019). 2. This terminological distinction is of course not Brentano’s own insight. It was first introduced by Aristotle (De interpretatione, 16 a 19–22) and discussed throughout scholasticism.

3  A Context Principle in Brentano? 

73

3. On the definition of ‘functionalist’ in Brentano and above all in Marty, see (Cesalli 2015, 255–257). 4. Although the reference has to be read cautiously for the reasons previously explained, (Brentano 1956, 312, note 29) provides an interesting quotation. On Marty’s double theory on the intention of the deliberate expression (sign as ‘announcement’ or as an ‘influence’ on someone’s else mental life), see (Marty 1908, 22, 280–287, 490–501). 5. As a consequence, he also distinguishes himself from late Scholastic or more contemporary views that consider the thought (in particular the conceptual abstraction) as a product of language. See e.g. (Bonald 1800, 47–48) and (Leblanc, in this volume). 6. However, as suggested in our introduction, his position presents some clear affinities with the last generation of analytical philosophers of language who also consider that linguistic, psychological and metaphysical questions have necessarily to be connected. 7. To my knowledge, Brentano’s distinction between the ‘meaning’ and ‘naming’ functions of an expression is not clearly fixed. Nevertheless, Brentano introduces some interesting clarification in his (Ms. EL 80 Manuscript): a name ‘means’ an internal object (or a mental content) ‘which is presented through the presentation’ whereas it ‘names’ ‘the external object of the presentation’ (see Ms. EL 80, 13.018). More precisely, if one can trust the controversial (Brentano 1956), adopting the classification also used in (Twardowski 1894), Brentano holds that a ‘name’ is characterised by three functions: it announces an act of presentation; it means a conceptual content and names an object (see e.g. Brentano 1956, 45; see also Fisette and Fréchette 2007, Introduction, 158 ff.; and Taieb in press). 8. See Twardowski’s 1894 distinction, which clarifies Brentano’s use of the notion of ‘content’. For a discussion, see also (Leclercq, in this volume). 9. See also (Austin 1962, 110): ‘The question of truth and falsehood does not turn only on what a sentence is, nor yet on what it means, but on, speaking very broadly, the circumstances in which it is uttered’. 10. See for instance (Marthelot 2012; Cesalli and Friedrich 2014; Gauvry in press; Mulligan, in this volume).

74 

C. Gauvry

References Austin, John Langshaw. 1962. Sense and Sensibilia. Oxford: Oxford University Press. Benoist, Jocelyn. 2003. Sprachkritik ou sémantique. Sur le schisme de l’école brentanienne. Les études philosophiques 1: 35–52. Brentano, Franz. 1911. Von der Klassifikation der psychischen Phänomene. Leipzig: Duncker and Humbol. ———. 1956. In Die Lehre vom richtigen Urteil, ed. Franziska Mayer-Hillebrand. Bern: Francke. ———. 1977a. In Die Abkehr vom Nichtrealen, ed. Franziska Mayer-Hillebrand. Hamburg: Felix Meiner Verlag. ———. 1977b. Zur Lehre von den entia rationis. In Die Abkehr vom Nichrealen, ed. Franziska Mayer-Hillebrand, 390–394. Hamburg: Felix Meiner Verlag. ———. 1995a. Psychology from an Empirical Standpoint. Trans. Aantos C.  Rancurello, D.B.  Terrell, Linda L.  McAlister. London and New  York: Routledge. ———. 1995b. On ens Rationis. Trans. Antos C. Rancurello, D.B. Terrell and Linda L.  McAlister. In Psychology from an Empirical Standpoint, 265–287. London: Routledge. ———. 2009. The True and the Evident. Trans. Roderick M.  Chisholm, Ilse Politzer and Kart Fischer. Taylor and Francis e-library. ———. n.d.-a. Ms. EL 72. Logik. Cambridge, MA: Houghton Library. ———. n.d.-b. Ms. EL 80. Logik. Ms. edited by Robin D. Rollinger. Cambridge, MA: Houghton Library. Text available online: http://gams.uni-graz.at/ archive/objects/o:bag.el.80-html-norm/methods/sdef:HTML/get. Cesalli, Laurent. 2015. Meaning in Action: Anton Marty’s Pragmatic Semantics. In Linguistic Content. New Essays on the History of Philosophy of Language, ed. Margareth Cameron and Robert J.  Stainton, 245–265. Oxford: Oxford University Press. Cesalli, Laurent, and Janette Friedrich. 2014. Anton Marty & Karl Bühler— Between Mind and Language. Basel: Schwabe. Coffa, J. Alberto. 1991. The Semantic Tradition from Kant to Carnap. Cambridge: Cambridge University Press. Bonald, Louis de. 1800. Essai analytique sur les lois naturelles de l’ordre social. Paris. Dummett, Michael. 1973. Frege. Philosophy of Language. London: Duckworth.

3  A Context Principle in Brentano? 

75

Fisette, Denis, and Guillaume Fréchette, eds. 2007. À l’école de Brentano. De Würzbourg à Vienne. Paris: Vrin. Frege, Gottlob. 1980. The Foundations of Arithmetic. Trans. John Langshaw Austin. Oxford: Blackwell. Gauvry, Charlotte. in press. Landgrebe’s Reading of Marty: on Name and Proper Name. Analecta Hermeneutica. Kraus, Oskar. 1934. Wege und Abwege der Philosophie. Prague: Robert Lerche. Marthelot, Perrine. 2012. Karl Bühler. Du contexte à la situation, la signification. Paris: Armand Colin. Marty, Anton. 1873. Letter from Marty to Brentano. May 23th 1873. Ms. Cambridge, MA: Houghton Library. ———. 1908. Untersuchungen zur allgemeinen Grammatik und Sprachphilosophie. Halle: Max Niemeyer. Mayer-Hillebrand, Franziska. 1977. Einleitung. In Brentano, Franz, 1977. Die Abkehr vom Nichtrealen, ed. Franziska Mayer-Hillebrand, 1–99. Hamburg: Meiner. Mill, John Stuart. 1974. A System of Logic. In Collected Works of John Stuart Mill, ed. John M. Robson, vol. 7–8. Toronto: University of Toronto Press. Mulligan, Kevin. 1990. Marty’s Philosophical Grammar. In Mind, Meaning and Metaphysics. The Philosophy and Theory of Language of Anton Marty, ed. Kevin Mulligan, 11–27. Dordrecht: Kluwer. ———. 2012. Wittgenstein et la philosophie austro-allemande. Paris: Vrin. ———. 2019. Misleading Pictures, Temptations and Meta-Philosophies: Marty and Wittgenstein. In Anton Marty and Contemporary Philosophy, ed. Giuliano Bacigalupo and Hélène Leblanc, 197–232. Cham: Palgrave Macmillan. Ryle, Gilbert. 1957. The Theory of Meaning. In British Philosophy in the Mid-­ Century, ed. J.H. Muirhead, 239–264. London: George Allen and Unwin. Simons, Peter. 1992. Philosophy and Logic in Central Europe from Bolzano to Tarski. Dordrecht: Kluwer. Sluga, Hans. 1980. Gottlob Frege. London: Routledge and Kegan Paul. Taieb, Hamid. in press. Ordinary Language Semantics: The Contribution of Brentano and Marty. British Journal for the History of Philosophy. Twardowski, Kazimierz. 1894. Zur Lehre vom Inhalt und Gegenstand der Vorstellungen. Munich: Philosophia Verlag. Wittgenstein, Ludwig. 1922. Tractatus Logico-Philosophicus. Trans. C.K. Ogden. London: Routledge and Kegan Paul. ———. 1953. Philosophische Untersuchungen/Philosophical Investigations. Trans. Elizabeth Anscombe. Oxford: Blackwell.

4 Brentano and Mauthner on Grammatical Illusions Denis Seron

In the current literature Brentano’s name is usually associated with British empiricism, the Aristotelian tradition, and the so-called ‘Austrian semantic turn’. This paper seeks to suggest a convergence with another tradition within Austrian philosophy—namely the critique of language developed, among others, by Mach, Mauthner, Karl Kraus, the German Vaihinger, and Wittgenstein. My starting hypothesis is that, despite significant differences, the late Brentano’s approach to grammatical illusions has a great deal in common with Fritz Mauthner’s critique of language. One of the first authors who noticed the similarities between the two philosophers was Oskar Kraus. In his Self-Presentation from 1929, Kraus says that the late Brentano’s search for a criterion for autosemantic expressions clearly belongs to the same movement as the critique of language of Vaihinger and Mauthner (Kraus 1929, 16, cited in Weiler 1970, 297). The connections between Brentano and Mauthner have been investigated at length later, above all by Liliana Albertazzi (1986, 1989, 1990) and, to

D. Seron (*) Department of Philosophy, University of Liège, Liège, Belgium e-mail: [email protected] © The Author(s) 2021 A. Dewalque et al. (eds.), Philosophy of Language in the Brentano School, History of Analytic Philosophy, https://doi.org/10.1007/978-3-030-52211-7_4

77

78 

D. Seron

a lesser extent, by Gershon Weiler (1970). Some other commentators have drawn attention to them without further discussion. For example, Cesalli and Mulligan rightly claim that Brentano’s critique of language was an ‘opening shot of the Austrian critique of language to which Marty, Mauthner, Oskar and Karl Kraus and Wittgenstein will all contribute’ (Cesalli and Mulligan 2017, 257). This paper is an attempt to bring forward some new or newly formulated ideas on the relation between Brentano and Mauthner.

1

Critique of Language

Fritz Mauthner terms his philosophical project as a whole ‘critique of language’ (Sprachkritik). In his view, the philosopher’s sole task is to liberate us from the illusions of language. The basic idea is twofold. First, language necessarily generates grammatical illusions that are pervasive and play a central role in science, philosophy, religion, politics, and everyday life. Secondly, Mauthner defends the extreme sceptical view to the effect that these illusions make language in general ‘inadequate for knowing the world and hence for communicating knowledge’ (Mauthner 1901–1902, I, 122). We cannot enrich our knowledge of the world through language; all language consists of only tautologies (Mauthner 1901–1902, I, 93; Weiler 1958, 82; Weiler 1970, 236 ff.). Brentano, too, holds that grammar spontaneously generates illusions, and that an essential part of the philosopher’s task is to dissolve these illusions through philosophical clarification. In his late correspondence with Marty, he calls these grammatical illusions ‘entia rationis’ or ‘fictions’, as opposed to entia realia or things (Brentano 1977, 324). Grammar leads us to introduce illegitimate entities—fictions—which as such are no more than entia linguae, or entia elocutionis in Marty’s terminology (Brentano 1977, 191). It is unclear whether Brentano defended this view only in his reist period or also in earlier writings. Some commentators, for example Massimo Libardi (1996, 68), claim that the struggle against entia rationis thus conceived is basic not only to Brentano’s late reism, but to his work as a whole.

4  Brentano and Mauthner on Grammatical Illusions 

79

Brentano’s line of reasoning can be summed up as follows: first, the theory of intentionality teaches us that not all phenomena really exist. Some thought objects are real things and some are fictions, that is, possible correlates of synsemantic expressions. Second, grammar is deceiving because synsemantic expressions are grammatically indistinguishable from autosemantic expressions. A fictitious name can have the same grammatical function within a sentence as the name of a real thing. For example you can say ‘A round square is impossible’ or ‘Freedom is better than slavery’. This may lure you into the illusion that round squares and freedom are real things. Since the term ‘round square’ is the grammatical subject of the sentence ‘a round square is impossible’, it looks as if the round square were a thing that has the property of being impossible. However, a round square is not a real thing, but a fiction. We mistake it for a real thing because the grammatical function of the term ‘round square’ in the sentence prompts us to do so. Actually, Brentano argues, there is ‘no more than a specific linguistic phrasing that leads one to the fiction of a new being’ (Brentano 1977, 172). Thirdly and finally, we need a criterion for distinguishing entia rationis from entia realia. Since synsemantic expressions are grammatically indistinguishable from autosemantic expressions, this criterion must come from elsewhere than from grammar itself. In Brentano’s view, it is psychological analysis that enables us to sort out synsemantic from autosemantic expressions and thus to discriminate fictions from real things (Brentano 1977, 173). The non-grammatical criterion the late Brentano proposes for distinguishing entia rationis from entia realia is, at least in part, psychological. It is stated as follows: ‘Every object of a presentation, if it exists, is real’ (Brentano 1977, 380). This criterion sets two conditions for something to be an ens reale. First, there must be a presentation (or intuition, as Brentano also says): ‘It is impossible, claims Brentano, to have a presentation without having a presentation of something. And this “something” means something real’ (Brentano 1977, 173). Being real is a necessary condition for being presented. In other words, if an intentional object is presented, then it must be real; if it is an ens rationis or fiction, then it cannot be presented and one or more acts of the two other classes can be identified through psychological analysis. Thus, wherever psychological analysis uncovers a genuine presentation, that is, an act that is not

80 

D. Seron

reducible to a judgement or feeling, the object presented must be real. This first requirement leads us to deny that abstract terms like ‘colour’ and ‘space’ as well as dispositional, negative, and most mathematical expressions denote entia realia.1 The second condition is that the ens reale must actually exist. It is not easy to work out exactly what this further requirement adds to the first one. On my interpretation, Brentano’s aim might be to confine entia realia to just the things given in present inner perception. It is well known that the late Brentano defended a form of presentism according to which, as he declares in his posthumous Versuch über die Erkenntnis, ‘“being” means “present”’ (‘Seiend’ heisst soviel als ‘gegenwärtig’) (Brentano 1970, 27). Since only what is perceived, that is, innerly perceived, is present, it follows from this that only mental phenomena can be present and hence real, while other phenomena must be fictions. In any case, there are obvious similarities between these views and Mauthner’s critique of language. Both Brentano and Mauthner hold that we all cling to fictions as if they were real objects, that the reason for this is grammatical, and that part or whole of the philosopher’s job is to denounce such grammatical illusions. This corresponds to what Mauthner calls ‘word fetishism’. Some expressions are misleadingly used in such a way that their surface grammar function or category induces us to take them as denoting real things, while they actually don’t: ‘Most people have the weakness to believe that, since a word is present, it must be a word for something, and that, since a word is present, something real must correspond to it’ (Mauthner 1901–1902, I, 159). This means, in Brentano’s terms, that some expressions are used as if they corresponded to presentations and thus to real things, while they actually don’t.

2

L anguage Doesn’t Refer to the Outside World

I come now to Mauthner’s scepticism. His most central view about language is that language does not refer to the world. Mauthner denies that language in general has a referential function at all. In his view, language

4  Brentano and Mauthner on Grammatical Illusions 

81

has instead mainly two functions, a practical and an expressive one. First, words serve to register memories within a linguistic community. The memories of a people are stored in its language just as the memories of an individual are stored in her brain (Mauthner 1901–1902, I, 179, 184–185; Mauthner 1906, 26–27). Secondly, words have a poetical or expressive function which is carried out by a small number of original thinkers. Such-and-such an experience A puts the original thinker in a specific mood (Stimmung), that is, brain state. The original thinker then associates this brain state with such-and-such a word or combination of words which is commonly associated with another experience B, so that the linguistic community can then use this new association in order to register experiences: the word used by the poet can now be used to register both experiences A and B, and then the experience A only. In Mauthner’s view, the expressive function is prior to the mnemonic function. The vocabulary of a language is constituted by poetical expressions that have lost their expressive value. To sum up, we associate a given word with some specific memories because our linguistic community prescribes to do so, and our linguistic community prescribes to do so because in the past a poet associated this word with a specific mood or brain state which had until then been associated with other words and memories. Words are primarily signs for moods, and it is only secondarily that they are taken to denote objects, namely those objects which cause the corresponding mood in the poet’s mind. For this reason, Mauthner claims, ‘words never produce knowledge, they are just tools for poetry’ (Mauthner 1901–1902, I, 151). The key point here is that language, for Mauthner, neither refers nor enables us to know reality. The ultimate function of language is solely expressive or aesthetic: language expresses moods or brain states that are associated with our presentations of the real world—and this association should not be conceived of as a relation of reference. There exist no expressions that denote real objects. Hence, language and knowledge are mutually incompatible and knowledge is impossible. In Mauthner’s view, words and concepts have no direct connection with presentations; they just express aesthetic moods that are incidentally associated with presentations. Of course, Brentano was not a sceptic. Unlike Mauthner, he did believe that some linguistic expressions really

82 

D. Seron

denoted entia realia, namely present mental phenomena. While Brentano claims that ‘all our concepts derive from intuitions’ (Brentano 1977, 329), Mauthner protests that ‘a concept amounts not to a real intuition, but only to the beautiful appearance of an intuition’ (Mauthner 1901–1902, I, 123–124). Nevertheless, Mauthner’s views may be regarded as close to Brentano’s in some respects. Both philosophers deny that language refers to external reality. In the Logikkolleg, Brentano characterises language as a ‘sign of thinking’ (Zeichen des Denkens) (Ms. EL 80, 12.978[9]; see Cesalli and Mulligan 2017, 257). This characterisation should be taken in a radical sense. In his monography on Brentano posthumously published by Franziska Mayer-Hillebrand, Alfred Kastil paraphrases Brentano’s claim as follows: ‘language is not a direct sign for things existing outside the speaker, but a sign for what goes on in her, in her soul’ (Kastil 1951, 100). Both Brentano and Mauthner deny that our language denotes external reality; both claim that language ultimately has to do with inner life, namely moods or present mental acts. The difference is that, for Brentano, language in some cases really denote internal things, while for Mauthner language never denotes anything. I’d like now to review three kinds of entities that both Brentano and Mauthner reject as linguistic fictions, namely intentional objects, the ego, and general objects.

3

Three Kinds of ‘Word Fetishes’

Some commentators view (rightly in my opinion) Brentano’s rejection of entia rationis as a consequence of his theory of intentionality (Huemer 2018). What is more, it is plausible to say that the approach just outlined is already at the basis of Brentano’s theory of intentionality in the Psychology of 1874. The basic tenet of the theory of intentionality is that present mental phenomena really exist and that they necessarily contain in them non-existent phenomena, namely physical or mental phenomena that are not presently experienced by the subject. Brentano’s theory of intentionality stems from his rejection of the view that all phenomena are objects of perception, that is, given as really existent in the present.

4  Brentano and Mauthner on Grammatical Illusions 

83

What Brentano advocates instead is his theory of intentionality, namely this view: not all phenomena are perceived; most of them—all physical and non-present mental phenomena—are rather imagined, remembered, conceptually thought, and the like. Those phenomena which are not perceived do not really exist in the present, although they do presently appear within mental acts insofar as they are presently imagined, remembered, etc. Thus, Brentano’s purpose in his theory of intentionality is to acknowledge and explain the fact that most phenomena are objects although they do not really exist. In other words: most of the objects that we represent and designate by nouns are inexistent objects. The use of a noun creates in us the illusion that a represented object exists outside of us. In the 1874 Psychology, Brentano claims that the existence of the external world is no more than ‘a fiction to which no reality of any sort corresponds’ (eine Fiktion, der keinerlei Wirklichkeit entspricht) (Brentano 1973a, 15–16; 1995, 10–11). And most importantly, Brentano, like Mauthner, holds that this fiction is of a linguistic nature. In a famous letter dated March 1905, he declares that the intentional object of a representation is not an existing object to which the representation corresponds, but merely the ‘linguistic correlate’ of an intentional verb (Brentano 1977, 120). The fact that a mental act can have an inexistent object is usually called ‘representational opacity’ and considered to be a constitutive property of intentionality. This property can be characterised as follows: it is possible that there exists a representation or intentional act x such that x represents an object α and α does not exist. The idea is that the verbs expressing intentional acts grammatically require a correlate—an α—and that this grammatical requirement sometimes misleads us into postulating an existing entity that is denoted by the name ‘α’. We could express the same idea by saying that Brentano here seeks to show that the rule in virtue of which one infers from ‘x represents α’ to ‘α exists’ is not valid. In other words, the rule of existential generalization is not valid for intentional sentences. It is not true that, from the existence of an act that has α as its object, one can safely infer that α exists. It is for this reason that it makes sense to regard Brentano’s rejection of entia rationis as a consequence of his theory of intentionality. A ‘fiction’, in

84 

D. Seron

Brentano’s sense of the word, is precisely something whose existence is erroneously inferred from the fact that it is presented and linguistically expressed by a name. Therefore, in a way at least, it makes sense to view Brentano’s theory of intentionality as part of a more general project in the vein of Mauthner’s critique of language. The illusion Brentano seeks to dispel through his theory of intentionality is very similar to what Mauthner calls ‘word fetishism’. In his Beiträge (Mauthner 1901–1902, I, 160 ff.), Mauthner says that the main task of the Sprachkritik is to warn us against ‘word fetishism’, and defines ‘word fetishism’ as the error consisting in inferring from the existence of a name to the existence of a named object: ‘Without a critique of language it will always remain possible to infer from the existence of a name to that of the named object, for example from the word deus to the existence of God’ (Mauthner 1901–1902, I, 173). Most interestingly, the same applies in the case of intentional objects. In his small essay on language published in 1906 in Martin Buber’s series ‘Die Gesellschaft’, Mauthner outlines an adverbial approach to intentionality. His idea is that the subject-object distinction is purely linguistic: it is no more than an illusion created by the grammar of the intentional idiom. Actually, sentences of the form ‘x represents α’ never involve any real duality and should be construed in the same way as ‘x builds a building’ or ‘x plays a game’.2 To illustrate his point, Mauthner tells a funny story from Voltaire’s Zadig (Mauthner 1901–1902, I, 174). Zoroaster has forbidden anyone to eat griffins. His disciples, who all know that griffins don’t exist, wonder how their master can have given them such a preposterous commandment. Some argue that the commandment is absurd, while others say: ‘Griffins must exist, since Zoroaster wants us not to eat them’. Zadig says this: ‘If there are any griffins, then let’s not eat them. If there aren’t any, then we will eat even less of them. That way we shall all be obeying Zoroaster’ (Voltaire 1879, 40; 2006, 118). What interests us here is the fetishist inference from the intentional sentence ‘Zoroaster wants us not to eat griffins’ to the sentence ‘griffins exist’. Once again, there remains a great difference between Brentano and Mauthner. The former’s claim is not that names never refer to real things. In sharp opposition to Mauthner, he maintains that there are names that

4  Brentano and Mauthner on Grammatical Illusions 

85

really refer to existent objects. As I said, the late Brentano makes a distinction between autosemantic expressions or real names (wahre, eigentliche Namen) and synsemantic expressions. The second category includes many expressions that are grammatically indistinguishable from real names. Leaving intentionality, I now pass to the two remaining kinds of fictions, namely the ego and concepts. At first sight, the gulf between the two authors’ views about the ego seems unbridgeable. Mauthner was an avowed pupil of Ernst Mach, who undoubtedly had a deep influence on his thought (Berlage 1994). In particular, Mauthner takes up Mach’s idea that psychophysical dualism is a discursive construct with no experiential basis for it. By contrast, Brentano strongly criticises Mach’s psychophysical monism, and his theory of intentionality as a whole can even be viewed as an attempt to rescue dualism from its empiricist critics (Seron 2019).3 In spite of this, Mauthner and Brentano defended quite similar views on mental entities. First, both follow up Friedrich-Albert Lange’s appeal to a ‘psychology without soul’ (Brentano 1973a, 16; 1995, 11; Mauthner 1922, 136; Mauthner 1906, 7 ff.; Le Rider 2012, 212). Mauthner, like Nietzsche and Mach with his thesis of the ‘unsalvageable ego’, holds that the ego is no more than a grammatical illusion or ‘fetish word’ (Mauthner 1922, 136). Secondly, it seems that Mauthner and Brentano agreed to some extent that mental entities were fictions. In his Vienna lectures on practical philosophy, Brentano claims that not only the ego, but also the entities denoted by psychological categories such as ‘passion’ or ‘pleasure’ are just abstracta or fictions and should be rephrased in mereological terms: We say, ‘I have overcome my passion (or my pain)’, but we would be more precise if we were to say ‘My inclination to do my duty overcame my leanings towards pleasure (or my self-pity)’ and still more precise—for abstractions are fictions—if we were to say, ‘One part of myself gained a victory over another part of myself ’. (Brentano 1952, 251; 1973b, 156)

This is not very different from Mauthner’s claim that ‘it is mythology to postulate abstracta (sensation, will) as causes of our mental states’ (Mauthner 1922, 136).

86 

D. Seron

The third and last kind of grammatical fictions to be considered are concepts. One of the most salient features of Mauthner’s critique of language is his vehement admonition against reifying universals—against what he calls ‘word realism’ (Wortrealismus).4 Similarly, Brentano’s late reism holds that only individuals can really exist, while universals— ‘Plato’s Idea of an animal, William of Champeaux’s general animal, “the animal” of Husserl’—are but fictions (Brentano 1977, 120, 178; Albertazzi 1990, 223). As said earlier, a fiction or ens rationis in Brentano’s sense is basically an ens linguae that is illusorily regarded as a real thing. A fiction can appear wherever a name does not express a concept psychologically analysable into intuitive elements, that is, into individual data. Such names, Brentano argues, create the ‘illusion’ (Täuschung) that ‘one has concepts whereas actually there are just words’ (Brentano 1977, 254). For this reason, psychological analysis centrally serves to demarcate fictions from real things.

4

Some Differences

Let us now pinpoint some substantive differences between Brentano and Mauthner. As Gershon Weiler has brilliantly shown in his monography on Mauthner, the most central point of difference between the two authors is that Brentano calls for a reformation of language, whereas Mauthner views any such reformation as doomed to fail (Weiler 1970, 296 ff.). Brentano thinks that ordinary language is misleading and should be amended; Mauthner claims that every language whatsoever must be misleading because it is in the nature of language to be misleading. For Brentano grammatical illusions are, so to speak, local illusions, while Mauthner rejects language in general as globally illusory. At least in his latest works, Brentano’s overall project is very similar to the project underlying Russell’s theory of descriptions (Weiler 1970, 297). Brentano’s view is that ordinary language is deceiving and should be replaced by a philosophical language which is not deceiving, namely a language grounded in psychological analysis. Ordinary language is deceiving because it uses expressions that are not referential as if they were referential. For example, Pegasus is the subject of the proposition

4  Brentano and Mauthner on Grammatical Illusions 

87

‘Pegasus is a poetical fiction’. Since the noun ‘Pegasus’ is the subject of a proposition, it seems that it must denote something that somehow exists. And of course this holds for the proposition ‘Pegasus doesn’t exist’ as well. Ordinary language misleads us into believing that even the objects that are correctly said not to exist must somehow exist. Thus, the philosopher’s task is to rephrase ordinary language in such a way that every subject of a proposition denotes something that really exists. For example, ‘Pegasus is a poetical fiction’ should be rephrased as ‘some representations that are experienced by poets and have Pegasus as their content exist’. The subject of the proposition—‘some representations that…’—here denotes something that really exists. This task of reformulating ordinary language involves at least two distinct things. First, as Weiler (1970, 297) emphasises, it involves ‘sorting out the referring expressions (those which are names of things) from those which are not’, that is, among other things, sorting out real from fake names. Secondly, the philosopher rephrases. Brentano here uses the word ‘translation’ (Übersetzung). In a 1907 dictation entitled “Vom Denken und vom ens rationis”, he says that the task is to ‘translate in a speech form with real names’ (Übersetzung in eine Sprechweise mit eigentlichen Namen) (Brentano 1977, 374). In a letter of 1906, he agrees with Marty that the assertions that accept entia rationis should be rephrased in equivalent assertions that have realia as their objects (Alle Behauptungen, welche Ihre entia rationis anerkennen würden, würden in Behauptungen, welche Realia zu Objekten haben, ihr Äquivalent haben) (Brentano 1977, 173). Mauthner is in total disagreement with such views. His rejection of them rests upon four theses which are extensively discussed in his works: (1) The first thesis is that the illusions of language are indestructible because they are motivated by practical interests and that these practical interests themselves cannot be eliminated (Mauthner 1906, 81–82). (2) The second thesis is that the notion of an ideal language is self-­ contradictory. Mauthner delivered a scathing critique of artificial ­languages like Esperanto, Volapük, and Leibniz’s lingua characteristica, arguing, among other things, that such languages are not different in character from natural languages (Mauthner 1906, 31 ff.).

88 

D. Seron

(3) The third thesis is that there is no universal or ‘philosophical’ grammar.5 Additionally, Mauthner considers all grammar intrinsically irrational and for this reason rejects the idea of a rational grammar as absurd (Mauthner 1906, 35–36). (4) Mauthner’s fourth and final thesis is that thought is reducible to language. As a nominalist, he claims that there are no such things as concepts or thoughts, and that what many call ‘thought’ is actually but ordinary language (Mauthner 1901–1902, I, 112, 176). Thus, there is no logic beside and beyond the grammar of ordinary language. The idea of a logical language that expresses thought more clearly and univocally than ordinary language is an illusion. Obviously, Brentano rejects all these theses. He is not a nominalist (Brentano 1977, 177–178). He believes that conceptual thought is different from language, that the latter can express the former more or less adequately, and therefore that logic—that discipline which prescribes how to express our thoughts adequately—is distinct from the grammar of ordinary language (Albertazzi 1990, 229). In his correspondence with Marty, Brentano holds that grammar (i.e. the grammar of ordinary language) is opposed to logic just as illusion is opposed to reality. A fiction corresponds to a fake name, namely to an expression which is used grammatically as a substantive while this is not its logical function: ‘The subject function is [here] merely grammatical, not logical’ (Brentano 1977, 373–374). Finally, Brentano believes that linguistic expressions have a logical or deep grammar function which is grounded in natural classifications revealed by psychological analysis. Mauthner, like Vaihinger later, rejects the idea of a natural classification as a myth. All our categories of thought come from ordinary language and are purely contingent (Mauthner 1901–1902, I, 76 ff.; Vaihinger 1922, 312 ff.; Weiler 1970, 228 ff.; Le Rider 2012, 195).6

4  Brentano and Mauthner on Grammatical Illusions 

5

89

Concluding Remarks

Brentano and Mauthner agree that ordinary language is deceptive insofar as it appears to refer to objects it actually does not refer to. They differ in the conclusions they draw from this fact. Mauthner concludes that language in general never refers to any object whatsoever. Linguistic reference itself is a deceiving appearance: the actual function of language in general is expressive rather than referential. As Le Rider emphasises, ‘it is not such-and-such usage of language that Mauthner criticises, but human language in general. In his view, it is not possible to rectify or amend language so as to make it usable as an instrument for knowledge’ (2012, 221–222). By contrast, Brentano, like Russell and others, claims that ordinary language actually refers to other objects than those it appears to refer to. Thus, it must be possible to elaborate a philosophical language that appears to denote what it really denotes—a language that explicitly refers to present mental life by univocally articulating concepts directly abstracted from inner perception. At a more general level, Mauthner’s and Brentano’s accounts of grammatical illusions may as well be viewed as different ways of dealing with one and the same paradox generated by the radical form of empiricism— something in the vein of Mach’s monism of sensation—they both largely adhere to. This paradox may be explained as follows: ( 1) Knowledge needs to somehow relate to objects that really exist. (2) All that can be said to really exist are the individual data of present perception. (3) As a matter of fact, knowledge (in the normal sense of scientific knowledge) is about objects that are not presently perceived and thus cannot be said to really exist. In Brentano’s view, the physical phenomena studied in the natural sciences do not exist in any sense. Likewise, psychological knowledge deals with either the psychologist’s own mental phenomena she remembers or mental phenomena she attributes to others. In either cases the psychologist pretends to ‘know’ objects that she does not presently perceive and thus cannot truly recognise as really existing (Seron 2017a, b).

90 

D. Seron

Such considerations—including the view that ‘all our representations, sensorially produced, can only relate to the past’ (1922, 137)—lead Mauthner to the sceptical conclusion that the concept of knowledge is self-contradictory. Indeed, knowledge conceptually involves both language and referring to reality.7 But language is essentially fictitious and thus ‘impotent as an instrument for knowledge’ (Mauthner 1901–1902, I, 94). Since the objects of knowledge cannot be both real things (1) and fictions (3), knowledge is therefore impossible. By contrast, Brentano’s theory of intentionality may be seen as an attempt to block the sceptical implications of the above paradox. His solution is to distinguish the object of knowledge from its experiential source: knowledge relates to really existing objects that are numerically distinct from the objects it is about. For example, psychological knowledge is rooted in perceptual experience not in the sense that it is about mental phenomena that are presently perceived (which is obviously not the case), but in the sense that the psychologist presently perceives her act of remembering the phenomenon her knowledge is about, that is, an act in which the object of psychological knowledge appears in another mode than perception. Brentano’s theory of intentionality seeks to clarify how some objects—as ‘intentional objects’—appear and are objects of knowledge without existing.8

Notes 1. Albertazzi (2006, 177–178) provides the following list of entia rationis: ‘1. All abstract nouns like “extension”, “colour”, “thought”, “space”, “time”, given that in reality there exists only the extended, the coloured, the thinking, something spatial, temporal. 2. Nouns expressing possibility like “capacity for thought”, “capacity for movement”. 3. The contents of presentations like “the non-being of the centaur is”, “a non-existent centaur is”. 4. Whatever belongs to the Aristotelian category of ens per accidens. For example, in “this man is armed” the predicate has solely the meaning of an extrinsic determinatio, given that there is no real identity between subject and predicate. 5. Whatever belongs to the Aristotelian category of ens tamquam verum. For example, one can only say that “an

4  Brentano and Mauthner on Grammatical Illusions 

91

oak is” if one recognises an oak; or “the non-being of the centaur” only means that one denies a centaur. 6. All scientific fictions, like unreal numbers, fractions, infinity and infinitesimals in mathematics’. 2. See (Mauthner 1906, 72–73; 1910, 370), where Mauthner refers to the medieval-scholastic notion of intentional object. Brentano would reject such a formulation due to his commitment to some form of empirical dualism. I will discuss this point later on. 3. For Mauthner’s Mach-inspired monism, see, for example, (Mauthner 1910, 370–371). On this difference between Brentano and Mauthner, see (Albertazzi 1986, 43–44; Bredeck 1992, 106 ff.; Le Rider 2012, 242). 4. ‘The medieval-scholastic realism—which I always call “word realism” for the sake of distinction—teaches that universals or concepts are something real, that, for example, to the […] concepts of stallion, horse, quadruped, animal, organism, thing, etc., corresponds in the real world something that is not an individual and yet is really a stallion, a horse, etc.’ (Mauthner 1901–1902, III, 616–617). See the related notion of ‘concept realism’ (Begriffsrealismus) in (Vaihinger 1922, 403 ff.). 5. ‘The human language in itself is an abstractum, an intangible shadow just like the old mental faculties; the human language in itself has absolutely no grammar, and hence no philosophical grammar’ (Mauthner 1901–1902, I, 193). ‘It would be timely to cease to dream of a philosophical grammar. There is no general grammar, let alone a philosophical one’ (Mauthner 1901–1902, III, 261, transl. by Weiler 1970, 146). Max Scheler was entirely right in his 1914 review of the third volume of Mauthner’s Beiträge (second edition), when he opposed to Mauthner Husserl’s project of a ‘pure grammar’ (Scheler 1914, 119). Mauthner criticises Husserl’s idea of an ‘objective-ideal meaning’ in (Mauthner 1910, 90). It is also worth noticing that herein lies a major difference between Mauthner’s and the Tractarian Wittgenstein’s critiques of language (Nájera 2007, 161; Gakis 2012, 89). The question is whether the illusions of ordinary language are constitutive of language in general, or whether a language grounded in logical or psychological analysis can liberate us from them. Also see below on Brentano’s distinction between logic and ordinary language grammar. 6. Albertazzi very rightly opposes Mauthner’s’substantial conventionalism’ to Brentano’s project of a ‘reform of language’ that ‘in no way reduces the possibility of an ontology, albeit strongly reist, guaranteed in the last analysis by the fact of inner perception relative to the representation of things,

92 

D. Seron

and guarantor in its turn of the symbolic translation performed by the common innere Sprachform’ (Albertazzi 1989, 149, 154; see also Albertazzi 1996, 103). 7. Mauthner rejects the idea of a non-linguistic knowledge. See (Weiler 1970, 292). 8. I have developed this interpretation of Brentano’s theory of intentionality in (Seron 2017a, b, 2019).

References Albertazzi, Liliana. 1986. Fritz Mauthner: La critica della lingua. Lanciano: Carabba. ———. 1989. Brentano and Mauthner’s Critique of Language. Brentano Studien 2: 145–159. ———. 1990. Nominalismo e critica della lingua in Franz Brentano. Idee 13–15: 217–232. ———. 1996. Anton Marty 1847–1914. In The School of Franz Brentano, ed. Liliana Albertazzi, Massimo Libardi, and Roberto Poli. Dordrecht: Springer. ———. 2006. Immanent Realism: An Introduction to Brentano. Dordrecht: Springer. Berlage, Andreas. 1994. Empfindung, Ich und Sprache um 1900: Ernst Mach, Hermann Bahr und Fritz Mauthner im Zusammenhang. Frankfurt am Main: Peter Lang. Bredeck, Elizabeth. 1992. Metaphors of Knowledge: Language and Thought in Mauthner’s Critique. Detroit: Wayne State University Press. Brentano, Franz. 1952. Grundlegung und Aufbau der Ethik. Hamburg: Meiner. ———. 1970. Versuch über die Erkenntnis. Hamburg: Meiner. ———. 1973a. Psychologie vom empirischen Standpunkt. Hamburg: Meiner. ———. 1973b. The Foundation and Construction of Ethics. Trans. Elizabeth Hughes Schneewind. London and New York: Routledge. ———. 1977. Die Abkehr vom Nichtrealen. Hamburg: Meiner. ———. 1995. Psychology from an Empirical Standpoint. Trans. Antos C.  Rancurello, D.B.  Terrell, Linda L.  McAlister. London and New  York: Routledge. ———. n.d. Ms. EL 80. Logik. Unpublished Manuscript. Provisional Online Edition. ed. Robin D. Rollinger. Franz Brentano Archiv Graz. http://gams. uni-graz.at/archive/objects/context:bag/methods/sdef:Context/get?mode= logik

4  Brentano and Mauthner on Grammatical Illusions 

93

Cesalli, Laurent, and Kevin Mulligan. 2017. Marty and Brentano. In The Routledge Handbook of Franz Brentano and the Brentano School, ed. Uriah Kriegel, 251–253. New York: Routledge. Gakis, Dimitris. 2012. Contextual Metaphilosophy: The Case of Wittgenstein. Amsterdam: ILLC. Huemer, Wolfgang. 2018. Franz Brentano. In The Stanford Encyclopedia of Philosophy (Fall 2018 Edition), edited by Eduard N.  Zalta. URL: https:// plato.stanford.edu/archives/fall2018/entries/brentano/. Kastil, Alfred. 1951. Die Philosophie Franz Brentanos: Eine Einführung in seine Lehre. Bern: A. Francke. Kraus, Oskar. 1929. Selbstdarstellung. Leipzig: Meiner (Separate reprint from Die Philosophie der Gegenwart in Selbstdarstellungen, vol. 7, 161–203). Le Rider, Jacques. 2012. Fritz Mauthner : Scepticisme linguistique et modernité. Une biographie intellectuelle. Paris: Bartillat. Libardi, Massimo. 1996. Franz Brentano (1838–1917). In The School of Franz Brentano, ed. Liliana Albertazzi et al., 25–79. Dordrecht: Kluwer. Mauthner, Fritz. 1901–1902. Beiträge zu einer Kritik der Sprache. Vol. 3. Stuttgart: J.G. Cotta. ———. 1906. Die Sprache. Frankfurt am Main: Rütten u. Loening. ———. 1910. Wörterbuch der Philosophie: Neue Beiträge zu einer Kritik der Sprache. Vol. 1. Munich and Leipzig: Georg Müller. ———. 1922. Fritz Mauthner. In Die Philosophie der Gegenwart in Selbstdarstellungen, ed. Raymund Schmidt, vol. 3, 121–144. Leipzig: Meiner. Nájera, Elena. 2007. Wittgenstein versus Mauthner: Two Critiques of Language, Two Mysticisms. Papers of the 30th International Wittgenstein Symposium 5–11 August 2007, edited by Herbert Hrachovec and Aloïs Pichler. Kirchberg am Wechsel: ALWS. Scheler, Max. 1914. “Review of (Mauthner 1901–1902)”, vol. 3, 2nd ed. Die weissen Blätter 6: 118–119. Seron, Denis. 2017a. Brentano’s Project of Descriptive Psychology. In The Routledge Handbook of Franz Brentano and the Brentano School, ed. Uriah Kriege, 35–40. New York: Routledge. ———. 2017b. Brentano on Appearance and Reality. In The Routledge Handbook of Franz Brentano and the Brentano School, ed. Uriah Kriegel, 169–177. New York: Routledge. ———. 2019. Intentionality vs. Psychophysical Identity. In Ernst Mach—Life, Work, and Influence / Ernst Mach—Leben, Werk und Wirkung, ed. Friedrich Stadler. Dordrecht: Springer.

94 

D. Seron

Vaihinger, Hans. 1922. Die Philosophie des Als-Ob: System der theoretischen, praktischen und religiösen Fiktionen der Menschheit auf Grund eines idealistischen Positivismus. 7th and 8th ed. Leipzig: Meiner. Voltaire. 1879. Zadig. In Œuvres complètes, vol. 21, 31–93. Paris: Garnier. ———. 2006. Zadig. Trans. Roger Pearson in Candide and Other Stories, 107–177. Oxford: Oxford University Press. Weiler, Gershon. 1958. On Fritz Mauthner’s Critique of Language. Mind 67 (265): 80–87. ———. 1970. Mauthner’s Critique of Language. London: Cambridge University Press.

5 Misleading Expressions: The Brentano-­Ryle Connection Arnaud Dewalque

Some linguistic expressions are misleading in the sense that they look as if they are about something while they actually are about something else. In this chapter I argue that Gilbert Ryle’s account of misleading expressions, which is rightly considered a milestone in the history of analytic philosophy, is continuous with Brentano’s critique of language. Not only did they identify roughly the same classes of misleading expressions, but their analyses are driven by a form of ontological parsimony which sharply contrasts with rival views in the Brentano School, like those of Meinong and Husserl. It is true that Brentano’s account, unlike Ryle’s, is put in terms of underlying mental phenomena. However, this difference, I submit, is mainly terminological and does not reflect any substantial disagreement. The chapter has four sections. Section 1 (‘Analysis’) suggests that Ryle and Brentano share a similar notion of analysis as paraphrase of misleading expressions. Section 2 (‘Two Senses of “About”’) spells out the notion of misleading expression by means of the surface-grammar/

A. Dewalque (*) Department of Philosophy, University of Liège, Liège, Belgium e-mail: [email protected] © The Author(s) 2021 A. Dewalque et al. (eds.), Philosophy of Language in the Brentano School, History of Analytic Philosophy, https://doi.org/10.1007/978-3-030-52211-7_5

95

96 

A. Dewalque

truth-­conditions distinction, which I argue is implicit in their accounts. Section 3 (‘Ficta’) zooms in on a specific class of misleading expressions, namely expressions about ficta. Finally, Sect. 4 (‘A Moral About the Meaning of “Meaning”’) draws the consequences of what precedes for a correct understanding of the notion of meaning.

1

Analysis

In this section I suggest that Ryle and Brentano share a similar view of analysis. On this view, the latter is best understood as a type of paraphrase which prevents professional philosophers of being misled by language into multiplying entities without necessity. In other words, their conception of analysis is motivated at the outset by a principle of ontological parsimony traditionally associated with Occam’s razor. Before getting there, though, let me very briefly recall some background information about Ryle’s relation to phenomenology in general and to Brentano in particular. It is common knowledge that Ryle, who started his career with a book review of Roman Ingarden in Mind in 1927 and visited Husserl in Freiburg (presumably) in 1929 (Schuhmann 1977, 340), was thoroughly familiar with the Austro-German phenomenological tradition, for which he showed a continued interest throughout his life. Over the last two decades, a growing number of studies have pointed at his mediating role between phenomenology and analytic philosophy (see e.g. Brandl 2002; Bourdeau 2003; Thomasson 2002, 2007; Vrahimis 2013, 110–159; Morran 2014, 254–259; Chase and Reynolds 2017). Unsurprisingly, one of the chief things those studies reveal is that there is something of a tension in Ryle’s position. On the one hand, indeed, he had strong sympathies for the phenomenological project and went as far as acknowledging that his major book, The Concept of Mind, might be described as ‘a sustained essay in phenomenology’ (Ryle 2009a, 196); on the other, he challenged dubious claims made by Husserl (and Heidegger)—claims he believed should be excised from the phenomenological project (Thomasson 2002, 137). My hypothesis is that this tension is best explained against the background of another tension which is internal to the Brentano School, namely the opposition between Brentano’s ontological

5  Misleading Expressions: The Brentano-Ryle Connection 

97

parsimony, on the one hand, and the more liberal views held by Husserl and Meinong, on the other. On Brentano’s account, indeed, both Meinong and Husserl are typical examples of philosophers who have been misled by language into multiplying entities without necessity, in blatant violation of Occam’s razor. Very roughly, I contend that the early Ryle endorsed the same line of thought and took sides with Brentano against Meinong’s object theory and Husserl’s so-called ‘Platonism of meaning’. In the course of this chapter, I will gather some evidence to substantiate this hypothesis (although I will focus more on Meinong’s object theory than on Husserl’s Platonism). My starting point lies in the following observation: however different their metaphilosophical views may be, Ryle and Brentano share the idea that philosophers may be misled by language not just occasionally but in some significant and systematic way. In his Psychology from an Empirical Standpoint of 1874, for example, Brentano bluntly states that, when it comes to describing and classifying mental phenomena, ordinary language ‘offers no sufficient guarantee and would mislead [in Irrthümer führen] those who relied on it too much, just as it would facilitate the discovery of the truth for those who utilise its determinations with caution’ (Brentano 1924, 63; 1995, 45; my translation). Later in the book, in a long footnote devoted to his debate with Mill on existential propositions, he writes: ‘I hope that people will finally, once and for all, stop confusing linguistic differences with differences in thought’ (Brentano 1925, 63, note; 1995, 220, note). What Brentano is up against in those passages is what he often calls ‘linguistic fictions’ or fictions created by language—a topic which is usually associated with his so-called ‘reistic turn’ but which is far from being absent from his earlier writings.1 As we shall see, ‘linguistic fiction’ is just another name for what Ryle calls ‘misleading expression’ or what may be termed ‘grammatical illusion’ (see Seron, this volume). I shall give examples of such expressions momentarily. Ryle’s account of misleading expressions was first presented in his speech to the meeting of the Aristotelian Society in London on March 21, 1932, the same year he wrote his well-known critical essay on “Phenomenology”. Both essays have initially been published in the Proceedings of the Society (Ryle 1932a, b; reprinted in 2009a, 174–185; 2009b, 41–65). Ryle’s study on misleading expressions is usually

98 

A. Dewalque

considered a significant document in the history of analytic philosophy—with good reason, since it aims at establishing the legitimacy of a philosophical research programme in which the notion of analysis plays the central role. His main concern is with what a philosophical analysis of linguistic meaning can achieve and whether such analysis is needed at all. His diagnosis has a negative and a positive side. On the negative side, Ryle insists that it cannot be a legitimate goal for philosophy to (i) clarify the meaning of linguistic expressions used in ordinary discourse, for it is already known by the speakers who use them, nor to (ii) paraphrase them in order to ensure the effectiveness or elegance of linguistic communication in real-life situations, for this is the business of linguistic, lexical or philological analysis (Ryle 1932b, 139–141; 2009b, 41–43).2 On the positive side, even once these misconceptions of philosophical analysis have been discarded, he argues, there remains ‘an important sense in which philosophers can and must discover and state what is really meant by expressions of this or that radical type’ (Ryle 1932b, 142; 2009b, 43). How, then, should we conceive of philosophical analysis? Very roughly, Ryle’s reply is as follows. The point of philosophical analysis is to avoid philosophers—and, generally speaking, ‘any man who embarks on abstraction’ (Ryle 1932b, 146; 2009b, 46)—to be misled by the surface grammar of some linguistic expressions into drawing false conclusions of a logical, ontological or epistemological nature. In other words, the ensuing programme of analysis is philosophical through and through: it is to be carried out, so to speak, by philosophers for philosophers, merely for the sake of avoiding philosophical mistakes. Thus understood, philosophical analysis is not a matter of paraphrasing some expressions into more elegant, familiar or readily intelligible ones. It is a matter of paraphrasing them into less elegant but less misleading expressions for the sake of pursuing philosophical (i.e. ontological, epistemological, etc.) inquiries. Interestingly, this understanding of philosophical analysis is in line with Brentano’s criticism of linguistic fictions. Here again, the point is not to reform ordinary language. As a matter of fact, Brentano has it that such a reform is impossible for misleading expressions are an integral part of ordinary language, to the effect that eliminating them would be

5  Misleading Expressions: The Brentano-Ryle Connection 

99

tantamount to replacing ordinary language with a different language altogether. In a dictation on ens rationis dated January 6 1917, he writes: Naturally, this variety of locutions [i.e. misleading expressions] arises from complications in our thinking. They make possible abbreviated discourse which is highly advantageous. The whole of ordinary language is so much under their influence that we could not possibly give them up without giving up the use of that language completely and resolving to invent an entirely new and extremely unwieldy language. (Brentano 1925, 275; 1995, 367)

As we shall see, the idea of ‘complications in our thinking’ (Verwicklungen in unserem Denken) has to be understood literally: on Brentano’s view, misleading expressions typically express mental phenomena directed at other mental phenomena, nested into each other like Russian dolls—or so I shall argue (more on that in Sect. 3). For now, though, suffice it to say that both Ryle and Brentano agree on the following claims: (i) misleading expressions are an integral part of ordinary language; (ii) in non-­ philosophical, ordinary context, they offer convenient and harmless abbreviations for complicated states-of-affairs; (iii) taken at face value, however, such expressions may mislead professional philosophers into multiplying entities without necessity. Accordingly, the point of analysing misleading expressions merely is to apply Occam’s razor (Ryle 1932b, 165; 2009b, 61) and avoid ‘infinite complications’ (Brentano 1925, 160; 1995, 292–293).3 What do such misleading sentences look like? Ryle takes it that the types of misleading expressions are ‘in principle unlimited’ in number, even though ‘the number of prevalent and obsessing types is fairly small’ (Ryle 1932b, 169; 2009b, 64).4 He identifies three main classes thereof, to which a fourth one may be added, namely:

Quasi-ontological statements ( 1) ‘Satan is not a reality’ (2) ‘Unicorns do not exist’

100 

A. Dewalque

Quasi-Platonic statements ( 3) ‘Unpunctuality is reprehensible’ (4) ‘Colour involves extension’

Quasi-referential the-phrases ( 5) ‘Poincaré is not the King of France’ (6) ‘Jones hates the thought of going to the hospital’

[Modifying expressions]5 ( 7) ‘Jones is an alleged murderer’ (8) ‘Smith is a probable Lord Mayor’ As far as I know, Brentano didn’t offer any classification of its own, but his account of linguistic fictions includes at least a critical discussion of quasi-ontological statements, quasi-Platonic statements and modifying expressions. On the whole, then, I think it is fair to say that Ryle and Brentano roughly pinned down the same main classes of misleading expressions. Now, if they are right, there must be a sense in which sentences (1)–(8), however different, may all be said to be misleading in a similar way. Spelling out this sense is the business of the next section.

2

Two Senses of ‘About’

In this section I explain what it means for sentences like (1)–(8) to be misleading in the sense here discussed. Drawing on (Ryle 1933, 2009b, 86–88), the proposed explanation hinges on the distinction between two senses of ‘about’. As stated at the outset, the notion of misleading expression applies to cases where a linguistic expression—typically, a sentence— looks as if it is about something while in fact it is about something else. But how are we to understand the difference between what a sentence seems to be about and what it actually is about? More pointedly, what clue do we have as to what a sentence actually is about? I argue that,

5  Misleading Expressions: The Brentano-Ryle Connection 

101

whereas what a sentence seems to be about is determined by its surface grammar (starting with its grammatical subject), what it actually is about is determined by its truth conditions. Accordingly, successfully analysing a misleading expression amounts to paraphrasing it into a non-misleading expression with the same truth conditions. This, I submit, is at bottom the view held by Ryle and Brentano, as I interpret them. Consider (1) ‘Satan is not a reality’. It has the same surface grammar than ‘Capone is not a philosopher’, and yet, whereas the latter denies a certain property (namely, that of ‘being a philosopher’) to the individual called ‘Capone’, it would be an error to say that the former denies a certain property (that of ‘being a reality’) to an individual called ‘Satan’. Despite the grammatical appearances, the sentence ‘Satan is not a reality’ is not about Satan in the way in which ‘Capone is not a philosopher’ is about Capone (Ryle 1932b, 148–149; 2009b, 48). (1) refers, if at all, to something else, namely the fact that there is no individual which is called ‘Satan’ and exhibits the related features (is devilish, etc.). A similar observation, Ryle suggests, applies to (2)–(8): (2) looks as if it is denying existence to unicorns in the way in which ‘Cows are not carnivorous’ (say) denies the property of being carnivorous to cows; (3) looks as if it ascribes the property of being reprehensible to unpunctuality in the way in which ‘Paul is eligible (to become governor)’ ascribes the property of being eligible to Paul; (4) looks as if it is ascribing to colour the property of involving extension in the way in which ‘The final list involves Paul’s name’ ascribes to the final list the property of involving Paul’s name; (5) looks as if it is about the King of France in the way in which ‘This is the picture which is on my desk’ is about the picture which is on my desk, etc. Sentences (1)–(8) are misleading in the sense that they look as if they refer to something while that actually refer to something else, which is not explicitly named in them. Here is yet another example: (9) ‘Centaurs are fictional creatures’. Now take the following, non-misleading sentence: (10) ‘Whales are marine mammals’.

102 

A. Dewalque

Plainly, the grammatical structure of (9) is identical to that of (10): they have exactly the same surface grammar. However, whereas (10) ascribes the property of being marine mammals to whales, (9) does not ascribe the property of being mythical creatures to centaurs. ‘Though the grammatical appearances are to the contrary’, as Ryle puts it (1932b, 144; 2009b, 45), (9) is not about centaurs in the way in which (10) is about whales, for the word ‘centaurs’ does not signify a ‘subject of attributes’. Rather, (9) is a shorthand for what would be less conveniently, albeit less misleadingly, expressed by means of another sentence like, perhaps: (9*) ‘There is no creature in the non-fictional world which is half-man and half-horse’ (or whatever properties are taken to be definitional of centaurs). It is important to note that systematically misleading sentences are not per se syntactically ill-formed (e.g. ‘green is where’), nonsensical (‘the number three is screaming loudly’) or false (‘Descartes is born in the 1400s’). On the contrary, they do not violate any rule of grammar or semantics. What is more, they are not to be conflated with equivocal expressions (‘this curry is hot’), which admittedly are occasionally misleading—even though, most of the time, the context of utterance usually contributes to remove any doubt as to their actual meaning. Again, the ‘plain man’ is in no way misled or puzzled by them. They are a source of trouble for philosophers only. Surely, at this point, it would be good to have a definition of the notion of misleading expression. Ryle himself does not offer any formal definition but describes the situation as follows: (9) and (10), despite having the same surface grammar, actually refer to ‘different types’ of states-of-­ affairs or to facts with a ‘different logical structure’. Now when two sentences with the same grammatical structure refer to different types of states-of-affairs, taking the linguistic expressions ‘at face value’ can result in one’s becoming oblivious to the differences between the types of facts referred to. Hence, Ryle suggests that the following holds true of all systematically misleading expressions: they ‘are couched in a syntactical form improper to the facts recorded and proper to facts of quite another logical form than the facts recorded’ (Ryle 1932b, 143; 2009b, 44).

5  Misleading Expressions: The Brentano-Ryle Connection 

103

Admittedly, this is hardly a definition. Besides, this description is utterly silent on what clue one might have as to what fact is actually ‘recorded’. In his three-page article for the first issue of the journal Analysis, though, Ryle introduces an interesting distinction between various senses of ‘about’ which, as it happens, makes it possible to spell out the notion of misleading sentence in a somewhat more precise fashion. He distinguishes between what a sentence is about in a linguistic sense and what it is about in a referential sense (Ryle 1933, 2009b, 86–88). Thus, (1) ‘Satan is not a reality’ is about Satan in a merely linguistic (and nominal) sense. Yet, it does not refer to Satan, since there is no such a thing as an individual called Satan. The distinction can be captured by means of the following notation: sentence (1) is ‘about (l)’ Satan (in the linguistic sense of ‘about’) while it is not ‘about (r)’ Satan (in the referential sense of ‘about’). Note that, contrary to what this example suggests, what a sentence is about (l) is not exhausted by its grammatical subject. On Ryle’s account, the sentence (5) ‘Poincaré is not the King of France’ is ‘nominatively’ (i.e. in virtue of its grammatical subject) ‘about (l)’ Poincaré, but it is also ‘about (l)’ the King of France, just like the sentence (8) ‘Smith is a probable Lord Mayor’ is both ‘about (l)’ Smith and ‘about (l)’ a probable Lord Mayor. The important thing, however, is that they are not ‘about (r)’ the King of France or a probable Lord Mayor. Again, they refer, if at all, to something else. My suggestion is that this distinction between the linguistic and the referential use of ‘about’ may be employed to define the notion of systematically misleading sentence. The definition I have in mind is as follows: Systematically Misleading Sentence For any sentence S, S is systematically misleading if, and only if, (i) S is ‘about (l)’ x and (ii) S is ‘about (r)’ y and (iii) x ≠ y.

On the face of it, this sounds quite promising. But we are still faced with the same difficulty: how are we supposed to find out what a given expression is ‘about (r)’? Ryle is not very explicit on that score, to say the least, but I suspect that the notion of truth conditions offers the needed leverage here. Recall that systematically misleading sentences like (1) ‘Satan is not a reality’ or (9) ‘Centaurs are fictional creatures’ are not false.

104 

A. Dewalque

In most cases they are true, and they are so in the perfectly common sense that some fact or state-of-affairs makes them true. Yet, even when they are not true, they have truth conditions and this, I submit, is of paramount importance whenever it comes to paraphrasing them into less misleading expressions. Note well: Ryle himself does not talk of truth conditions; yet, if I’m not mistaken, this notion captures what he and Brentano are after in their analyses. Take (6) ‘Jones hates the thought of going to the hospital’. It looks as if it is about an entity referred to as ‘the thought of going to the hospital’, but this grammatical appearance is deceptive, Ryle says. How can we tell? Reply: by considering its truth conditions. ‘For it to be true’, he writes, ‘the world must contain a Jones who is sometimes thinking and sometimes, say, sleeping; but it need no more contain both Jones and “the thought or idea of so and so” than it need contain both someone called “Jones” and something called “Sleep”’ (Ryle 1932b, 161; 2009b, 58; my emphasis). Compare Brentano: ‘The proposition, “A centaur is a poetic fiction,” does not imply […] that a centaur exists, rather it implies the opposite. But if it is true, it does imply that something else exists’ (Brentano 1925, 61; 1995, 219; my emphasis). As I interpret them, those passages speak in favour of reframing Ryle and Brentano’s conception of analysis in terms of truth conditions. In order to find out what a sentence really is about (r), all you have to do is inquire what should be the case for the sentence to be true. Let me take stock. Saying that linguistic expressions may be misleading in a systematic or non-occasional way amounts to saying that there is sometimes a gap between what an expression seems to be about and what it actually is about. On the proposed interpretation, this difference is best understood when applied to sentences—that is, truth-assessible expressions. What a sentence seems to be about is determined by its surface grammar, including its grammatical subject. (9) looks as if it is about centaurs in virtue of having the word ‘centaurs’ as grammatical subject, just like (5) looks as if it is about the King of France as having ‘the King of France’ as grammatical predicate. By contrast, what a sentence actually is about is determined by its truth conditions. For (9) to be true, it is not necessary that there exist centaurs which have the property of being fictional creatures, etc. As a result, I suggest understanding systematically misleading expressions as those which generate systematically misleading

5  Misleading Expressions: The Brentano-Ryle Connection 

105

sentences and defining the latter as follows, with ‘aboutSG x’ meaning ‘about x in virtue of its surface grammar’ and ‘aboutTC y’ meaning ‘about y in virtue of its truth conditions’: Systematically Misleading Sentence* For any sentence S, S is systematically misleading if, and only if, (i) S is aboutSG x and (ii) S is aboutTC y and (iii) x ≠ y.

To my opinion, this understanding of misleading expressions is implicit in both Ryle and Brentano. It captures the sense in which sentences (1)– (8) may be said to be misleading. In virtue of their surface grammar, indeed, (1)–(8) are aboutSG Satan, unicorns, unpunctuality, colour, the King of France, the thought of going to the hospital, a murderer and a Lord Mayor. Yet, they are not aboutTC Satan, unicorns, unpunctuality, etc. In other words, they are all misleading for the same reason: what they are aboutSG is not identical to what they are aboutTC. In Sect. 1 I have suggested that both Brentano and Ryle share a certain notion of philosophical analysis. Philosophically analysing misleading expressions like (1)–(8) amounts to successfully paraphrasing them into non-misleading (or, at any rate, less-misleading) ones. We are now in a position to get a better grip on the relevant notion of analysis. All is needed is to define in turn the notions of non-misleading sentence and successful paraphrase. Given what precedes, the following definitions suggest themselves: Non-Misleading Sentence For any sentence S, S is a non-misleading sentence if, and only if, (i) S is aboutSG x and (ii) S is aboutTC y and (iii) x = y. Successful Paraphrase For any pair of sentences S1 and S2, S2 is a successful paraphrase of S1 if, and only if, (i) S1’s surface grammar ≠ S2’s surface grammar and (ii) S1’s truth conditions = S2’s truth conditions and (iii) S2 is a non-­misleading sentence.

With those definitions in hand, I now turn to the analysis of some misleading sentences. For reasons of limited space, I will zoom in on

106 

A. Dewalque

Ryle’s first class of misleading expressions, namely ‘quasi-ontological statements’ or statements about ficta, and compare Brentano’s analysis thereof with Ryle’s.

3

Ficta

Recall Ryle’s examples of quasi-ontological statements: ( 1) ‘Satan is not a reality’. (2) ‘Unicorns do not exist’. Although those sentences are aboutSG Satan and unicorns (respectively), they are not aboutTC Satan and unicorns. For (1) to be true, indeed, it is not required that there is an individual which is called ‘Satan’ and which is devilish and of which it is true that it is not a reality. Similarly, for (2) to be true it is not required that there are horses having a horn and of which it is true that they do not exist. What is required, then? What exactly are the truth conditions of (1) and (2)? In this section I review three options I call the Lockean, the Meinongian and the Brentanian option. According to the Lockean option, quasi-ontological statements are aboutTC ideas; according to the Meinongian option, they are aboutTC non-existent objects; and according to the Brentanian option, they are aboutTC mental phenomena. I argue that Brentano and Ryle both reject options 1 and 2 and endorse option 3. The Lockean option: quasi-ontological statements are aboutTC ideas.— Although it is associated with Locke’s notion of ‘ideas in the head’, the claim that a name does not mean the thing itself but our idea of the thing is usually traced to Hobbes. On this account, (1) is true in virtue of there being an idea of Satan in my mind. John Stuart Mill famously objected that this view is incompatible with a correct analysis of the underlying mental phenomena. When we use the word ‘sun’, Mill argues, we do not name our idea of the sun, but the sun itself. His argument rests on the view that we do not only use words in order to communicate some presentations of a thing, but also in order to indicate that we judge a thing to be so-and-so. Now such judgements are not about our idea of the

5  Misleading Expressions: The Brentano-Ryle Connection 

107

thing, but about the thing itself. For example, when I say that ‘the sun shines’, surely I do not mean that my idea of the sun shines or that my idea of the sun involves the idea of shining. Rather, this sentence expresses an act of judging which is aboutTC the sun itself: my judgement is true if the sun actually shines (of course, it is also aboutSG the sun; ‘the sun shines’ is a non-misleading sentence). Mill writes: Names are not intended only to make the hearer conceive what we conceive, but also to inform him what we believe. Now, when I use a name for the purpose of expressing a belief, it is a belief concerning the thing itself, not concerning my idea of it. When I say, ‘the sun is the cause of the day’, I do not mean that my idea of the sun causes or excites in me the idea of day; or in other words, that thinking of the sun makes me think of day. I mean, that a certain physical fact, which is called the sun’s presence […] causes another physical fact, which is called day. (Mill 1974, 25)

Of course, Mill’s example is not about a fictum, so the question arises whether it is any different with names of ficta like ‘Satan’ and ‘unicorns’. Shouldn’t we retain Hobbes’ suggestion when it comes to sentences such as (1) and (2)? Both Ryle and Brentano answer this question negatively: no, pace Hobbes, (1) and (2) are not aboutTC ideas (Ryle 1932b, 144; 2009b, 45).6 Plainly, (1) cannot possibly be paraphrased by saying something like ‘my idea of Satan is not a reality’ or ‘my idea of Satan does not involve the idea of a reality’ or ‘my idea of Satan does not involve the feature “real”’ (whatever that would mean). Likewise, (2) does not mean something like ‘my idea of unicorns does not exist’ or ‘my idea of unicorns does not involve the idea of existence’. The moral seems to be that, when uttering (1) and (2), we are not talking aboutTC ideas in the mind. So much for option 1. What about option 2? The Meinongian option: quasi-ontological statements are aboutTC non-­ existent objects.—This is the option favoured by Meinong and the supporters of object theory. In fact, on Meinong’s view, (1) and (2) are not misleading sentences. What they are aboutSG is identical to what they are aboutTC, namely fictional objects or ficta, objects of which it is true to say that they don’t exist. Therefore, there is no need to paraphrase (1) and (2). This view, however, raises well-known difficulties. Like option 1, option

108 

A. Dewalque

2 is based on the thought that something may be without existing. Now this thought is highly problematic. Brentano rejects it outright: ‘I confess that I am unable to make any sense of this distinction between being and existence’ (Brentano 1925, 137; 1995, 274). Brentano’s main reason for rejecting the distinction between being and existing is that it generates contradictions. For example, if we follow Meinong, we are led to attribute a being to things which we correctly deny, i.e. which are not, and thus we are led to speak of ‘existing unicorns’, say, as being and non-being at the same time, which contravenes the principle of non-contradiction. Russell makes a similar point against Meinong in his well-known essay, “On Denoting” (Russell 1905, 1973, 117–118). Ryle shares Brentano and Russell’s opinion: Some argued that the statement [(1)] was about something described as ‘the idea of Satan’, others that it was about a subsistent but non-actual entity called ‘Satan’. Both theories in effect try to show that something may be (whether as being ‘merely mental’ or as being in ‘the realm of subsistents’), but not be in existence. But as we can say ‘round squares do not exist’, and ‘real nonentities do no exist’, this sort of interpretation of negative existentials is bound to fill either the realm of subsistents or the realm of ideas with walking self-contradictions. (Ryle 1932b, 144; 2009b, 45)

Furthermore, the Meinongian option violates the principle of ontological parsimony, for it would imply that, besides an apple (say), there would be ‘the existence of an apple’, ‘the subsistence of the existence of an apple’, and so on—which is hardly acceptable in light of Occam’s razor (Brentano 1925, 160; 1995, 292–293). Thus, we are left with the third, Brentanian option. The Brentanian option: quasi-ontological statements are aboutTC mental phenomena.—The key idea here is that of a complication of mental phenomena: (1) and (2) are aboutTC mental phenomena inasmuch as they express mental phenomena directed at other mental phenomena. To fully understand this idea, though, it is necessary to make a brief detour via Brentano’s logic course (Brentano Ms. EL 80). On Brentano’s view, it is the primary function of language to express (or arouse) mental phenomena. He calls categorematic any linguistic

5  Misleading Expressions: The Brentano-Ryle Connection 

109

expression which by itself expresses (or arouses) a mental phenomenon, where ‘by itself ’ means ‘without being supplemented by any further word or group of words’. He then calls name any categorematic expression that expresses (or arouses) an act of presentation, and contrasts names with speech (Rede), which is any categorematic expression that expresses another type of mental phenomena, namely an act of judgement or an act of interest.7 Now central to Brentano’s analysis of names is the distinction between what a name means (bedeutet) and what it refers to or names (nennt). What a name means depends on the ‘content’ of the corresponding presentation, whereas what it names depends on what thing in the world, if any, is picked out by means of this content. In other words, it depends on the object of the presentation.8 For the sake of illustration, consider the phrase ‘my cat’. Unlike ‘and’, ‘or’, ‘is’ and the likes, ‘my cat’ is a categorematic expression for it expresses by itself a mental phenomenon occurring in the mind of the speaker or aroused in the mind of the addressee. Unlike ‘my cat is sleeping on the mat’, which expresses an act of judging, ‘my cat’ is a name for it expresses an act of presentation. It means whatever is the content of the related presentation and names the animal which was chasing birds in the garden earlier in the morning and is presently sleeping on the mat in my living-room. On Brentano’s view, a similar distinction applies to pieces of discourse: the assertion ‘my cat is sleeping on the mat’ means whatever is the content of the related judging act and indicates (anzeigt) that which is accepted or rejected by this judging act, what is judged about. Those distinctions are captured in Fig. 5.1 and Fig. 5.2. Brentano has it that many errors in philosophical theories about language derive from a confusion between those three fundamental

Name

names the object of

expresses/arouses the related act of presentation Fig. 5.1  Brentano’s analysis of a name

means the content of

110 

A. Dewalque

indicates

the object of

Speech

means the content of

expresses/arouses the related act of judgement/interest

Fig. 5.2  Brentano’s analysis of a speech

functions, namely expressing (respectively, arousing), naming (respectively, indicating) and meaning.9 It is plain, for example, that ‘my cat’ does not name or mean the related act of presentation but expresses it, and does not express or mean the animal which is sleeping on the mat but names it. As it happens, this idea of a confusion between the aforementioned functions places Brentano’s rejection of the Lockean and Meinongian options under a new light. One way of interpreting Brentano’s criticism indeed is to say that the Lockean option rests on a confusion between what is named and what is expressed, while the Meinongian option rests on a confusion between what is named and what is meant. Plainly, the presentation of Satan is what is expressed by the name ‘Satan’, not what it refers to or what it names. Likewise, the non-existence of Satan is what is meant in (1)—it is the content of the related act of judgement—but it is not what is indicated or judged about. This explains why, on Brentano’s account, (1) is not aboutTC the non-existence of Satan, any more than it is aboutTC Satan or the non-existent object called ‘Satan’. When I utter (1) in a communicational context, I’m not communicating to my addressee, pace Meinong, that I’m performing a judging act to the effect that the non-existence of Satan obtains or that there is an individual called Satan of which it is true to say that it does not exist. Again, such analyses would violate Occam’s principle according to which entities should not be multiplied without necessity: If a content, e.g. the being or the non-being of Napoleon, could become an object, then it would also have to hold true of it that it either is or is not, and we would have to be able to say in the strict sense not only of Napoleon

5  Misleading Expressions: The Brentano-Ryle Connection 

111

but of the being of Napoleon as well, that it has being at one time and at another time it does not. (Brentano 1925, 161; 1995, 293)

Let me return to our opening question: what are (1) and (2) aboutTC? Brentano has it that, to answer this question, it is necessary to ‘peep beneath’ their surface grammar in order to identify the actual objects of the underlying acts of judgement, i.e. what must be the case for (1) and (2) to be true. Take Mill’s well-known example: ‘A centaur is a poetic fiction’. Plainly, there is no centaur in the world. Still, as we have seen, there must exist something else, which makes the sentence true. I quote Brentano again: The proposition, ‘A centaur is a poetic fiction’, does not imply […] that a centaur exists, rather it implies the opposite. But if it is true, it does imply that something else exists, namely a poetic fiction which combines part of a horse with part of a human body in a particular way. If there were no poetic fictions and if there were no centaurs imaginatively created by poets, the proposition would be false. (Brentano 1925, 61; 1995, 219)

Accordingly, ‘A centaur is a poetic fiction’ might be less conveniently but less misleadingly paraphrased by ‘There is a poetic fiction which conceives the upper parts of the human body joined to the body and legs of a horse’ or ‘Some people have imagined a creature which is half-man and half-horse’. Note that, since mental acts are not just ‘floating in the air’, they always presuppose someone, some individual or group of individuals, who is performing them—hence the second paraphrase. Therefore, ‘A centaur is a poetic fiction’ may be said to be aboutTC an act of imagining or, more pointedly, aboutTC the one who performs this act, namely an undetermined ‘imaginer’. If this analysis is correct, one should say that (1) ‘Satan is not a reality’ expresses a judgement which is not aboutTC a thing called ‘Satan’ but aboutTC another judgement, for example ‘The act of acknowledging an individual called “Satan” and being so and so, is incorrect’, or: (1*) ‘Those who acknowledge an individual called ‘Satan’ and being so and so, are in error’.

112 

A. Dewalque

Likewise, (2) ‘Unicorns do not exist’ might be paraphrased by saying that ‘The act of acknowledging creatures which resemble horses and have a horn, is incorrect’, or again: (2*) ‘Those who acknowledge creatures which resemble horses and have a horn, are in error’. Let me briefly comment on that. First, since what (1*) and (2*) are aboutSG is identical to what they are aboutTC, they are non-misleading sentences in the above-defined sense. Furthermore, if Brentano’s analysis is correct, their truth conditions are identical to those of (1) and (2), respectively, to the effect that they are successful paraphrases of (1) and (2). Next, we see how the whole process of analysing misleading expressions is driven by the thought of ontological parsimony underlying Occam’s razor. Clearly, (1*) and (2*) are assertions: they are pieces of discourse which express some judging acts. The key to Brentano’s account is that those judging acts are not intentionally directed at Satan or centaurs but at other mental acts: (1*) makes it manifest that (1) really is aboutTC the act of acknowledging an individual called ‘Satan’, etc., just like (2*) makes it manifest that (2) really is aboutTC the act of acknowledging creatures which resemble horses and have a horn. If one nonetheless wants to say that they are ‘about’ Satan and unicorns, one should say that they are about them only in modo obliquo, which is Brentano’s way of capturing the thought of mental phenomena nestled into other mental phenomena (see Dewalque 2014). Admittedly, this account implies that expressions about ficta rest upon a complication of mental phenomena. Yet, this complication is quite acceptable in Brentano’s eyes, for unlike the Lockean and the Meinongian options, it does not imply the existence of anything above and beyond mental phenomena or mentally active subjects. Ontologically speaking, mental acts or mentally active subjects are regular entities. They have nothing in common with weird entities like ‘ideas in the head’ or ‘non-existent objects’. So far I have briefly reconstructed (what I take to be) Brentano’s analysis of (1) and (2). What about Ryle? Regarding (1), he maintains that ‘the fact recorded would have been properly or less improperly recorded in

5  Misleading Expressions: The Brentano-Ryle Connection 

113

the statement “‘Satan’ is not a proper name” or ‘No one is called ‘Satan’” or “No one is both called ‘Satan’ and is infinitely malevolent, etc.” or perhaps “Some people believe that someone is both called ‘Satan’ and infinitely malevolent, but their belief is false” (Ryle 1932b, 149; my emphasis). The last paraphrase is very close to (1*). To be sure, it is not explicitly couched in terms of a complication of mental phenomena and I’m not sure Ryle would be happy with such a terminology. But I see no reason why his own analysis in 1932 should be constructed as incompatible with that of Brentano. After all, Brentano’s theory of mental phenomena has nothing to do with the ‘Lockean demonology’ Ryle rejects (Ryle 1932b, 161). On the contrary, as Ryle himself points out, mental phenomena in Brentano’s sense are nothing but ‘directly discernible manifestations of mental functioning’, the variety of which is in no way reducible to ideas in the head (Ryle 2009a, 175). Similarly, Ryle paraphrases (2) by saying that ‘Nothing has the property of resembling a horse and having a horn’, which on Brentano’s account should in turn be paraphrased in terms of (2*), for ‘nothing’ is not a categorematic expression and does not refer to anything, as Ryle himself recognises. Brentano’s analysis, therefore, does not seem to me to be incompatible with Ryle’s. If anything, it goes a step further by investigating the structure of the underlying mental phenomena, while Ryle rather sticks to case-by-case paraphrases and refrains from embarking on psychological descriptions (the question of which option is better is a question that I will not address here). Admittedly, this difference is not negligible. It probably reflects the fact that they have different views of philosophy— not to mention the fact that Ryle’s anti-Cartesianism, which will come to the fore in The Concept of Mind, will still increase the gap between them. Still, Ryle’s 1932a account of misleading expressions is, if not identical to, at least continuous with that of Brentano.

114 

4

A. Dewalque

 Moral About the Meaning A of ‘Meaning’

I have argued that Brentano and the early Ryle share a strikingly similar conception of philosophical analysis and misleading expressions. The backbone of this conception is the principle of ontological parsimony associated with Occam’s razor. I have illustrated this claim by comparing their analyses of statements about ficta. I suspect (although I cannot show it here) that a similar observation can be made for their analyses of quasi-­ Platonic statements (statements about universals) and modifying expressions. To conclude, I would like to draw a consequence of what precedes for a correct understanding of the notion of meaning. On Brentano’s view, as we have seen, what is called ‘the meaning of a name’ just is the content of the underlying act of presentation. Likewise, ‘the meaning of an assertion’ just is the content of the related act of judgement. Because Brentano has it that the content-object distinction must be preserved, the contents of those mental phenomena cannot become the true objects of further assertions. Although it sometimes looks as if we are referring to the meaning of an expression, we do not actually judge aboutTC it. In other words, meanings are not objects we can talk aboutTC. Whenever we talk aboutSG the meaning of a linguistic expression, what we really are talking aboutTC is the underlying mental phenomenon with such and such content. For example, saying that ‘the meaning of x is vague’ is a convenient, albeit misleading, way of saying that ‘there is an underlying mental phenomenon whose content is not sharply determined’. One of the main consequences of this view is that all questions of philosophy of language which presuppose that meanings are objects are wrongheaded. Worse, they are senseless. Treating meanings as if they were objects distinct from, and on a par with, physical things and mental episodes, is just another way of allowing oneself to be misled by language into multiplying entities without necessity. Ryle, of all people, had perfectly grasped this fundamental lesson from Brentano.10 The sentence ‘I have just grasped the meaning of x’, he writes, is systematically misleading in that it grammatically resembles

5  Misleading Expressions: The Brentano-Ryle Connection 

115

sentences such as ‘I have just met the village policeman’, whereas in fact there is no object corresponding to ‘the meaning of x’ (no more than there is an object corresponding to ‘Satan’ or ‘the present King of France’). Hence, enquiring into the status of the object allegedly referred to by ‘the meaning of x’ is absurd, since there simply is no such object. As Ryle puts it: “It is as pointless to discuss whether word-meanings (i.e. ‘concepts’ or ‘universals’) are subjective or objective, or whether sentence-meanings (i.e. ‘judgements” or “objectives’) are subjective or objective, as it would be to discuss whether the equator or the sky is subjective or objective. For the questions themselves are not about anything’ (Ryle 1932b, 163; 2009b, 65). On this approach, anti-Platonism about meaning does not imply that meaning is psychological or subjective. Rather, it implies that the question whether meaning is subjective or objective is senseless. Here, I suspect, Ryle is more ‘Brentanian’ than many philosophers of the Brentano School.

Notes 1. My own feeling is that Brentano’s position is much more continuous than suggested by the talk of a reist ‘turn’, but I will bracket this issue here. For a recent account of Brentano’s reism and related developments in the Prague School, see (Sauer 2017; Janoušek and Rollinger 2017). 2. Interestingly, Ryle is not the only analytic philosopher to defend that view. G.E. Moore makes approximately the same points in 1933–1934 (see Moore 2004, 165–171). 3. Ryle himself retrospectively talks of the ‘Occamizing zeal’ deployed in his early writings (Ryle 2009b, xx). 4. Compare (Brentano 1925, 305; 1995, 367): ‘If one wished to make a complete survey of entia rationis, one would have to go into the great variety of locutions which make words the subject and predicate of propositions which do not refer to real things in and of themselves’. 5. Ryle does not give any name to this fourth class, to which he devotes only a couple of lines (see Ryle 1932b, 165; 2009b, 61). For the sake of convenience, I use here the Brentanian label. 6. See e.g. (Brentano Ms. EL 80, 13.001[3]–13.002[7]): ‘Regarding names, the question arises as to what they mean. They mean: (1) not themselves;

116 

A. Dewalque

(2) not the act of presentation or the presentation; (3) not what is presented as presented; (4) and yet they do not seem to denote the things (a) for many names are not names of things, they are fictions, e.g. Jupiter, [and] (b) hoc animal and hic homo would not have a different meaning. They denote something presented, though not as presented but as that as what it is presented [als das, als was es vorgestellt wird]. This accommodates (a) and (b), for we are presented here with a thing, albeit through the mediation of various presentations’. 7. I won’t comment here on Brentano’s classification of mental phenomena. For a reconstruction, see (Dewalque 2018). 8. See (Brentano Ms. EL 80, 13.018[1]–13.018[5]): ‘The name designates [bezeichnet] in some way the content of a presentation as such, the immanent object. In some way, [it designates] what is presented by means of the content of a presentation. The former is the meaning of the name. The latter is what the name names [nennt]. We say about it that it has the name [es komme der Name ihm zu]. When it exists, it is an external object of presentation. One names through the mediation of meaning’. 9. Another source of errors lies in the fact of regarding syncategorematic expressions as categorematic ones, like when one takes at face value expressions such as ‘the truth of p’ of ‘the impossibility of A’ (see e.g. Brentano 1995, 322). I won’t address this kind of confusion here. 10. Thomasson rightly noted that the anti-Platonistic approach to meaning is a major commonality between Ryle and Brentano: ‘Ryle—like Brentano—takes it to be a systematic mistake to conceive of the procedure of conceptual analysis as involving a description of relations among Platonized meanings or concepts’ (Thomasson 2002, 129).

References Bourdeau, Michel. 2003. Ryle et la phénoménologie. Revue Internationale de Philosophie 223: 13–35. Brandl, Johannes L. 2002. Gilbert Ryle: A Mediator between Analytic Philosophy and Phenomenology. Southern Journal of Philosophy 40 (S1): 143–151. Brentano, Franz. n.d. Ms. EL 80. Logik. Unpublished Manuscript EL 80. Provisional Online Edition. Edited by Robin D. Rollinger. Franz Brentano Archiv Graz. http://gams.uni-graz.at/archive/objects/context:bag/methods/ sdef:Context/get?mode=logik.

5  Misleading Expressions: The Brentano-Ryle Connection 

117

———. 1924. Psychologie vom empirischen Standpunkt, ed. Oskar Kraus, vol. 1. Leipzig: Meiner. ———. 1925. Psychologie vom empirischen Standpunkt, ed. Oskar Kraus, vol. 2. Leipzig: Meiner. ———. 1995. Psychology from an Empirical Standpoint. Trans. Antos C. Rancurello, D. B. Terrell, and Linda L. McAlister. London: Routledge. Chase, James, and Jack Reynolds. 2017. Russell, Ryle and Phenomenology: An Alternative Parsing of the Ways. In Analytic Philosophy: An Interpretive History, ed. Aaron Preston, 52–69. New York: Routledge. Dewalque, Arnaud. 2014. Intentionnalité in obliquo. Bulletin d’Analyse Phénoménologique 10 (6): 40–84. ———. 2018. Natural Classes in Brentano’s Psychology. Brentano Studien 16: 107–138. Janoušek, Hynek, and Robin D.  Rollinger. 2017. The Prague School. In The Routledge Handbook of Brentano and the Brentano School, ed. Uriah Kriegel, 313–322. New York and London: Routledge. Mill, John Stuart. 1974. A System of Logic, Ratiocinative and Inductive. In The Collected Works of John Stuart Mill, ed. John M. Robson, vol. VII. Toronto-­ London: University of Toronto Press – Routledge and Kegan Paul. Moore, George Edward. 2004. Lectures on Philosophy. London and New York: Routledge. Morran, Dermot. 2014. Analytic Philosophy and Continental Philosophy: Four Confrontations. In Phenomenology: Responses and Developments, ed. Leonard Lawlor, 235–266. London and New York: Routledge. Russell, Bertrand. 1905. On Denoting. Mind (n.s.) 14: 479–493. ———. 1973. Essays in Analysis. New York: George Braziller. Ryle, Gilbert. 1932a. Phenomenology. Proceedings of the Aristotelian Society. Supplementary Volumes 11: 68–83. ———. 1932b. Systematically Misleading Expressions. Proceedings of the Aristotelian Society (New Series) 32: 139–170. ———. 1933. About. Analysis 1 (1): 10–12. ———. 2009a. Collected Papers. 1: Critical Essays. London and New  York: Routledge. ———. 2009b. Collected Papers. 2: Collected Essays. London and New  York: Routledge. Sauer, Werner. 2017. Brentano’s Reism. In The Routledge Handbook of Brentano and the Brentano School, ed. Uriah Kriegel, 133–143. New York and London: Routledge.

118 

A. Dewalque

Schuhmann, Karl. 1977. Husserl-Chronik. Denk- Und Lebensweg Edmund Husserls. Husserliana Dokumente 1. Dordrecht: Springer. Thomasson, Amie L. 2002. Phenomenology and the Development of Analytic Philosophy. Southern Journal of Philosophy 40: 115–142. ———. 2007. Conceptual Analysis in Phenomenology and Ordinary Language Philosophy. In The Analytic Turn, ed. Michael Beaney, 270–84. London: Routledge. Vrahimis, Andreas. 2013. Encounters Between Analytic and Continental Philosophy. Basingstoke: Palgrave Macmillan.

6 Sign and Language in Anton Marty: Before and after Brentano Hélène Leblanc

Before the genetic queries of On the Origin of Language (1875), long before the Investigations towards the Foundation of General Grammar and Philosophy of Language (1908), before becoming Brentano’s ‘Minister for Linguistic Affairs’—before all that, Anton Marty was a young seminary student in Mainz. There, in 1867, he wrote a dissertation on “Saint Thomas’s Doctrine of Abstraction of Supersensory Ideas from Sensory Images, with an Exposition and Critique of Other Theories of Knowledge”.1 The present paper is an analysis of this early work, the so-called Preisschrift. The relevance of this almost entirely unexplored material2 lies in the fact that in it Marty was already developing the framework of his linguistic thought, just before his meeting with Brentano in the winter semester of 1868–1869 at Würzburg (Fréchette and Taieb 2017, 2). For this reason, I shall not consider the main subject of this text, namely, abstraction (on this see Fréchette 2017, 173–176) and the origin of ideas,

H. Leblanc (*) Institut des Civilisations, Arts et Lettres (INCAL), UCLouvain, Louvain-la-Neuve, Belgium © The Author(s) 2021 A. Dewalque et al. (eds.), Philosophy of Language in the Brentano School, History of Analytic Philosophy, https://doi.org/10.1007/978-3-030-52211-7_6

119

120 

H. Leblanc

but the linguistic considerations that the young Marty developed, which I shall compare with the philosophy of language he elaborated later. From a historical point of view, the descriptions in this early work of the notion of sign and of the relation between words, ideas, and things provide some help in understanding the background of the School of Brentano, especially with regard to those associated with it who received a theological-philosophical education. From a doctrinal point of view, I hope to show how Marty’s originality is to be assessed with respect to the scholastic tradition, on the one hand, and within the School of Brentano, on the other. Finally, a careful study of the Preisschrift will suggest a new description of the ‘functionalist’ or ‘instrumentalist’ label, as applied to Marty’s later philosophy of language. I shall begin by presenting Marty’s critique of Louis de Bonald’s traditionalism and his claim that language is a necessary condition for thought. Next, I shall consider the young Marty’s conception of what a sign is, since language is a system of signs. I shall stress the traditional aspects of this conception, especially as compared to scholastic semiotic theories. Finally, the analysis of the notion of sign in Marty’s dissertation will be compared to the semantic and semiotic theories developed in his later works, in particular, the two broad and narrow senses of ‘meaning’ (Bedeutung), and the definition of ‘being a sign’ (Zeichensein). In doing so, I shall take into account Marty’s Austrian context.

1

The Destruction of the Cult of the Word

The main analysis of language in the Preisschrift occurs—and this is already a distinctive feature of Marty’s style—not in the descriptive section on Aquinas’s abstraction, but in the second, critical section, in which he aims to refute competing theories. To the question of the origin of our ideas, says Marty, there are four main answers that can be given (Marty 1867, 2): a) Ideas come from the senses. b) Ideas are innate. c) God gives us ideas by immediate contact. d) God gives us ideas through language.

6  Sign and Language in Anton Marty: Before and after Brentano 

121

In relation to these theories, Aquinas is the middle way: the subject and the object of knowledge work together by virtue of God’s cooperation. Thesis (d) was maintained by Louis de Bonald (1754–1840), a leading author of the traditionalist school who was important enough at the time that Marty dedicated a whole section to him (Marty 1867, 265–292). Bonald claims that language is not a human invention but a divine gift, and that it is the necessary condition of thinking (see e.g. Bonald 1800, 47–48; and the appendix to chapter 2). Marty strongly opposes this position, and follows instead the ‘true realism’ that he attributes to Aquinas (Marty 1867, 8), according to which abstraction is not the product of language, even if language is useful for it. Marty criticises Bonald for being a kind of mystic, and lacking in rigour. More precisely, Marty argues that if all words and ideas were founded on an act of belief, any hope of scientificity would be ruined (Marty 1867, 269). Moreover, Bonald’s thesis is based on a misunderstanding of the relation between words and ideas. According to Marty, thought and language must be separated, but also words should not be considered to be the source of ideas. Rather, ideas are prior to words, to the point that ‘the word is for the one who does not already carry in herself the idea, an empty sign, unable ever to produce by itself any understanding in the listener’ (Marty 1867, 277).3 Marty is clearly referring not just to Aquinas, but to Augustine, who, as we shall see, plays a paramount role in Marty’s account of language and sign. Augustine wrote in his treatise De magistro (XI.36) that hearing a word is not enough for it to be understood: the word can be understood only when the thing is understood. Marty (1867, 280) proves this correct by emphasising the arbitrary nature of linguistic signs. The word is like clothing for the idea, and in this sense, it is arbitrary, accessory, and artificial. Thus, the listener must first know not only the idea, but also the conventional relation between this idea and the word by which it is signified. Here Marty appeals explicitly to Augustine to claim that we can learn either from the things themselves, or by means of signs as a kind of stopgap (Lückenbüsser) when things are not accessible to our senses (Marty 1867, 277–278; see Augustine, De magistro X). Even in the case of knowledge through signs,

122 

H. Leblanc

one must have a prior idea of the thing signified. The priority of the thing is even more obvious in the case of conventional signs, since these are arbitrary, unlike natural signs. This is why, in the absence of the idea, the word would be the most defective amongst all the signs. Thus, linguistic signs are necessary for thought to be communicated to other beings, but not for thinking per se. Aquinas’s verbum mentis is a purely intellectual content, independent of and prior to the verbum oris by which it is signified. Bonald misunderstands Aquinas’s concept of the ‘mental word’ by turning it into a silent language, bound to imagination. Even though Bonald does not deny that thought precedes language, he definitely claims that ‘we cannot think without speaking within ourselves, that is, without tying some words to our thoughts’ and that ‘the intelligent being thinks her word before speaking her thought’ (1800, 224). Language is the necessary condition for thinking in this imaginative sense. By contrast, for a Thomist like the young Marty, this is simply misunderstanding the verbum mentis as a silent verbum oris. This does not preclude, however, that language is useful for the intellectual activity (Marty 1867, 283). First, it is useful for an individual as an image: a word is an ‘abbreviation’ (Abkürzung) of the corresponding sensible idea. It simplifies it and makes it easy to remember. Second, it is useful for communication as a sign: words are the means by which ideas are exchanged among speakers. And third, it is an aid to teaching. This view anticipates Marty’s later position on the relation between thought and language, which he developed in particular in “Über subjectlose Sätze und das Verhältnis der Grammatik zu Logik und Psychologie” (1884). Marty will advocate the relative independence of grammar from logic, against Heymann Steinthal who, at the opposite pole from Bonaldism, was a defender of the total separation of each discipline. Coming back to the Preisschrift, it is worth stressing that Marty’s critique of traditionalism has a strong Augustinian flavour. Marty argues for a restricted function of language, understood as a tool that is useful for reason, but not necessarily to its development. This point is particularly important, since Marty’s later conception of language has been characterised as ‘functionalist’ (see, for instance, Cesalli and Majolino 2014; as well as Cesalli et al. 2018; although the label has been applied to Marty

6  Sign and Language in Anton Marty: Before and after Brentano 

123

since Karl Bühler and Edmund Husserl). However, the term ‘functionalist’ has a more complex sense for the later Marty than it seems to in the Preisschrift. In the 1908 Untersuchungen, linguistic meaning is understood in terms of action and intention: Voluntary speech is a special kind of action whose ultimate aim is to trigger certain psychical phenomena in other beings. (Marty 1908, 284, transl. Cesalli 2013, 150)

Language must be addressed from the point of view of its users: the speaker’s goal is to bring about mental phenomena in the hearer. In this respect, Marty opposes most of the linguistic theories that were prevalent in his time, which gave precedence to the content of language, whether as some kind of hypostatised meaning (Bolzano, Frege, Husserl), or as the idea ‘in the mind’ of the speaker (Aquinas and most medieval thinkers, as well as most philosophers before the nineteenth century). Since Marty gives primacy to the functional goals of language, it will be conceived of as a ‘tool’ (instrument) that is used and optimised by its speakers. This idea, which can already be seen in Marty’s thesis “Über den Ursprung der Sprache” (1875), has been described as an ‘empirico-teleological’ account of the origin of language, which is formed ‘purposively but without planning’ (absichtlich aber planlos).4 The description of language as a tool continues in the Untersuchungen: Language is an organ, which, like any other tool, must be understood from its purpose and the task it is charged with. (Marty 1908, 53)

Thus, Marty’s theory of language is correctly defined as ‘functionalist’ or ‘instrumentalist’. Moreover, as we shall see, the production of a mental act in the hearer is not only the function of a sign but also its very meaning. On this basis, Marty has been portrayed as an ancestor of the main contemporary trend in philosophy of language, that is pragmatics, and in particular as a precursor of Paul Grice (Liedtke 1990; Cesalli 2008; Cesalli 2013; Cesalli 2015; Longworth 2017; Recanati 2019). I shall argue in the next section that the term ‘functionalist’ assumes an additional sense, which we see emerging in the Preisschrift, linked both to the

124 

H. Leblanc

reception of Augustine’s theory of signs and to Marty’s scholastic background.

2

The Semiotics of the Young Marty

In the section dedicated to language as a teaching aid, Marty gives a definition of the concept of sign: Every sign, and hence also the word, is—insofar as it is the manifestation of thought—a means of teaching. In the broadest sense, the sign is all that can evoke an idea in the mind; in the proper sense, however, it is only what gives rise to an idea not of itself but of something else. The sound of the voice arouses not an idea of itself but of the thing expressed. (Marty 1867, 285)

Beyond the critique of Bonaldism, which I shall now set aside, a full understanding of this definition requires the consideration of three philosophical contexts: (1) the reference to Augustine and the idea of the sign as a means of teaching; (2) the discussion of the two senses of ‘sign’, which was crucial in late scholasticism; (3) the discussion, in the Austrian context and in Marty’s later philosophy, of the equivocity of the sign. First, the claim that teaching is the goal of any sign, and more specifically of any word, is an Augustinian idea that can be found in the De magistro (I.1), which is quoted by Marty. However, Augustine’s definition of sign is to be found in the De doctrina christiana (II.1.1): A sign is a thing which, besides the form that it presents to the sense, evokes something different in the mind. (Augustine 1962, 32)

This definition quickly became standard in the main domains in which scholastic semiotic theories were developed: theology, grammar, and logic. It stresses the sensible nature of the sign and the fact that the sign is different from the thing signified. By contrast, Marty gives not one but two definitions of sign. The first (and broader) definition encompasses ‘all that can evoke an idea in the

6  Sign and Language in Anton Marty: Before and after Brentano 

125

mind’. In this sense, an idea can be the sign of another idea, and a thing, or a sensation, can be the sign of an idea, since things and sensations are able quite literally to ‘provoke’ ideas. Marty is not explicit about the implications of this broad definition, but it is fairly clear what he has in mind, since his formulation echoes medieval developments of the Augustinian definition. Such a reformulation aims at broadening the scope of the definition so as to encompass not just linguistic signs but also non-sensible processes such as angelic communication, the Eucharistic Sacrament, and—with major consequences in philosophy— mental presentation (for the sake of brevity, let us mention just Marmo 2010, 16–22; and Meier-Oeser 2011). Such a broadened version of Augustine’s definition has medieval roots that can be traced back to, among others, Aquinas himself, who gives the following definition: [The sign is] something manifest for us from which we are led, almost by the hand, to the knowledge of something hidden. (Aquinas 1947, 13)

However—and this is my second point—Marty’s broad definition of the sign goes further, since it does not posit any dichotomy between what is manifest and what is hidden, unlike Aquinas’s definition, notwithstanding its extension of signs beyond the sensible realm. In this respect, it is closer to a later version of a broad definition of sign, one that reached a climax in the early seventeenth century with the late scholastic debate on the ‘formal sign’ (signum formale), which was defined as ‘what represents something to a knowing power’ (Conimbricenses 2001, 39). This definition aims at explaining mental presentation by the notion of sign: intelligible species and concepts are called ‘formal signs’, while words (standing for ideas), smoke (indicating fire), and footprints (traces of the animal that left them) are ‘instrumental signs’, matching the old Augustinian definition (on this debate see Meier-Oeser 1997, 171–272; and Leblanc [in press]). Notwithstanding the various positions in this debate, all agree that within the narrow definition, what counts more than the sensible aspect of the sign is the difference between the knowledge of the sign and the knowledge of the thing signified. Prior knowledge thus becomes the criterion to distinguish between instrumental and formal signs. In the case

126 

H. Leblanc

of instrumental signs, there is a dual knowledge (one must perceive the smoke so as to have the idea of fire), whereas in the case of formal signs, the sign is a species intelligibilis, or a concept that does not need to be known per se to convey the knowledge of the thing) (see Conimbricenses, 59 and 78–79). In this respect, Marty’s narrow definition, like the one endorsed by many of his scholastic predecessors, does not mention sense perception but instead emphasises the twofold nature of signs. This is a further clue confirming that what Marty has in mind is the late scholastic formulation of the ‘equivocity of the sign’. This would be consistent with the scholastic background of a young seminarian at the end of the nineteenth century. Indeed, the whole of his Preisschrift shows a thorough knowledge of late scholastic authors, like Cajetan (alias Tommaso de Vio), Francesco Silvestri da Ferrara, and Francisco Suárez. Direct influence does not really matter here: even if Marty did not know the specific features of the debate, the dual definition had been transmitted and taught for centuries according to the late scholastic formulation. I shall here make two remarks, which will lead to the third and last context of Marty’s definitions of sign. First, the position adopted by the young Marty is consistent with his later rejection of mental presentations as signs. As is well known, Marty later describes the relation between ideas and things as a relation of assimilation, which after 1906 he calls ideelle Verähnlichung (see for instance Marty 1908, 421), referring to another scholastic debate (Cesalli 2017, 67–72; see also Cesalli and Taieb 2012; and Majolino 2017). Now, the concept of ‘mental assimilation’ is foreshadowed in the Preisschrift, where Marty translates similitudo or repraesentatio as Ähnlichkeit, Nachbildlichkeit, or Abbildlichkeit, and he translates assimilatio as Verähnlichung (Marty 1867, 25, 35, 232). Marty’s later conception of intentionality and ideas should therefore be reassessed in the light of the Thomist passages of the Preisschrift related to the species intelligibilis and the verbum mentis. Such a task would deserve a study on its own, but for present purposes I suggest only that the young Marty, in spite of his dual definition of sign, already limits the vocabulary of image and resemblance to his account of mental presentation, whereas ‘sign’ is employed to refer to linguistic communication.

6  Sign and Language in Anton Marty: Before and after Brentano 

127

This lexical distribution is consistent with the fact that in the Preisschrift Marty quotes the two definitions along the lines of the prevalent view in late scholasticism in order to focus on language, which matches the second, narrower, definition. Marty uses it against Bonald to describe the linguistic sign, with a focus on the difference between the sign and what it signifies—namely, ideas, which are always prior to words. Apart from the text quoted above, he makes no further reference to the dual definition or to a broad notion of sign. In this respect, Marty’s early thought on signs can be called ‘instrumental’ precisely as the late Scholastics understood the term: These are called instrumental signs either because we signify to others our concepts through them as if they were instruments; or because, as it is necessary for the craftsman, as long as he stirs the matter with an instrument, to move the instrument, so it is necessary for the powers able to know a certain thing through this kind of sign, that they perceive these signs. (Fonseca 1614, 14)

In the case of words, both characterisations fit: through words we signify our concepts, which are independent of them, and it is necessary to perceive the spoken sound or to see the graphic mark in order to grasp the meaning. We have here a traditional sense of ‘instrumental’, but it opens the way to an understanding of signs from the standpoint of their use, that is, their function. My second remark emphasises the connection of the dual definition in the Preisschrift with another debate on the equivocity of ‘sign’, one that this time occurred within the context of Austro-German philosophy.

3

Marty and His Contemporaries on Signs

As far as we know, the first philosopher in the Austro-German tradition to introduce the problem of the equivocity of ‘sign’ was Bernard Bolzano, who was also familiar with scholasticism. In § 52 of the Wissenchaftslehre, Bolzano introduces the problem of the equivocity  of ‘sign’ in order to determine whether a mental presentation (Vorstellung) is a sign.

128 

H. Leblanc

Instead of the word ‘image’, some use the word ‘sign’, and thus take presentations to be signs of things. This word is equivocal (zweideutig). (Bolzano 1837, I, 232)

The issue at stake is the traditional scholastic discussion about the semiotic status of mental presentations. On this question, Bolzano says clearly that ‘sign’ refers to either a conventional sign or a symptom, and: In neither of these senses is it possible to call the presentations the signs of things. (Bolzano 1837, I, 232)

Nevertheless, after resolving the issue of mental presentations, Bolzano comes back to the question of the equivocity of ‘sign’ in the third book of the Wissenschaftslehre, defining it as something used […] when we want to know, by its presentation that another presentation linked to the former is reactivated in a thinking being. (Bolzano 1837, III, 67)

Here the context is a discussion of linguistic signs (see Centrone and Künne 2011). The first note at the end of § 285 deepens the issue of the unity of the concept of sign by addressing the following question: how could both natural signs (such as smoke for fire) and conventional communicational signs (such as the word ‘God’—which does not even refer to any perceptible entity) possibly fall under the same concept? In Bolzano’s view, if there is any concept from which both indication and word derive, it is most likely not the general concept of sign. He (Bolzano 1837, I, 75–77) concludes that the two senses of ‘sign’ are autonomous, stating the difference between natural and communicational signs as follows: Natural signs fall within an epistemic framework. A natural sign is a property the perception of which lets us infer the existence of another property (on Bolzano’s inferential sign and its relation to Marty, see Kasabova 2006). Communicational signs belong to pragmatics. A conventional sign is not a property but an object, which is perceived by the listener as having

6  Sign and Language in Anton Marty: Before and after Brentano 

129

been produced by the intention of the speaker to provoke in her a presentation. The narrow version of the issue of equivocity, which draws a contrast between indications and expressions, reappears famously in 1900. Openly following in Bolzano’s footsteps (see Majolino 2010, 2013), Husserl opens his first Logical Investigation with this issue: § 1. An ambiguity in the term ‘sign’ (Zeichen) The terms ‘expression’ (Ausdruck) and ‘sign’ (Zeichen) are often treated as synonyms, but it will not be amiss to point out that they do not always coincide in application in common usage. (Husserl 1913–1921, II, I, 23, transl. Findlay 2001, 183)

Zeichen improperly refers to both expression and indication. Properly speaking, the term should be applied only to indication, to the extent that: Signs in the sense of indications [Anzeichen] (notes, marks, etc.) do not express anything, unless they happen to fulfil a significant [Bedeutungsfunktion] as well as an indicative function [der Funktion des Anzeigens]. (Husserl 1913–1921, II, I, 23)

Husserl strictly limits the meaning of ‘sign’. Mental presentations are definitely not signs, nor, properly speaking, linguistic means. Signs are only things indicating other things on a single level, such as the flag for a nation, and marks aiding memory, like a knot in a handkerchief. Even in the case of an abstract referent (a nation), the indication is an existing object or state of affairs standing for another existing (although not immediately perceptible) object or state of affairs. By contrast, expressions or meaningful signs belong to the level of meanings, and, in this, they radically differ from indications. Many Austro-German authors wrote about this ambiguity in the word ‘sign’, contrasting natural (inferential) and linguistic (conventional) signs. Following in Bolzano’s footsteps, Robert Zimmermann (1867, 233–239) and Richard Grätschenberg (1901, 74–85), were sympathetic to the thesis of equivocity  (see Bacigalupo 2012, 78). However, as Majolino (2010,

130 

H. Leblanc

79–81) points out, within the School of Brentano the issue moves from the level of presentation to that of judgement: conventional signs do not rest on associations of presentations; they require judgement. This brings them closer to natural signs. For instance, Eduard Martinak (1901, 57–62) claims, contra Alois Höfler (1897), that conventional signs do not require the activity of the association of ideas, for the knowledge of conventions appears to be sufficient. Thus, the same inferential relation operates in natural (‘real’) and conventional (‘final’) signs alike (Martinak 1901, 9–13). On the other side, some authors reverted to a broad notion of sign merging the psychological and ontological levels. Kazimierz Twardowski is the most striking example of this position (see Fréchette 2012). The scope of this paper does not allow discussing each of these authors, but let us say at least that in many cases, Brentanian theories of signs seem to prefer moderate versions of equivocity, such that ‘sign’ has two different but related meanings. In this sense, Husserl is an exception, favouring a strong version of equivocity. The later Marty, by contrast, is a good example of a moderate equivocity, albeit in an original formulation. Indeed, in the Untersuchungen Marty formulates, instead of two definitions of ‘sign’, two ways of ‘being a sign’: § 58. When considering the meaning of our linguistic expressions, it is necessary, first of all, to understand whether the very term ‘to signify’ is univocal or whether it is used in various senses […] [A]re the terms ‘to signify’, ‘to express’, ‘to communicate’ used in various meanings and for various modes of ‘being a sign’, and if so, what various meanings could ‘being a sign’ then have? (Marty 1908, 280)

It is no longer the issue of the putative semiotic status of mental presentations that is at stake. The formulation deals instead with the division between natural and conventional signs. Unlike Bolzano’s account, Marty’s does not depend on the assumption that ‘sign’ is an equivocal word, but assumes rather that there are two distinct modes of ‘being a sign’. In both of these modes, the issue of equivocity is secondary to identifying the mode.

6  Sign and Language in Anton Marty: Before and after Brentano 

131

Marty distinguishes non-deliberate externalisation of our psychic life (a scream) and deliberate externalisations (human language). In the first case, ‘being a sign’ is univocal: it invariably means something that makes known or announces something else. In the second case, two intentions should be distinguished: first, the deliberate use of a sign aims at influencing the mental life of other persons; and second, the sign announces our mental life (the mental phenomenon indicated by the words). The pragmatic function (1) is accomplished through the expressive one (2). The whole process may be visualised as follows (Fig. 6.1). Marty calls the second intention Kundgabe (indication) and the first Bedeutung, which corresponds to ‘meaning’ in a broad sense. The broad sense of ‘meaning’ contrasts with the narrow one, namely, the content that the speaker aims at expressing. (Marty 1908, 22, 280–287, 490–501) The whole scheme developed by Marty has been thoroughly described (starting with Mulligan 1990, 13–20). As I pointed out above at the end of the first section, it leads to Marty being labelled ‘functionalist’ or ‘instrumentalist’ (see for instance Cesalli 2015, 255–257). I would like to suggest in the next section that a comprehensive view of Marty’s philosophy of signs and language, including the early text of the Preisschrift, could help us get a better understanding of such labels.

'being a sign' Zeichensein

non-deliberate externalization = announces or makes something known (INFERENCE)

second intention: announces our mental life

deliberate externalisation first intention: influences the mental life of others

A

B

Fig. 6.1  Marty’s analysis of being a sign

C

132 

4

H. Leblanc

What Does ‘Instrumentalist’ Mean?

Like most labels—and like many words—‘instrumentalist’ has many senses. With respect to Marty, I shall distinguish at least three. a) ‘Instrumentalist’ is a synonym of ‘functionalist’ (see Cesalli 2008, 18). Both terms at least share interchangeable uses. Saying that Marty is an ‘instrumentalist’ in this sense means that he conceives of language as a tool such that linguistic signification is the production of a mental act in the hearer: speaking is a kind of action. We have already seen that this leads us to see Marty as a forerunner of contemporary pragmatics. This strikes me as correct, but I would add that it is also useful to emphasise that the diagram above suggests a strong heterogeneity, which differs from the equivocity discussed earlier. Indeed, to write, as I did above, that Marty transforms the issue of the ‘equivocity of sign’ in describing ‘two modes of “being a sign”’ is no mere rhetorical turn. This goes beyond the fact that the formulation— ‘being a sign’ rather than ‘sign’—once again highlights that to speak is to act. Maintaining that there are two ways of being a sign is not tantamount to claiming that the word ‘sign’ is equivocal: rather, these are modes, and the second mode includes the first as a part, since there is always a moment of announcement within the deliberate externalisation. However, the second mode is not a species of the first (which would then be the genus). While the B part of the diagram is about a semiotic division, the C part describes a process in terms of intentions. It is here that the true equivocity lies, that is, the heterogeneity: the deliberate externalisation (such as linguistic activity) brings together two intentions that are radically different, and the first and prior one (to influence the mental life of others) is extrinsic to the original meaning of ‘being a sign’ (to announce or make something known). However, the two intentions are linked in the process, since the second intention is the means by which one reaches the first: by announcing our mental life, we influence the mental life of others.

6  Sign and Language in Anton Marty: Before and after Brentano 

133

Thus, there are two modes of being a sign. In the first mode, ‘being a sign’ is univocal: it means only ‘what announces something’. In the second mode, it is equivocal: it means both ‘what announces our mental life’, which corresponds to the second intention of the speaker, and ‘what influences the mental life of the hearer’, which corresponds to the first intention of the speaker. Within the second mode, there are two heterogeneous senses of ‘being a sign’. This gives a strong accent, I think, to Marty’s version of equivocity, though it is sometimes considered somehow weaker as compared with Husserl’s more radical one (see Majolino 2010, 85). b) I have suggested that what unifies these two heterogeneous parts of deliberate externalisation is the process: the fact that a speaker intends to influence the mental life of others through the announcement of her mental life. It should be added that both of these acts are intentional. This is the second sense of ‘instrumentalist’. While the function emphasises the action, this sense of ‘instrument’ emphasises deliberation. Before describing Marty’s pragmatic semantics, Cesalli (2015, 248) identifies this sense as already at work in the medieval philosopher Roger Bacon. One could also note that the seminal Augustinian division between signa naturalia and signa data (Augustine 1962, 7, 32–33) also surfaces in Marty’s scheme, but because of the role played by inference the division is somewhat blurred and turns out to be more complex. This leads us finally to reconsider Marty’s mature description of non-­ deliberate externalisation in the light of the Preisschrift. The first meaning of ‘being a sign’ in the Untersuchungen is very close to the broad definition presented by the young Marty, but the change in the formulation is significant. As I hope to have shown, the formulation ‘all that can evoke an idea in the mind’, as opposed to a narrower definition stressing the difference between the sign and what it signifies, referred to the late scholastic debate about the formal sign. Similarly, ‘makes something known’ seems to refer to the radical position, which immediately preceded this debate. In fact, it echoes not only the tradition of Boethius’s commentary on Aristotle’s Peri hermeneias 16b, but

134 

H. Leblanc

also the influence of this tradition on the late Scholastics, in particular on the nominalist milieu of the school of Paris, in which signification was defined very broadly as ‘facere cognoscere’ (see Meier-Oeser 1997, 117). The return of a narrow Augustinian conception of sign, as the outcome of the debate on formal signs, was turned precisely against such excesses. Thus, in the Untersuchungen Marty takes the broadest definition of ‘sign’ and transforms it by adding the notion of ‘announcement’, which includes an inferential structure. The notion of announcement remains only in the description of the second intention of the deliberate externalisation. By contrast, the description of non-deliberate externalisation remains quite vague: ‘etwas kund und zu erkennen geben’ (Marty 1908, 280). In this sense, Marty brings together the opposition between mental presentation (the formal sign) and linguistic signification (the instrumental sign) which appeared in the Preisschrift, and the division between Kundgabe and Bedeutung, which is specific to the deliberate externalisation in the Untersuchungen. c) Finally, ‘instrumentalist’ has also its scholastic sense. The ‘instrumental’ sign must be itself known so as to lead to the knowledge of something else. This sense appears plainly in the Preisschrift and it is necessary to have it in mind so as to highlight a point little considered in the scheme of the Untersuchungen. Let us consider the first intention of the deliberate externalisation. ‘What gives rise to an idea not of itself but of something else’ and ‘what influences the mental life of others’ seem to have little in common. Yet in the first mode of being a sign there is no specific object related to the inference to be made, or to the knowledge to be acquired. The idea is simply to ‘make something known’, whatever it is. By contrast, in deliberate externalisation, what is to be known is ‘our mental life’, namely, something that in all circumstances will remain inaccessible to others (since we are bound to our bodies). According to this third, very Augustinian sense of ‘instrumental’, deliberate externalisation draws its complex structure from human corporeity. It stresses the absolute necessity of passing through language as a means, which is strictly different from thought, in order to accomplish human communication. The Preisschrift is not so far away.

6  Sign and Language in Anton Marty: Before and after Brentano 

5

135

Austro-German Scholasticism

The previous section identified three senses of ‘instrumentalism’ in Marty: in the first, it is a synonym of ‘functionalism’ and emphasises the action, the second stresses the intention and the third highlights the role of corporeity in human communication. Taking the Preisschrift into account played an important role in distinguishing these meanings, but also in providing a more accurate description of the traditional divisions and references at stake in Marty’s very complex notion of ‘being a sign’. For this reason, I shall conclude by stressing first the value of this early text, which calls into question in particular the view that before the Untersuchungen, Marty had an unclear notion of sign (Majolino 2013, 21–22). More generally, it contributes to the reassessment of Marty’s originality with respect to Brentano. Secondly, it is important to take into consideration the scholastic background of the Austro-German tradition. This is obvious with regard to authors like Bolzano, Marty and, of course, Brentano, who were actually taught in a scholastic milieu. Studies in this regard have proven their worth since Martin Grabmann’s claim (1926, 104–106) that the interdisciplinarity of medieval logic (which unites semantics, ontology, and philosophy of mind), is better understood if considered as a Sprachphilosophie, the model of which is provided by the Austro-German philosophy of the beginning of the twentieth century (see also Libera 1986 and Cesalli 2010). With different degrees of attention to direct influence, such studies aim at highlighting the conceptual relations between medieval and Austro-German philosophy. In addition, many of them present the underlying idea that both traditions prefigure the analytic style of philosophy. What is interesting here, from a metahistorical perspective, is the philosophical relevance of late scholasticism. As we have tried to show, scholasticism did not end with the Middle Ages, but was a continuous tradition that transmitted its concepts—and a certain way of doing philosophy—up to the beginning of the twentieth century. It also invites us to question the meaning that ought to be attributed to a current, a tradition, or even a school. This will undoubtedly make sense for anyone interested in the complex nature of a movement such as the School of Brentano.5

136 

H. Leblanc

Notes 1. “Die Lehre des hl. Thomas über die Abstraction der übersinnlichen Ideen aus den sinnlichen Bildern. Summa theol. I q. 85 art. 1–3. nebst: Darstellung und Kritik der übrigen Erkenntnißtheorien” (title transl. in Rollinger 2014). 2. Although it was briefly described by Oskar Kraus (1916, 3), the Preisschrift remained unknown until its rediscovery in the Masaryk Archives. These archives contain a huge number of unpublished writings of remarkable value, but because they have not yet been systematically catalogued they have been little studied. See however O. Funke’s introduction to the first volume of the Nachgelassene Schriften (Marty 1940); see also Raynaud (1982, 290). The Preisschrift has been transcribed by Giuliano Bacigalupo as a part of the SNF project “Meaning and Intentionality in Anton Marty: At the Crossroads of the Philosophy of Language and Mind”, 2014–2017. Laurent Cesalli, Guillaume Fréchette, Hélène Leblanc and Hamid Taieb are currently preparing an edition of the text. 3. Unless otherwise specified, all translations are mine. 4. The expression is frequently quoted, but it appears more often in the Untersuchungen, although in less compact versions. See for instance Marty (1908, 76), with reference to Marty (1875). 5. I am especially grateful to Laurent Cesalli, Guillaume Fréchette, Claudio Majolino, and Hamid Taieb, who provided very helpful comments on previous drafts of this paper.

References Aquinas, Thomas. 1947. In Scriptum Super Sententiis magistri Petri Lombardi, ed. Maria Fabianus Moos, vol. 4. Paris: Lethielleux. Augustine, S. 1962. In De doctrina christiana; De vera religione, ed. Joseph Martin. Turnhout: Brepols. Bacigalupo, Giuliano. 2012. Semiotica e semantica in Martinak e Meinong. In Paradigmi, ed. Giuliano Bacigalupo and Claudio Majolino, vol. 2, 77–91. Bolzano, Bernard. 1837. Wissenschaftslehre: Versuch einer ausführlichen und größtentheils neuen Darstellung der Logik mit steter Rücksicht auf deren bisherige Bearbeiter. Vol. 4. Sulzbach: Seidel. Bonald, Louis de. 1800. Essai analytique sur les lois naturelles de l’ordre social. Paris.

6  Sign and Language in Anton Marty: Before and after Brentano 

137

Centrone, Stefania, and Wolfgang Künne. 2011. Bolzanos Zeichentheorie: Eine Untersuchung zu § 285 der Wissenschaftslehre. Grazer Philosophische Studien 83: 171–198. Cesalli, Laurent. 2008. ‘Faire Sens’: La sémantique pragmatique d’Anton Marty. Revue de théologie et de philosophie 140: 13–29. ———. 2010. Medieval Logic as Sprachphilosophie. Bulletin de philosophie médiévale 52: 117–132. ———. 2013. Marty’s Intentionalist Theory of Meaning. In Themes from Brentano, ed. Denis Fisette and Guillaume Fréchette, 139–163. Amsterdam and New York: Rodopi. ———. 2015. Meaning in Action: Anton Marty’s Pragmatic Semantics. In Linguistic Content: New Essays on the History of Philosophy of Language, ed. Margaret Cameron and Robert Stainton, 245–265. Oxford: Oxford University Press. ———. 2017. Mental Similarity: Marty and the Pre-Brentanian Tradition. In Mind and Language: The Philosophy of Anton Marty, ed. Guillaume Fréchette and Hamid Taieb, 63–82. Berlin and New York: De Gruyter. Cesalli, Laurent, Janette Friedrich, and Claudio Majolino. 2018. Les actes de parole et la tradition austro-allemande. In De l’action du discours: Le concept de speech act au prisme de ses histoires, ed. Bruno Ambroise, 83–113. London: Iste. Cesalli, Laurent, and Claudio Majolino. 2014. Making Sense: On the Cluster signification-intentio in Medieval and ‘Austrian’ Philosophies. Methodos 14. Cesalli, Laurent, and Hamid Taieb. 2012. The Road to ideelle Verähnlichung: Anton Marty’s Conception of Intentionality in the Light of Its Brentanian Background. Quaestio 12: 171–232. Conimbricenses. 2001. Some Questions on Signs, Ed. and Trans. J.P.  Doyle. Milwaukee: Marquette University Press. Fonseca, Pedro da. 1614. Institutionum dialecticarum libri octo. Lyon: sumptibus Petri Rigaud. Libera, Alain de. 1986. La logique du Moyen Âge comme logique naturelle (Sprachlogik). In Sprachphilosophie in Antike und Mittelalter: Bochumer Kolloquium, 2.–4. Juni 1982, ed. Burkhard Mojsisch, 403–437. Amsterdam: Bochumer Studien zur Philosophie 3. Fréchette, Guillaume. 2012. Twardowski on Signs and Products. In Paradigmi, ed. Giuliano Bacigalupo and Claudio Majolino, vol. 2, 61–75.

138 

H. Leblanc

———. 2017. Marty on Abstraction. In Mind and Language: The Philosophy of Anton Marty, ed. Guillaume Fréchette and Hamid Taieb, 169–194. Berlin and New York: De Gruyter. Fréchette, Guillaume, and Hamid Taieb. 2017. Anton Marty: From Mind to Language. In Mind and Language: The Philosophy of Anton Marty, ed. Guillaume Fréchette and Hamid Taieb, 1–20. Berlin and New  York: De Gruyter. Grabmann, Martin. 1926. Die Entwicklung der mittelalterlichen Sprachlogik. In Mittelalterliches Geistesleben, Bd I, ed. Martin Grabmann, 104–146. Munich: M. Huber. Grätschenberger R. 1901. Grundzüge einer Psychologie des Zeichens. Regensburg: Mainz. Reprint in 1987, ed. A. Eschbach, Amsterdam / Philadelphia: John Benjamins. Höfler, Aloïs. 1897. Psychologie. Vienna: Tempski. Husserl, Edmund. 1913–1921. Logische Untersuchungen (Husserliana XVIII, XIXX/1 XIX/2), 2nd ed. Niemeyer: Halle. Trans. John N. Findlay as Logical Investigations. New York: Routledge, 2001. Kasabova, Anita. 2006. Bolzano’s Semiotic Method of Explication. History of Philosophy Quarterly 23 (1): 21–39. Kraus, Oskar. 1916. Marty’s Leben und Werke: Eine Skizze. In Marty, Anton. Gesammelte Schriften I.1, edited by Joseph Eisenmeier, Alfred Kastil and Oskar Kraus, 1–68. Halle: Max Niemeyer. Leblanc, Hélène. in press. Théories sémiotiques à l’Âge classique. Translatio signorum. Paris: Vrin. Liedtke, Frank. 1990. Meaning and Expression: Marty and Grice on Intentional Semantics. In Mind, Meaning and Metaphysics: The Philosophy and Theory of Language of Anton Marty, ed. Kevin Mulligan, 29–49. Dordrecht: Kluwer. Longworth, Guy. 2017. Grice and Marty on Expression. In Mind and Language: The Philosophy of Anton Marty, ed. Guillaume Fréchette and Hamid Taieb, 263–284. Berlin and New York: De Gruyter. Majolino, Claudio. 2010. Structure de l’indice et équivocité du signe: À l’origine du partage Anzeige/Ausdruck dans les Recherches logiques. Histoire, Épistémologie, Language 32 (2): 75–92. ———. 2013. Il trilemma bolzaniano: Prolegomeni al ‘problema del segno in fenomenologia’. Blityri 2 (2): 137–159.

6  Sign and Language in Anton Marty: Before and after Brentano 

139

———. 2017. Talking about Intentionality: Marty and the Language of ‘Ideal Similarity’. In Mind and Language. The Philosophy of Anton Marty, ed. Guillaume Fréchette and Hamid Taieb, 83–104. Berlin and New  York: De Gruyter. Marmo, Costantino. 2010. La semiotica del XIII secolo. Milan: Bompiani. Martinak, Eduard. 1901. Psychologische Untersuchungen zur Bedeutungslehre. Leipzig: J.A. Barth. Marty, Anton. 1867. Die Lehre des hl. Thomas über die Abstraction der übersinnlichen Ideen aus den sinnlichen Bildern. Summa theol. I q. 85 art. 1–3. nebst: Darstellung und Kritik der übrigen Erkenntnißtheorien. Preisaufgabe für 1867 von A. Marty alum. Rukopis 301 str. Ústřední Archiv Českolovenské Akademie Věd. Fond: Anton Marty. Signatura: IIIa inv. č. 19. ———. 1875. Über den Ursprung der Sprache. Würzburg, A.  Stuber. Trans. Robin D. Rollinger. In Rollinger, Robin D. 2010. Philosophy of Language and Other Matters in the Work of Anton Marty. Analysis and Translation: 133–234. Amsterdam: Rodopi. ———. 1884. Über subjectlose Sätze und das Verhältnis der Grammatik zu Logik und Psychologie I–III. Vierteljahresschrift für wissenschaftliche Philosophie 8: 56–94, 161–192, 292–340. Reprinted in Marty, Anton. 1918. Gesammelte Schriften, vol. II/1, ed. Joseph Eisenmeier, Alfred Kastil and Oskar Kraus, 1–35, 36–62, 62–101. Halle: Max Niemeyer. ———. 1908. Untersuchungen zur allgemeinen Grammatik und Sprachphilosophie. Halle: Max Niemeyer. ———. 1940. Nachgelassene Schriften: Aus ‘Untersuchungen zur Grundlegung der allgemeinen Grammatik und Sprachphilosophie’ I: Psyche und Sprachstruktur, edited with an introduction and comments by Otto Funke. Bern: A. Francke. Meier-Oeser, Stephan. 1997. Die Spur des Zeichens: Das Zeichen und seine Funktion in der Philosophie des Mittelalters und der früher Neuzeit. Berlin and New York: De Gruyter. ———. 2011. Medieval Semiotics. In The Stanford Encyclopedia of Philosophy, edited by Eduard N.  Zalta, https://plato.stanford.edu/archives/sum2011/ entries/semiotics-medieval. Mulligan, Kevin. 1990. Marty’s Philosophical Grammar. In Mind, Meaning and Metaphysics: The Philosophy and Theory of Language of Anton Marty, ed. Kevin Mulligan, 11–27. Dordrecht: Kluwer. Raynaud, Savina. 1982. Anton Marty, filosofo del linguaggio. Rome: La Goliardica Editrice.

140 

H. Leblanc

Recanati, François. 2019. Natural Meaning and the Foundations of Human Communication: A Comparison Between Marty and Grice. In Anton Marty and Contemporary Philosophy, ed. Giuliano Bacigalupo and Hélène Leblanc, 13–31. Cham: Palgrave Macmillan. Rollinger, Robin D. 2014. Anton Marty. In The Stanford Encyclopedia of Philosophy, ed. Eduard N. Zalta, https://plato.stanford.edu/archives/spr2014/ entries/marty/. Zimmerman, Robert. 1867. Philosophische Propaedeutik. Vienna: Wilhelm Braumüller.

Part II The Brentano School: Act, Meaning and Object

7 De Significatione: The Brentano-­ Ingarden Axis Sébastien Richard

1

Introduction

The Controversy over the Existence of the World (Ingarden 1962, 1964, 2013) and The Literary Work of Art (Ingarden 1931, 1973a, 1988) are some of the most important contributions to the ontology and the philosophy of art from the twentieth century. It is less known that Ingarden also wrote extensively on the theory of meaning. He even dedicated a whole course on language and meaning in Cracow in (1972). The lack of study of Ingarden’s work on language is due to the fact that this course is only available in Polish and that Ingarden’s texts on language and meaning translated into English are scattered in several books and papers (see mainly 1931, 1972, 1973a, b, 1988, 1997). The Polish phenomenologist never published a paper exclusively devoted to his theory of meaning.

S. Richard (*) Department of Philosophy, Free University of Brussels (ULB), Brussels, Belgium e-mail: [email protected] © The Author(s) 2021 A. Dewalque et al. (eds.), Philosophy of Language in the Brentano School, History of Analytic Philosophy, https://doi.org/10.1007/978-3-030-52211-7_7

143

144 

S. Richard

This is a pity because, in my opinion, this theory is highly original and deserves some attention (see however Malherbe, in this volume1). It is well known that Ingarden was a pupil of Husserl and it is therefore no surprise that his theory of meaning was heavily influenced by the theory of meaning that can be found in the Logical Investigations. Outside Poland, it has been less noted that Ingarden was also strongly influenced by Twardowski (for an exception, see Miskiewicz 2017) and through him by the whole Brentanian tradition. This is true in particular for his theory of meaning. In many respects it can be seen as a synthesis of some of the main contributions made in this tradition to the theory of meaning. My aim in this chapter is to cast some light on the influence of the Brentanian tradition in Ingarden’s theory of meaning. For reasons of lack of space, I will restrict my presentation to the meaning of names. I will first set out the most common theory of meaning that can be found in the Brentanian tradition, a theory that finds its origin in the work of the founding father of this tradition. This theory conceives meaning essentially as a subjective entity. Under the influence of the critique of psychologism, some Brentanians (in particular Meinong, Husserl and Twardowski) developed alternative theories that try to highlight the objective aspect of meaning. In Sects. 3–6, I will set out their attempts. In Sect. 7, I will finally show how some traces of these can be found in Ingarden’s own theory of meaning.

2

Brentano and others

I will naturally begin my journey through the Brentanian theories of meaning with the founding father of this tradition: Brentano himself. It is in his work that the view of meaning that was discussed and modified by the members of this tradition finds its origin. In his Psychology from an Empirical Standpoint, he associates names with presentations (Vorstellungen): When I hear and understand a word that names something, I have a presentation of what that word designates; and generally speaking the purpose of such words is to evoke presentations. (1995, 153)

7  De Significatione: The Brentano-Ingarden Axis 

145

Names can arouse by themselves presentations and are for this reason ‘categorematic expressions’. They differ from ‘syncategorematic expressions’, such as ‘and’, ‘or’, ‘however’, ‘each’, and so on, that can evoke mental phenomena only together with other expressions. This distinction between two kinds of expressions is one of the central hypotheses of the Brentanian research programme in philosophy of language. It can be found, for instance, in Marty (1884, 63), Twardowski (1982 [1894], 22; 1977, 19–21) and Husserl (1993, II.2, IV, 302–303; 2001, II, 54), even if they do not always agree on the class of expressions that are categorematic. Because they can arouse mental phenomena by themselves, categorematic expressions also have a complete meaning by themselves (Brentano Ms. EL 80, 13.010). Syncategorematic expressions, on the other hand, only have a meaning when they are joined with other words. Quite naturally Brentano identifies the meanings of categorematic expressions with the mental phenomenon that they evoke. The meaning of a name, in particular, is nothing else than the presentation that it arouses. More precisely it is the ‘content’ (Inhalt) of this presentation: What do names denote [bezeichnen]? The name denotes in a certain way the content of a presentation as such, the immanent object. In some other way, [it denotes] what is presented through the content of a presentation. The first is its meaning. The second is what is named by the name. We say about it that the name belongs to it [kommt ihm zu]. When it exists, this is the external object of the presentation. (Ms. EL 80, 13.016)

The meaning of a name is thus the content of a presentation and it is through this content that the name refers to the object it names. This is, in a nutshell, Brentano’s theory of the meaning of names. It was later expanded and made more precise by Twardowski (1982 [1894]; 1977). He tried in particular to clarify the distinction between the content and the object of a presentation. The content is a part of the mental act that directs it toward its object. Being mental, the content exists as really as the act of presentation. The content of a presentation, and thus the meaning of a name, is therefore a mental entity. The Brentano-Twardowski conception of meaning is a kind of semantic subjectivism, i.e. a view that

146 

S. Richard

makes meaning something private. At this stage, it can be schematised in the following way (Fig. 7.1). This schema is ambiguous, because it does not specify the mental content which is the meaning of a name. To determine it more precisely we must take into account something that was implicit in our presentation: the theory of ‘linguistic functions’ that comes into play in communication. This theory was a powerful tool in the Brentanian research programme in philosophy of language. Even if scholars generally refer to Marty for this theory, it seems to have been introduced first by Brentano in his 1877 lectures on logic (see Fisette and Fréchette 2007, 158).2 Let me focus on names. Such expressions denote objects: it is their ‘naming’ (Nennung) function. When I want to communicate something to somebody, I do not only want to name something, I also want to ‘indicate’ (kundgeben) to my addressee what I have in mind, i.e. a certain mental act (a presentation in the case of names). If she understands me, I managed to ‘awaken’ (erwecken) in her another mental act (a presentation whose content is similar to the content of my presentation). Besides the naming function, we must thus also distinguish the awakening function and the indicating function. What is now the content that constitutes the meaning of a name? Is it the mental content of the speaker who pronounces a word (the indicated content) or is it the mental content of the addressee who hears this word (the awakened content)? Even if the first branch of the alternative could seem more natural, Brentano (1956, 46–48), but also Marty (1884, 68–69) and Twardowski (1982 [1894], 11; 1977, 9) have identified the name

means

denotes

content of presentation

gives access to

object Fig. 7.1  Simplified Brentanian theory of meaning

147

7  De Significatione: The Brentano-Ingarden Axis 

awakens

indicates speaker

pronounces

hears

name

has

has act of presentation is a part of content of presentation

addressee

denotes intends

gives access to

means intends

object

gives access to

act of presentation is a part of content of presentation

Fig. 7.2  Complete Brentanian theory of meaning

meaning of a name with the mental content aroused in the addressee. This being said, I can now give a full schema of Brentano’s subjective view of the meaning of names (Fig. 7.2).

3

Meinong I

As his master Brentano, Meinong thought that psychology had to play a fundamental role in philosophy. However, he became progressively an opponent of psychologism. He was led on this path by his works on wholes and parts that he carried out in the context of the debate over Gestalt qualities (see Richard 2014) and the question of the ontological status of non-existents that he first encountered in the theory of value (see Lindenfeld 1980, chapter V; see also Sierszulska 2005, 17–21). Meinong thought that a careful distinction between the content and the object of mental acts would allow him to avoid psychologism. As far as the theory of meaning is concerned, this means that one should embrace an objectivist conception of meaning. Contents cannot serve as meaning, because they may differ from one act to another. From Meinong object-oriented perspective, a natural way to guarantee semantic objectivism amounts to externalising the meaning of a word by identifying it with the object intended by the mental act indicated by this word. This

148 

S. Richard

object faces the act that intends it and can be said to be transcendent with respect to the mental sphere. What is the meaning of the name ‘Donald Trump’ if not Donald Trump himself? This identification of the meaning with the object intended by a mental act explains the communication between two persons: they mean the same thing, because this thing is the same object existing outside the mind. This kind of semantic objectivism, that we could call ‘externalism’, has been maintained by Meinong in the second edition of Über Annahmen (1910, 1983). For him, an ‘expression’ (Ausdruck) expresses (indicates) a certain mental event (1910, 24–25; 1983, 24). When somebody pronounces the words ‘sun’, she expresses her presentation of the sun. Unlike Brentano, Meinong makes a distinction between what an expression expresses—the content of the mental act that accompanies it—and its ‘meaning’ (Bedeutung). This meaning is constituted by the object ‘presented’ (präsentiert) by the mental act that the word expresses (1910, 28; 1983, 27). A presentation presents its object through its content. What we want to say with words is thus not what these words express (the content), but what they mean, i.e. the object they are about. This Meinongian theory of the meaning of names can be schematised in the following way (Fig. 7.3). It must be noted that the objects of mental acts are not necessarily real objects. For instance, the meaning of the word ‘Pegasus’ is not an existing winged horse, but a transcendent non-existent object, an object ‘outside being and non-being’ (Außersein). This object is what Meinong calls a name

expresses

means

presentational content is a part of

gives access to

pronounces expresses speaker

has

Fig. 7.3  Meinong’s first theory of meaning

act of presentation

intends

object

7  De Significatione: The Brentano-Ingarden Axis 

149

‘pure object’. It is not a subjective entity, because, for him, ‘all objects, regardless of their existence or its lack, are non-subjective, mind-­ independent entities’ (Sierszulska 2005, 21). By assuming that there are objects that do not exist and are independent from the mind (see Leclercq, this volume), Meinong can claim that every well-formed expression has an objective meaning. The Meinongian externalist theory of meaning is a kind of hyperreferential semantic objectivism. This theory faces an obvious problem: it is impossible to associate different meaning to the same object, since meaning and object are one and the same. For instance, the expressions ‘the winner of Jena’ and ‘the loser of Waterloo’ name the same object, i.e. Napoleon, and thus have the same meaning according to the Meinongian theory of meaning. This counter-intuitive consequence appeals for some modification to this theory. In Über Möglichkeit und Wahrscheinlichkeit (1915), Meinong developed a second theory of meaning that is immune to this defect.

4

Meinong II

When I roll a die, I obtain a certain number between one and six. Once executed, my roll is thus determinate with respect to a number. By contrast, my next roll—the one I have not yet executed—is only a possible object whose result is indeterminate. It is what Meinong calls an ‘incomplete’ (unvollständig) object. This object is incomplete because it does not satisfy the restricted ‘ontological principle of the excluded middle’. Let me consider the property of being even. If I apply this property to the possible object denoted by ‘the result of my next die roll’, the application of the ontological principle of the excluded middle is: (1) The result of my next die roll is even or the result of my next die roll is not even. What is the truth value of this sentence? We cannot answer this question, because the negation in (1) is ambiguous. Is it propositional or nominal? If the negation is propositional, we obtain an ‘extended’ (erweiterte) version of the ontological principle of the excluded middle (Meinong 1915, 171):

150 

S. Richard

(2) The result of my next die roll is even or it is not the case that the result of my next die roll is even. If the negation is nominal, we obtain a ‘restricted’ (unerweiterte) version of the ontological principle of the excluded middle (Meinong 1915, 171): (3) The result of my next die roll is even or the result of my next die roll is non-even, which sentence is in fact equivalent to: (4) The result of my next die roll is even or the result of my next die roll is odd. (2) is true, since it is not the case that my next die roll is even (as far as it is not the case that it is odd). By contrast, (3) and (4) are false, since neither is it the case that the result of my next die roll is even, nor is it the case that it is odd. The result of my next die roll is indeterminate with respect to the property of being even, as it is with respect to the ‘complementary property’ of being odd. In general, an object o is incomplete if it does not satisfy the following ontological principle: (4) o is P or o is non-P with respect to at least one property P.

Meinong modified his theory of meaning with the concept of incomplete objects. Let me consider the proper name ‘Socrates’. This name denotes Socrates, i.e. the object intended by means of the content of the presentation associated with this name. In this sense, Socrates can be called the ‘target object’ (Zielgegenstand) of this presentation (Meinong 1915, 196). When I have a presentation of Socrates, I do not only intend him, I also have some presentation of him. For instance, I can have a presentation of him as the master of Plato. In that case, according to Meinong, I deal with another object: Socrates as (als) the master of Plato. This object is not presented by the content of my presentation, but

7  De Significatione: The Brentano-Ingarden Axis 

151

‘quasi-­presented’ (quasi-präsentiert) by the content of my presentation. Socrates as the master of Plato is what Meinong calls an ‘auxiliary object’ (Hilfsgegenstand). It is an incomplete object that incarnates the way the target object is intended. It is that object that constitutes the meaning of the name ‘Socrates’. It is thus no longer Socrates himself, who he is only the object named by this name. Different auxiliary objects can be used to intend the same object (Meinong 1915, 197). I can have a presentation of Socrates as the master of Plato or as the philosopher who drank the hemlock. Different meanings can in this way be associated with the same object and with the same expression. What is the relation between an auxiliary object and the corresponding target object? Meinong says that the former is ‘implected’ (implektiert) in the latter (see Leclercq, this volume). This means, roughly, that all the properties of the implected object are properties of the object in which it is implected. These precisions having been made, I can now schematise Meinong’s second theory of meaning for names (Fig. 7.4). When the object intended by a mental act exists, in one way or another, the auxiliary object can be identified with it. As it is the case with empty names, there may also be no transcendent (real or ideal) object intended by a mental act. However, even in this case, there is always an auxiliary object of such an act, as is proven by the linguistic fact that we can communicate about objects which do not have any kind of being. The expressions which are then used do not denote anything, but still have a meaning. From this point of view, Meinong’s second theory of meaning name

expresses

presentational content quasipresents

means is a part of

pronounces expresses

speaker

has

act of presentation

Fig. 7.4  Meinong’s second theory of meaning

presents

auxiliary object is implected in intends

object

152 

S. Richard

is still a kind of semantic objectivism, but it is no more a hyperreferential one.

5

Husserl

Like Frege, Husserl disdained psychologism (see Fisette, this volume). Around the time of the Logical Investigations, he claimed that to assimilate the meanings to mental contents, and consequently to reduce their analysis to psychology, was a way toward psychologism (Fisette and Fréchette 2007, 137). It is therefore no surprise that he criticised the subjective theory of meaning that I have previously set out (see Gallerand 2013). He did not directly discuss Brentano’s version of it, but rather Twardowski’s (see Husserl 1979). For him, meaning is ‘identically the same “in” an unlimited multiplicity of acts’ (1979, 350, note), whereas content is something individual that only exists here and now. The meaning of a name, in particular, cannot be the content of the corresponding presentation, because then the same name used in the same context could have different meanings for different persons if the content awakened in them is different. For instance, when pronounced, the word ‘tree’ could arise the content ‘lime’ for one person and the content ‘fir’ for another (Husserl 1979, 349, note). Husserl wants to elaborate a non-psychological concept of meaning.3 Meaning is a logical concept, and not a mental one. But contrary to what Frege did, this does not mean that meaning must be completely disconnected from mental acts. Without them the directionality of words toward their objects would remain mysterious. It is thanks to some acts that words get their meaning: it is ‘conferred’ (verliehen) to them by ‘meaning intentions’, also called ‘meaning-giving acts’. There are thus two aspects of meaningful expressions: their physical aspect and their meaningful aspect, which is given to them by some acts (Husserl 1993, II.1, I, 37; 2001, I, 191–192). This does not mean that any meaning can be arbitrarily attributed to an expression. For Husserl, we ‘discover’ the meaning of an expression in it (1993, II.1, I, 44; 2001, I, 196). As stressed by Smith (1994, 174), the physical aspect and the meaning-giving act cannot exist independently—they are only separated through

7  De Significatione: The Brentano-Ingarden Axis 

153

abstraction. They are depending parts of this whole that is the meaningful expression. When I am directed toward a concrete word through a meaning intention, I do not add to it a meaning, I rather understand this word as a meaningful expression. What is modified then is my intention: it is not any more the intention of a phenomenon that is only physical, but a meaning intention. After this modification, my interest is not in the sign as a concrete object, but in the object denoted by this sign in virtue of its meaning. With a meaning-giving act, something objective is ‘intended’ (vermeinte). Meaning as a phenomenal experience is thus intentional: it is a mental act with a content—what Husserl calls the ‘matter’ (Materie) of the act (1993, II.1, V, 411; 2001, II, 120)—and an object that is intended through its content. As for Brentano and Twardowski, the content of the meaning intention is what determines how we relate to the object denoted by an expression. For instance, the presentations expressed by ‘the winner of Jena’ and ‘the loser of Waterloo’ intend the same object, i.e. Napoleon, but they not intend it in the same way. As a consequence, we should not identify meaning with the object intended by the meaning intention, as Meinong did in his first theory of meaning. For Husserl, meaning is intentional, but it also has an objective aspect. When I state that the three perpendiculars of a triangle intersect in a point, I indicate to someone that I have a judging act. The addressee recognises me as the one who judges in such a way. But is the content of my judging act (or the one awakened in the addressee) the meaning of the sentence ‘the three perpendiculars of a triangle intersect in a point’? Certainly not, Husserl claims, because meaning is something that remains identical, contrary to judging acts. What this assertion asserts is the same whoever may assert it, and on whatever occasion or in whatever circumstances he may assert it, and what it asserts is precisely this, the three perpendiculars of a triangle intersect in a point, no more no less. (1993, II.1, I, 43; 2001, I, 195)

For Husserl, meaning is something that is ‘nothing subjective’ (1993, II.1, I, 44; 2001, I, 195), contrary to physical experiences. He thus maintains a kind of semantic objectivism, which he inherited from Bolzano,

154 

S. Richard

more than Frege. For Husserl, the objective character of meaning is the only one that explains the identity of meaning and its publicity, i.e. the fact that we can communicate. How can we explain the objective aspect of meaning? Husserl claims that meaning is what is identical in different meaning intentions. It must then be understood as the ‘ideal unity’ of those acts: Meaning is related to varied acts of meanings […] just as Redness in specie is to the slips of paper which lie here, and which all ‘have’ the same redness. Each slip has, in addition to other constitutive aspects (extension, form etc.), its own individual redness, i.e. its instance of this colour-species, though this neither exists in the slip nor anywhere else in the whole world, and particularly not ‘in our thought’, in so far as this latter is part of the domain of real being, the sphere of temporality. (1993, II.1, I, 100–101; 2001, I, 230)

Meaning is neither something real in our thought (it is not a mental content) nor something in the real world (it is not an empirical object), but an ideal ‘species’ instantiated in the individual contents of mental acts. In this sense, meanings are ideal entities. In this view, it is still the content of the mental act that is responsible for the directedness toward the object of a name. Ideal species only guarantee the objective character of meaning and help to explain communicability: different language users can understand each other not because the content aroused in the mind of the listener is sufficiently similar to the content indicated in the mind of the speaker, but because their contents are instantiations of the same ideal species (Fig. 7.5). Since he does not identify meanings with objects, Husserl needs not to assume that meaning acts always have an object. Empty expressions have a meaning, because the content of the mental act that they indicate instantiates an ideal species, but they do not denote anything. In this case, the meaning intention is simply ‘empty’, it is not fulfilled by anything (see Husserl 1993, II.1, 1, 54–55; 2001, 1, 201–203).

7  De Significatione: The Brentano-Ingarden Axis  means

name

155

ideal species

denotes

is instantiated in

indicates pronounces speaker

has

object

act of meaning

intends gives access to

is a part of matter Fig. 7.5  Husserl’s theory of meaning

6

Twardowski

The reading of Husserl’s Logical Investigations and debates with Łukasiewicz led Twardowski to reject the psychologistic conception of meaning that he had sustained in 1894. However, he could not make sense of the Husserlian ideal species. His 1912 paper “On Actions and Products” is an attempt to overcome psychologism and to make the ontological status of ideal meanings more precise. Twardowski identifies the origin of the psychologistic conception of meaning that he once held in the confusion between what he calls ‘actions’ (czynność) and ‘products’ (wytwór).4 The difference between actions and products is, first of all, a ‘logico-­ grammatical’ difference (see Fisette, in this volume). It is the difference, for instance, between to jump and the jump, to think and the thought, to draw and the drawing. Among actions and products, we need to make a distinction between those that are purely ‘physical’ and those that are ‘mental’. For instance, a jump is the physical product of the physical action of jumping. By contrast, the thought and the action of thinking are mental. The drawing, because it is the expression of a mental presentation, is a ‘psychophysical’ product. It is the result of the physical action of drawing, but this action must be accompanied by some mental action

156 

S. Richard

that ‘affects the course of the physical activity, and thereby the product that results from it’ (Twardowski 1965 [1912], 230; 1999b, 120). With this new distinction Twardowski is able to reformulate his former distinction between the act of presentation and its content: the content of a presentation is the product of the action of presenting and the object of the presentation is what is intended by the presentation through its product. The meaning of a name is thus the product of an act of presentation. What did Twardowski gain with this reformulation? To understand it, we must introduce a supplementary distinction between ‘enduring’ (trwały) and ‘non-enduring’ products (see Twardowski 1965 [1912], 228; 1999b, 118). A non-enduring product is a product that lasts as long as the action that produces it. Yet, meaning seems to outlast the mental act that comes with it. Let me here consider the case of written words. These are enduring products that continue to exist after the action of writing. It is thanks to them that the meanings that they express can be ‘preserved’. They potentially exist eternally in these enduring physical products that are the written words. When I read a word, this word awakens in me a mental product that constitutes its meaning. This meaning can be actualised each time that I read a word expressing it. In this way, meanings are enduring products, but enduring products that only exist in a non-actual way (Twardowski 1965 [1912], 233–234; 1999b, 124–125). Words confer on meanings the ‘semblance of endurance by being the enduring effects’ (Twardowski 1965 [1912], 234; 1999b, 125) of non-enduring mental actions. It is this semblance of endurance that explains the intersubjective aspect of meanings: different persons can access the meaning of a word, because this meaning exists potentially in this word, i.e. because when these persons read this word their reading arouses in them mental acts that actualise the meaning of this word. However, at this stage, we have no guarantee that the same meaning will be actualised in these persons. Being mental products, there still seems to be as many different meanings that there are different individuals. How can we be sure that meanings are something objective? Twardowski makes a distinction between two kinds of meanings. He notices that, despites the variety of mental products that arise in different persons reading the same words, there is ‘a group of common attributes

7  De Significatione: The Brentano-Ingarden Axis 

157

in these individual mental products’ (1965 [1912], 236; 1999b, 127). According to him, these common elements are precisely what we ‘ordinarily regard as the meaning’ of words. This second and ordinary sense of meaning is not a concrete product individually determinate, but the result of an act of ‘abstraction’ performed on concrete products. When read by two persons, the word ‘tree’ does not have the same meaning, understood as a mental content, but it has the same meaning, understood as the product of an act of abstraction, because the contents aroused by the reading of this word are sufficiently similar to extract from them what they have in common. I can now give a schema of Twardowski’s second theory of meaning for names (Fig. 7.6). It can be doubted that Twardowski successfully overcame psychologism with respect to meaning with his new theory.5 The problem lies in his theory of abstraction. Even if Husserl agrees that a form of abstraction is necessary to grasp meanings, he does not understand this process in the same way as Twardowski. The latter conceives of abstraction as an act of ideal meaning is abstracted from act of presentation

produces

intends

presentational content

means in the second sense

means in the first sense

indicates act of writing

produces

Fig. 7.6  Twardowski’s second theory of meaning

written name

denotes

object

158 

S. Richard

attention that allows us to mentally neglect some of the characteristics of an object (see Twardowski 1965 [1924], 301–302; 1999a, 84–85; Schaar 2016, 107–108). But if abstraction is a mental act performed on the product of another mental act, how could the product of such a second-­ order action be something ideal? Abstraction in this sense only produces something individual from an individual object. For Husserl, abstraction is rather a kind of ‘ideation’ in which the general is intended on the basis of something individual. While meaning is primarily something ideal for Husserl, for Twardowski it is the concrete result of a mental activity, something that exists as really as this act.

7

Ingarden

Ingarden considered the Husserlian critic of psychologism as ‘devastating’. It would have overcome the theory in which meanings are mental entities (Ingarden 1931, 96; 1988, 152; 1973a, 95–96). However, he was more critical of Husserl’s theory according to which meanings are ideal species. He blamed it for having made meanings something timeless, and consequently unable to change. Ingarden thought that it became then impossible to understand some of the fundamental characteristics of meaning, and, first of all, the fact that the same word can have different meanings according to its position in a sentence. Let me consider the two following sentences: ( 1) In the Roman state, the consul was a person of great influence. (2) The consul, who was murdered by Brutus, had previously crossed the Rubicon. Ingarden claims that the meaning of every word contains an ‘intentional directional factor’ (intentionaler Richtungsfaktor). This directional factor is not the same in sentences (1) and (2): in (1) it is indeterminate and in (2) it is determinate and directed toward Julius Caesar (see Ingarden 1931, 93; 1988, 149; 1973a, 93). Facing such a modification, Husserl would say that, between the two sentences, we moved from one meaning to another one—in each sentence a different meaning is

7  De Significatione: The Brentano-Ingarden Axis 

159

instantiated in the content of our meaning acts. Ingarden does not accept such an explanation for several reasons. Firstly, it renders incomprehensible the idea that the author of a text ‘creates’ expressions, i.e. units gathering a physical phenomenon and a meaning, and not simply sequences of concrete words. Husserl indeed claims that we do not posit the meaning in our expressions, but that we find it. For him, therefore, an author creates a sentence as a sequence of concrete words, but not as a unit of words and meaning, i.e. of what he calls ‘expressions’.6 Secondly, the Husserlian conception of meaning as ideal species arrives at a ‘monstrous proliferation’ of meanings (Ingarden 1931, 99; 1988, 155; 1973a, 98). Ingarden thinks that words have only one meaning (1997, 25; 1973b, 25), but this meaning can be modified. Admittedly, meaning has a certain kind of stability, since we can use the same expression in different contexts to express an identical meaning, but it is not completely stable, since it does not always remain the same—the meaning of a word can change in different contexts, as illustrated in the two sentences above. To underline this ‘contextual’ dependency of meanings, Ingarden speaks about a ‘quasi-static unity’ of meanings (1997, 28; 1973b, 28). Words rarely appear in isolation, except in dictionaries but they are then most of the time polysemous. They are tied to other words, pronounced or written by different speakers at different times, in different places and sentences. It is the context of utterance that allows us to identify the right meaning of a word. If Ingarden rejects Husserl’s theory of meaning as ideal species, he does not, however, embrace a kind of semantic subjectivism. His position is more subtle. He tries to keep his distance from Charybdis and Scylla, i.e. from a purely idealist position such as Husserl’s and from a purely subjectivist position such as Brentano’s.7 Meaning cannot be a real part of a mental act, something private and incommunicable, because communication would then be a guessing game: if someone wants to understand what I am saying when I utter an expression, he must guess the content of my mental acts. If this was really the case, communication would be a ‘miracle’. For Ingarden, an expression is something ‘intersubjective’, an entity whose meaning is accessible to different persons. It is not something private (1997, 30; 1973b, 29). Meaning must be neither real nor ideal.

160 

S. Richard

Ingarden synthesises several aspects of the theories of meaning advocated in the Brentanian tradition that I have set out before. From Husserl, he takes up the idea that we confer meanings to words (1997, 25; 1973b, 25). From Twardowski, he takes up the idea that meaning is produced by subjective operations. Meaning is therefore a ‘purely intentional object’ (see Ingarden 1931, § 66; 1988; 1973a), an object that exists in a different way than ideal and real objects. As purely intentional, meanings are objects that owe their ‘existential foundation’ (Seinsfundament) to meaning acts and are, for this reason, existentially ‘heteronomous’. From a descriptivist point of view, meaning appears to us not as a part of a mental act, but as a unitary whole, ‘a self-enclosed unit of meaning’ (Ingarden 1931, 109; 1988; 1973a, 107) that somewhat faces us. This is particularly clear in ‘derived intentionality’: the fact that meaning exists potentially in expressions and can be actualised by different persons implies that it can be separated from them. In other words, meaning ‘transcends’ every mental act. Being a mental product, meaning is non-enduring, i.e. it exists only as much as the act that creates it exists. Twardowski had, however, explained that non-enduring products could be fixed through written words and sentences that express them. The existence of the meaning of a word or a sentence is therefore only potential. We must intend a concrete word as an expression for it to actualise its meaning and to denote some object. When someone writes a word, she confers to it a meaning, which meaning refers to an object. It is only when this word is read that this originary meaning intention is reactivated,8 actualised (1997, 34; 1973b, 32), and that what this word means can be understood. We can, of course, be mistaken when we re-actualise the meaning intention of a word, but this is not what happens most of the time and this kind of error can be corrected. Ingarden also deeply modifies Twardowski’s theory of meanings as products. This theory assimilates meanings to pure products of subjective operations (Ingarden 1931, 107–108, note 2; 1988, 166, note 2; 1973a, 106, note 61), but if meanings have their ontological foundation in such operations, they also have a part of their ontological foundation in some idealities. These are not ideal species instantiated in expressions. As said earlier, the reason for this is that each time a word is used with a different

7  De Significatione: The Brentano-Ingarden Axis 

161

intention, it would instantiate a different meaning and there would be an inflation of ideal meanings. Nevertheless, some ideal elements must come into play. For Ingarden, the creation of meanings is not a creation ex nihilo. It is carried out from an ideal material that is structured into an expression by a cognitive agent. When someone produces an expression, on the one hand, she actualises some ‘pure qualities’ in its material parts and, on the other hand, she organises these ‘meaning elements’ into a whole. In other words, an expression does not instantiate a whole ideal meaning (Husserl), but contains (material) parts that instantiate pure qualities and that are structured (given a form) by subjective ‘forming operations’ (see 1931, 102–103; 1988, 159–160; 1973a, 101–102). There are here two kinds of ideal entities that must not be confused in Ingarden’s ontology: pure qualities and ideas. In first approximation, we could think that ideas are Husserl’s ideal species, but, contrary to these, ideas are not types of mental content. They rather are ideal concepts of objects, ideas that subsume the objects to which our words refer (Chrudzimski 1999, 308), such as, for instance, the idea of a triangle in general which subsume all the individual triangles. As for pure qualities, they are sort of ‘bare universals’ that can be (ideally) concretised in ideas and instantiated in (realised in) real objects and (actualised in) meanings. It is, for instance, the ideal red that is instantiated in a red object. A meaning does not instantiate an idea, but the ‘meaning elements’ of the corresponding idea. The meaning of a word is a whole in which these ideal elements are structured. Therefore, the meaning of an expression is something new, a product that must be distinguished from the idea on which it is modelled, i.e. the idea that is intended by the person who created this product through a sense-giving act. It is the actualisation, in a meaning, of the meaning elements contained in ideas that guarantees the stability and ‘intersubjective identity’ of the meaning. It is this ontological foundation of meanings in pure qualities that makes communication possible and that guarantees its objective aspect. Here is the schema of Ingarden’s theory of meaning for names (Fig. 7.7). As can be seen on this schema, Ingarden distinguishes the meaning of an expression both from the purely intentional object projected by it and from the real or ideal object denoted by it. It can be asked why he does not identify the meaning with the projected object. His theory would

162 

S. Richard are pure concretised qualities in are actualised in name

expresses

creates indicates

speaker

has

meaning act

intends is a part of

are realised in refers to

denotes

is the foundation of

pronounces

subsume

purely intentional object

projects

meaning

ideas

object gives access to mental content

Fig. 7.7  Ingarden’s theory of meaning

then be similar, in some respects,9 to Meinong’s second theory. To my knowledge, Ingarden does not give any explanation for distinguishing the meaning of an expression from the object projected, but the examination of fictional expressions offers us some clues. Firstly, Ingarden claims that every expression denotes an object, even fictional names. For instance, Hamlet exists and is the object denoted by the name ‘Hamlet’. Fictional objects are neither real nor ideal objects: they exist purely intentionally; they are projected through the meanings of the expressions that denote them. When an expression denotes a real or ideal object, it still projects a purely intentional object, but this object is not noticed—it is ‘transparent’ (see e.g. Ingarden 1997, 39, 170; 1973b, 37, 149). Secondly, as already denounced by Husserl (1993, II.1, I, § 12; 2001, 1, § 12), the meaning of an expression must not be confused with its object, because two expressions can have different meanings and denote one and the same object. For instance, the expressions ‘the Prince of Denmark’ and ‘the son of the late King Hamlet’ have different meanings, but they denote the same object, i.e. Hamlet. Thirdly, the meaning of an expression cannot be the object projected by this expression, because it is precisely in virtue of this meaning that the object can be projected. It is

7  De Significatione: The Brentano-Ingarden Axis 

163

because she produces the meaning of the name ‘Hamlet’ that someone who reads Shakespeare’s play projects Hamlet. To conclude, it must be admitted that Ingarden’s theory of meaning is quite complex. It contains elements from Twardowski’s and Husserl’s theories of meaning, while sending back to back the identification of meanings with mental contents and ideal species. Meaning is neither real nor ideal. It is a purely intentional object produced by subjective forming-­ operations from instantiated ideal meaning elements. Contrary to the Husserlian ideal meaning, it can be modified by actualising different pure qualities and by structuring them in another way. It however contains a stable core and is transcendent with respect to the individual acts of consciousness. In this respect, Ingarden’s theory of meaning can be viewed as the peak of the attempts made in the Brentanian tradition to conciliate the subjective and objective aspects of meaning. It was also probably the last one.

Notes 1. My approach of Ingarden’s philosophy of language in the present paper focuses on his theory of meaning and what it owes to other members of the Brentanian tradition, while Malherbe’s paper is more internal with respect to Ingarden’s work. 2. See (Brentano Ms. EL 108, 21, quoted in Marek 2001, 282; and Marty 1884, 300). In the Brentanian tradition, this theory was also adopted by Twardowski, Husserl and Ingarden, among others. They did not all agree on the names and the number of the functions. 3. Husserl’s view about meaning in the Logical Investigations is now rather well-known. I can thus be content to be succinct with my explanations here. On this view, see, among others, (Smith 1994; Simons 1995; Benoist 2001; Gallerand 2013). 4. These notions were introduced in the Brentanian tradition by Stumpf (see 1906 [2006]; see also Taieb 2018). 5. For Twardowski psychologism with respect to logic is overcome by a delimitation of philosophical disciplines and their objects. He does not dispute the foundational role of psychology, but the fact that meanings are its subject-matter. Psychology only studies mental acts. If meanings are

164 

S. Richard

products of such mental acts, and are thus ontologically dependent on them, they can also be studied autonomously by logic. (1965 [1912], § 45; 1999b, § 45; see also 1999c). 6. As Husserl, Ingarden considers that an expression is a unity in which the material aspect of the sign and its meaning (what the expression expresses) are simultaneously grasped. It is only through abstraction that we can make a distinction between these two aspects of expressions. 7. He thinks that it was also the intention of Twardowski with his second theory of meaning, but that he did not succeed (see Ingarden 1948, 28–29). 8. Two kinds of sense-giving acts must be distinguished. On the one hand, we have the act in which someone creates a new word and confers a meaning to it. In this case, the meaning intention is direct. On the other hand, we have the act in which a word already exists and has a commonly accepted meaning. In this case, the reader reconstitutes the originary meaning intention that has been associated with the word when it was created. The meaning intention is then not direct, but derived (abgeleitet) (see Ingarden 1997, 25; 1973b, 25). 9. An important difference between a pseudo-Ingardenian theory of meaning that would identify the meaning of an expression with the object projected by this expression and Meinong’s second theory is that, for Meinong, the auxiliary object, i.e. the meaning of a word, neither exists nor does not exist—it is a pure object beyond being and non-being–, while, for Ingarden, the object projected by an expression exists purely intentionally.

References Benoist, Jocelyn. 2001. Intentionalité et langage dans les Recherches logiques de Husserl. Paris: Presses Universitaires de France. Brentano, Franz. 1956. In Die Lehre vom richtig Urteil, ed. Franziska Mayer-­ Hillebrand. Bern: Francke Verlag. ———. 1995. Psychology from an Empirical Standpoint. Trans. Antos C. Rancurello, D.B. Terrell, and Linda L. McAlister. London: Routledge. ———. n.d.-a. Ms. EL 80. Logik. Unpublished Manuscript EL 80. Provisional Online Edition. Edited by Robin D. Rollinger. Franz Brentano Archiv Graz. http://gams.uni-graz.at/archive/objects/context:bag/methods/sdef:Context/ get?mode=logik.

7  De Significatione: The Brentano-Ingarden Axis 

165

———. n.d.-b. Ms. EL 108. Alte und neue Logik. Unpublished Manuscript EL 108. Cambridge (Mass.): Houghton Library. Chrudzimski, Arkadiusz. 1999. Are Meanings in the Head? Ingarden’s Theory of Meaning. Journal of the British Society for Phenomenology 30: 306–326. Fisette, Denis, and Guillaume Fréchette. 2007. Le legs de Brentano. In À l’école de Brentano. De Würzbourg à Vienne, 13–160. Paris: Vrin. Gallerand, Antoine. 2013. La critique husserlienne de la théorie psycho-­ descriptive de la signification. Archives de philosophie 134: 35–51. Husserl, Edmund. 1979. Besprechung von: K.  Twardowski, Zur Lehre vom Inhalt und Gegenstand der Vorstellungen. Eine psychologische Untersuchung, Wien 1894. In Husserliana XXII: Aufsätze und Rezensionen (1890–1910), ed. Bernhard Rang, 349–356. The Hague: Nijhoff. ———. 1993. Logische Untersuchungen. Vol. 3. Tübingen: Max Niemeyer. ———. 2001. Logical Investigations, 2 vols. Trans. John N. Findlay. London: Routledge. Ingarden, Roman. 1931. Das literarische Kunstwerk. Halle: Max Niemeyer. ———. 1948. The Scientific Activity of Kazimierz Twardowski. Studia Philosophica 3: 17–30. ———. 1962. O spór o istnienie świata. Vol. 1. Warsaw: PWN. ———. 1964. Der Streit um die Existenz der Welt. Vol. 1. Tübingen: Max Niemeyer. ———. 1972. O języku i jego roli w nauce. In Z teorii języka i filozoficznych podstaw logiki, 29–119. Warsaw: PWN. ———. 1973a. The Literary Work of Art. Trans. George G. Grabowicz. Evanston: Northwestern University Press. ———. 1973b. The Cognition of the Literary Work of Art. Trans. Ruth A. Crowley and Kenneth R. Olson. Evanston: Northwestern University Press. ———. 1988. O dziele literackim. Warsaw: PWN. ———. 1997. Vom Erkennen des literarischen Kunstwerks. Tübingen: Niemeyer. ———. 2013. Controversy over the Existence of the World, 1. Trans. Arthur Szylewicz. Frankfurt am Main: Peter Lang. Lindenfeld, David F. 1980. The Transformation of Positivism. Alexius Meinong and European Thought, 1880–1920. Berkeley: University of California Press. Marek, Johann Christian. 2001. Meinong on Psychological Content. In The School of Alexius Meinong, ed. Liliana Albertazzi, Dale Jacquette, and Roberto Poli, 261–286. London: Routledge. Marty, Anton. 1884. Über subjektlose Sätze und das Verhältnis der Grammatik zu Logik und Psychologie. In 1918. Gesammelte Schriften, vol. II.1, ed. Joseph

166 

S. Richard

Eisenmeier, Alfred Kastil, and Oskar Kraus, 1–101. The Hague: Niemeyer, Halle. Meinong, Alexius. 1910. Über Annahmen. 2nd ed. Leipzig: Johann Ambrosius Barth. ———. 1915. Über Möglichkeit und Wahrscheinlichkeit. Beiträge zur Gegenstandstheorie und Erkenntnistheorie. Leipzig: Johann Ambrosius Barth. ———. 1983. On Assumptions. Trans. James Heanue. Berkeley: University of California Press. Miskiewicz, Wioletta A. 2017. On Actions and Products (1911) by Kazimierz Twardowski: Its Historical Genesis and Philosophical Impact. In Act-Based Conceptions of Propositional Content, ed. Friederike Moltmann and Mark Textor, 161–177. Oxford: Oxford University Press. Richard, Sébastien. 2014. De la forme à l’être. Sur la genèse du projet husserlien d’ontologie formelle. Paris: Ithaque. van der Schaar, Maria. 2016. Kazimierz Twardowski: A Grammar for Philosophy. Leiden-Boston: Brill-Rodopi. Sierszulska, Anna. 2005. Meinong on Meaning and Truth. Frankfurt: Ontos Verlag. Simons, Peter. 1995. Meaning and Language. In The Cambridge Companion to Husserl, ed. Barry Smith and David Woodruff Smith, 106–137. Cambridge: Cambridge University Press. Smith, Barry. 1994. Husserl’s Theory of Meaning and Reference. In Mind, Meaning and Mathematics. Essays on the Philosophical Views of Husserl and Frege, ed. Leila Haaparanta, 163–183. Dordrecht: Kluwer. Stumpf, Carl. 2006. Erscheinungen und psychische Funktionen. Abhandlungen der Königlich-Preußischen Akademie der Wissenschaften. Berlin: 3–40. Taieb, Hamid. 2018. Building Objective Thoughts: Stumpf, Twardowski and the Late Husserl on Psychic Products. Archiv für Geschichte der Philosophie 100 (3): 336–370. Twardowski, Kazimierz. 1965 (1912). O czynnościach I wytworach. In Wybrane pisma filozoficzne, 217–240. PWN: Warsaw. ———. 1965 (1924). O istocie pojęc. In Wybrane pisma filozoficzne, 292–312. PWN: Warsaw. ———. 1977. On the Content and Object of Presentations. Trans. Reinhardt Grossmann. The Hague: Nijhoff. ———. 1982 (1894). Zur Lehre vom Inhalt und Gegenstand der Vorstellungen. Munich and Vienna: Philosophia Verlag. ———. 1999a. The Essence of Concepts. Trans. Arthur Szylewicz. In On Actions, Products and Other Topics in Philosophy, 73–97. Amsterdam: Rodopi.

7  De Significatione: The Brentano-Ingarden Axis 

167

———. 1999b. Actions and Products. Trans. Arthur Szylewicz. In On Actions, Products and Other Topics in Philosophy, 103–132. Amsterdam: Rodopi. ———. 1999c. The Humanities and Psychology. Trans. Arthur Szylewicz. In On Actions, Products and Other Topics in Philosophy, 133–140. Amsterdam: Rodopi.

8 Meaning(s) in Roman Ingarden’s Philosophy of Language Olivier Malherbe

1

Introduction

Roman Ingarden (1893–1970), a phenomenologist and (highly) critical disciple of Husserl, took an early interest in the question of language. He indeed devoted his first major work, Das literarische Kunstwerk (The Literary Work of Art, published in 1931), to this question, aiming in particular to dismiss both the psychologist approach and the logicist approach of the Vienna Circle. Unlike many philosophers of his time, Ingarden directly dived into the analysis of a particularly complex form of language production—the literary work of art. His approach is therefore not tainted by a desire to reject ordinary language or to purify language by removing its irrationality. On the contrary, Ingarden attaches great importance to the finest possible articulation between the ‘rational’ and ‘irrational’ components of language.

O. Malherbe (*) Faculté de Philosophie et Sciences sociales, Université Libre de Bruxelles, Brussels, Belgium © The Author(s) 2021 A. Dewalque et al. (eds.), Philosophy of Language in the Brentano School, History of Analytic Philosophy, https://doi.org/10.1007/978-3-030-52211-7_8

169

170 

O. Malherbe

That said, The Literary Work of Art, undoubtedly one of Ingarden’s master works, is also one of his earliest (1931), and one of the few that were not revised and enhanced during the 60s (as were for example, Ingarden’s other major books: Der Streit um die Existenz der Welt or Vom Erkennen des literarischen Kunstwerk—translated in [Ingarden 2016] and [Ingarden 1973]).1 It is also, to some extent, aporetic. Towards the end of the book, Ingarden suddenly reveals—but only briefly describes—a brand new and quite mysterious concept in order to account for another, second type of ‘meaning’, which is certainly not the mere meaning of language, but which nevertheless finds itself carried by language: the metaphysical quality (Ingarden 1965, §§ 47–50). As such, though this early masterpiece offers the first complete elucidation of Ingarden’s theory of meaning, his subsequent analyses, particularly in aesthetics and axiology, offer major developments of this other realm of ‘meaning’ that was barely touched upon in The Literary Work of Art and seems essential to get a full account of the meaning conveyed by language, both in its rational and irrational components. The goal of this paper will be to try and give such a ‘full account’ of Ingarden’s theory of meaning(s). To this end, I will especially propose to draw a sharp distinction between two concepts of ‘meaning’ in Ingarden’s theory. The first will prove to encompass the rational core of meaning, while the second will prove to be of an axiological nature and will account for the more irrational and emotional components of meaning. This journey will also allow us to underline a few other salient features of Ingarden’s conception, notably his theory of reference and his understanding of language as an intentional object. Another major feature of Ingarden’s theory of language is its foothold in ontology. I will therefore start with some ontological considerations that will prove indispensable later on, as they will lead us to Ingarden’s conception of language as a polystratic, intentional entity (Sect. 2). I will then analyse these various strata and focus on the rational components of meaning (Sect. 3). Finally, I will analyse this ‘second’ meaning and show its axiological nature (Sect. 4), before summarising my results (Sect. 5).

8  Meaning(s) in Roman Ingarden’s Philosophy of Language 

2

171

Language as a Polystratic Intentional Entity

It was Ingarden’s interest in ontology that first led him to the question of the literary work of art as the paramount example of a purely intentional object (Ingarden 1965, xiii–xiv). We must therefore briefly address this ontological frame in order to understand his theory of meaning, notably as Ingarden develops a certain parallelism between meaning and being. Ingarden’s tripartite ontology is divided between material ontology (the qualities of the object in the broadest sense, including what traditionally counts as form: colours, shapes, textures, and so on), formal ontology (the subject of properties and the being-a-property structure, that transforms mere qualities into properties of an object) and existential ontology (the ways of being). Ingarden indeed distinguishes between several ways of being. Three of them are of use for us: the real, the purely intentional and the ideal. Without diving too deeply into the details of Ingarden’s ontological pluralism, we should remember that (a) a purely intentional being finds its foundation in acts of consciousness (it is heteronomous while real and ideal beings are autonomous). This entails that (b) intentional entities are, generally speaking, incomplete (they only carry properties given to them by the consciousness and thus ‘lack’ an infinite number of properties) (see Richard, in this volume) and (c) there is a distinction between original and derived purely intentional objects (Ingarden 1965, 122–123): the former are objects created by a monosubjective consciousness (the unicorn I am currently thinking of ), while the latter have a physical foundation (a real object) in which they are anchored.2 As a result, they (i) generally have fewer properties than original intentional objects but (ii) become intersubjectively accessible for competent observers (e.g. the work of art anchored in a book, but also the acoustic vibrations that we produce to speak). In Ingarden’s perspective, every single cultural object is some kind of derived, purely intentional object (Ingarden 1983, 19) and language is no exception to this. These ontological features indeed imply that the process of hearing/reading a text/speech creates so to speak ‘in the reader’s mind’ a new original intentional object (Ingarden refers to this process as ‘concretisation’ [see 1965, 362]), whose

172 

O. Malherbe

properties (and possibly meaning) are different from those of the text/ speech itself. Ingarden then demonstrates that language is composed of various heterogeneous strata (layers)—it is thus polystratic (on this see Simons 1944). Simply put, a stratum is composed of unified homogenous elements and always maintains organic relations with the other strata. Strictly speaking, language3 is composed of two strata: a) the stratum of linguistic sound formations, that is, the phonic side of language (but not the vocal vibration pertaining to the real world— this is already a purely intentional stratum); b) the stratum of meaning units, that is, the core of linguistic signification. But it also implies other strata to a total ranging from three to five: c) the stratum of represented objectivities, that is, the objects, situations or even world(s) projected by the meaning units; d) the stratum of schematised aspects, that is, the one ‘preparing’ aspects through which we can picture the represented objectivities; e) the stratum of writings, also intentional and not real, that I will not discuss further here.4 Strata (a) to (c) are essential, while stratum (e) is optional,5 and Ingarden explicitly states that stratum (d) may be missing when we cannot get any intuition (hence no aspects) of the represented objectivities (e.g. a round square, a quark’s spin, etc.) (Ingarden 1965, 129). With this general frame now set up, we can further investigate those layers and shall, as announced in our introduction, first focus on the more rational parts of language, which require all four layers but are articulated around the layer of meaning units.

8  Meaning(s) in Roman Ingarden’s Philosophy of Language 

3

173

Meaning, Rationality and Reference

In Ingarden’s view, sound formations (a) and meaning units (b) are two sides of the same coin: without the phonic side, we would not be able to access the layer of meaning; but it is because this phonic material has a meaning that it stabilises and becomes a distinguishable word (a word-­ sound [Wortlaut]) (Ingarden 1965, 30–31), that is to say a ‘typical phonic Gestalt’ (Ingarden 1965, 33) that reaches over the purely ontic sound. This correlation between a word-sound and a meaning is so strong that even perfectly ‘identical’ words (meaning: with the same word-sound or spelling) differ from each other through ‘Gestalt characteristics’ (Gestaltcharaktere) that are neither phonic nor optical but are ‘like strange reflections falling from the meaning on the word-sound’ (Ingarden 1965, 35). Here is a good example: ‘bat’ may be a baseball bat or a nocturnal mammal; even though those words are pronounced in the exact same way, ‘something’ ‘sounds’ (though it is no actual sound) different. Ontologically speaking, every word-sound is an intentional object whose constitutive nature is a Gestalt quality (Ingarden 1965, 41, 58).6 A Gestalt quality is a particular quality that springs through the conjunction (e.g. a human figure) or the succession (e.g. a melody) of its grounding qualities. It thus only manifests itself when those qualities are present. It then surpasses them all and brings something qualitatively new in being (Ingarden 2016, 60–61). Some Gestalts—and this is the case here—allow variations in their grounding qualities. This means that every word has a typical quality that can manifest itself despite sometimes important variations (to which the Gestalt is impervious) in the concrete phonic material. As we shall see in Sect. 4.2, these variations may be linguistically insignificant, or may convey some important information. The layer-­ structure of the sound formations implies that the word-sounds unite in some way when put together. As a consequence, a whole set of new phonic phenomena appear: rhythm, tempo, even melodies which also constitute new Gestalt qualities (Ingarden 1965, 45, 60). The layer of meaning units (b) is Ingarden’s first and most thorough analysis of the structure of meaning.7 It is, of course, impossible to reproduce it here in its full extent and richness, so we will focus on the

174 

O. Malherbe

meaning of isolated nouns. Ingarden distinguishes four to five interdependent components in their meaning (Ingarden 1965, 62–63): 1. The intentional directional factor, 2. The material content, 3. The formal content, 4. The moment of existential characterisation, 5. The moment of existential position (optional). When the noun is entangled in a sentence or in a text, it gains a sixth, apophantic-syntactic meaning component (Ingarden 1965, 63; 1962, 214) Components (2) to (4) are the easiest to grasp as they directly relate to Ingarden’s ontological frame discussed earlier: a noun’s meaning determines the form (3) of the objectuality it deploys (object for ‘table’, process for ‘a run’, event for ‘a collision’, etc.), its material content (2) (that can enclose constant and variable qualities: every table must have a colour, but this colour is variable) and its mode of being, that is, its existential characterisation (4): real for, let’s say, an apple or Harry Potter (the character; not the book) or ideal for a mathematical triangle. The existential position (5) is optional and closely related to the existential characterisation moment. As said earlier, Harry Potter is intentionally aimed at as a real being. But as we know, that he does not ‘really’ exist, he lacks the moment of existential position (he may nevertheless have a quasi-position allowing the ‘make believe’ to kick in). Existential position can be modalised: questions do not have the same position as assertions, and position also differs amongst affirmations—judgements have a strong existential position; quasi-judgements (sentences from a work of fiction) have a quasi-position; mere assumptions have no position at all. Finally, the intentional directional factor (1) is the moment that refers the noun to an object. The intentional factor may be of several kinds. It may be ‘single-rayed’ or ‘multi-rayed’ (‘one’ or ‘three’ apples); it may be constant and actual (‘Olivier Malherbe’, ‘the table [in front of me]’), or variable and potential (‘a table’), and so on. (Ingarden 1965, 63–65, Falk 1981, 40–41). Independently of those variations, the referred-to object is always the purely intentional one, projected by the meaning, even though

8  Meaning(s) in Roman Ingarden’s Philosophy of Language 

175

in some circumstances—language uses that intend to refer to the actual world—the intentional factor is so shaped that it is transparent (durchsichtig). Therefore we do not notice the existence of a purely intentional object and focus directly on the real object (Ingarden 1976, 65). Building on those five meaning components, Ingarden then analyses the meaning of other types of words. For example, the formal and material content of an adjective is always that of a property, that is, a quality accruing to an object. When it is linked with a noun, its intentional directional factor changes and attaches itself to this very noun to attribute this property to it (‘a red apple’). Deictics mainly have an intentional directional factor; prepositions have a specific material-content: they attribute relational features between two totally undetermined entities, etc.8 In linguistic formations, the words combine and gain a sixth, ‘apophantic-­ syntactic’ meaning component. There is some kind of decompartimentalisation of the meaning of a word which is united with the meanings of the other words from the sentence to form a higher meaning unit (Sinneinheit). The different components of the meaning mix according to the functions of the different words: the intentional factor of the adjective aims at the noun it accompanies, the orientation factor of the noun in subject position ‘prepares’ this word in such a way that the orientation factor of the verb can ‘find’ it and define it as a subject that performs the action, and so on. In the end, every sentence projects a state of affairs (Ingarden 1965, 113) which constitutes the purely intentional correlate of the sentence. Of course, this uniting of the meaning does not stop at the sentence but continues and complexifies at the level of text or speech. The projection of this state of affairs fulfils the core (but not sole) function of language: that of representation (Darstellung), which leads to the projection of represented objectivities. This is certainly one of Ingarden’s most crucial theses: every sentence deploys a purely intentional correlate (Ingarden 1965, 113). The question of meaning is thus (almost) entirely severed from the question of reference, which only concerns some uses of the language,9 i.e. mainly linguistic formations with a claim to truthfulness, or, as Ingarden calls them: judgements. Ingarden understands truth as adequacy between the intentional state of affairs projected by the sentence

176 

O. Malherbe

and the state of affairs being described (Ingarden 1965, 170–171). In those ‘transparent’ uses of language, the judgement’s intentional state of affairs only becomes apparent when a hiatus appears between it and the state of affairs it was supposed to describe (‘My laptop is on the table’, except it isn’t). Finally, it should also be noted that the parallelism here sketched between language and meaning (particularly through components [2] to [4]) is far from absolute. Ingarden does not adhere to a bundle theory of being in which every object is composed of a set of properties with the same importance (Chrudzimski 2016, 214–217). An object’s matter is highly structured, and distinctions are to be made (ontologically, if not always linguistically) between mere properties, constitutive nature, quasi-­ nature, relational features, essence, etc. Most Gestalt qualities are also incredibly difficult (or impossible) to refer to with words as they are often of a singular nature. The layer of meaning units naturally leads us to the layer of represented objectivities (c) projected by the various linguistic formations. I will not elaborate this at length, but it is worth emphasising that (i) these intentional objects (and situations, and worlds) are essentially incomplete and possess countless so-called spots of indeterminacy, (ii) these intentional objects are not bound by logical laws: they may violate the law of excluded middle (see Seron 2016), the law of non-contradiction (Ingarden 2016, 503—in this case it can often make them unpresentable), etc., (iii) the phenomenological importance of this stratum varies a great deal according to the type of language used—it is on the forefront in works of fiction (especially when they unfold in another universe, such as The Lord of the Rings), and completely transparent (but still present) in works of science (unless some error appears to us) or language aimed at practical goals. Finally, (iv) this stratum is paramount to another meaning of ‘meaning’, one that is beyond pure rationality and that we will analyse below. All those represented objectivities are in themselves purely unintuitive and empty. To receive intuitiveness, they must be given to us through aspects (Husserlian Abschattungen), which form the last layer (d). Of course, ‘real’ aspects can only belong to an actual perceiving consciousness, but language can contain ‘schematised aspects’ (Ingarden 1965, 271 f.) that can be fulfilled by the interlocutor (Mitscherling 1997, 138).

8  Meaning(s) in Roman Ingarden’s Philosophy of Language 

4

177

Axiological Meaning

We have thus far focused on the ‘rational’ side of Ingarden’s meaning theory, analysing meaning as an interlocked network of conceptually graspable moments (formal, material, etc.) that are contextually unified to form meaning units and deploy represented objectivities. That said, this rational side of meaning does not exhaust Ingarden’s analysis of language. As no literary work of art can be thought of solely as a purely rational construction, Ingarden devoted serious analysis to more irrational parts of language, linking them mostly with the three other layers: phonic formation, represented objectivities and aspects. This is of course one of the advantages of Ingarden’s theory: even if something cannot be explained through the meaning unit layer alone, we still have three to four other layers to work with. We will now tackle this question of a ‘second meaning’ of meaning (hereafter ‘meaning2’) that we announced in our introduction and that Ingarden links to metaphysical qualities. We will try and show that this meaning2 is of an axiological nature (1). We will then sustain this thesis through an analysis of the phonic formations layer (2) and through some concrete examples of value-shaped languages according to Ingarden (3).

4.1

 he Layer of Represented Objectivities: T Metaphysical Qualities and Meaning2

The concept of ‘metaphysical qualities’ aims to address some peculiar qualities that sometimes spring from an emotionally intense life-­situation. When they do, they completely transform this situation and shed a totally new light both on it and on our lives. As he puts it: There are qualities […] such as the sublime, the tragic, the terrible, the moving, the elusive, the demonic, the sacred, the sinful, the sad, the indescribable clarity of joy, but also the grotesque, the delightful, the light, the serenity, etc. [They] manifest themselves in complex and very different situations and events, like a specific atmosphere which hovers over men and things caught in these situations, and which nevertheless penetrates and

178 

O. Malherbe

transfigures all by its light. […] Whatever the quality of this atmosphere is, terrifying or enchanting to the ecstatic self-oblivion, it is that which tears itself away from everyday greyness like the radiance of a bright range of colours, and which makes this event a summit of life […]. These are these ‘metaphysical’ qualities—as we call them—revealed from time to time, that give life an experiential value [Erlebenswert]; and after these concrete revelations, all our acts and actions are charged with a nostalgia that lives in us and drives us, whether we like it or not. Their revelation represents the summits and the ultimate depths of being. (Ingarden 1965, 310–311)

Such qualities also ‘unseal a “more profound” meaning [Sinn] of life and being’ (Ingarden 1965, 312), which prompts Ingarden to immediately share his discomfort regarding this peculiar use of the term ‘meaning’ in such a context: In this case the word ‘meaning’ [Sinn] obviously has nothing in common with the meaning [Bedeutung] we are thinking of when we talk about the meaning [Sinn] of a sentence. It is also inappropriate for the reason that the word ‘meaning’ [Sinn] usually means something rationally seisable. I would prefer to avoid this term here. But I can’t find any other that would be more appropriate. (Ingarden 1965, 312, note)

Ingarden’s account of metaphysical qualities does not go much further in The Literary Work of Art. He simply proceeds to explain that successful literary works of art are able to manifest such metaphysical qualities, though in a tuned down (and now aesthetically valuable) version (Ingarden 1965: 313f.; Stróżewski 2012: 28). In order to obtain this result, all the strata have to work together to build the required emotional atmosphere (it is enough to picture Oedipus at Colonus written as a police report, peppered with spelling mistakes, to realise that no metaphysical quality would ever appear). Despite Ingarden’s reluctant use of the word ‘meaning’, this concept entails that language can also convey, through its layer of represented objects, ‘another’, irrational, ‘meaning’, that we now need to cover. One can find other occurrences of this ‘other’ meaning when Ingarden addresses the problematic of values. He indeed states that it is through the contact with values that our life becomes ‘meaningful’ (1983, 18), or

8  Meaning(s) in Roman Ingarden’s Philosophy of Language 

179

that values ‘raise’ their bearer ‘above all other objectualities which merely exist but do not “mean” [bedeuten] anything’ (1969, 116). This fragmentary trail regarding meaning2 then gains momentum if we acknowledge that metaphysical qualities in works of art constitute an aesthetic value and remember Ingarden’s statement that such metaphysical qualities (appearing in the real world) have a special ‘experiential value’ (Ingarden 1965, 311). Value is one of the core concepts of Ingarden’s late philosophy. It will be enough for our topic to state that (a) Ingarden advocates a material conception of values, meaning that they ground themselves in some of their concrete bearer’s qualities (this beauty is rooted in that painting’s colours, the usefulness of this tool in its shape, etc.), (b) they are either positively or negatively connoted and, therefore, carry a particular ought-­ to-(not)-be, (c) we feel called to realise/destroy value bearers, and values are thus the ratio of most major human acts, (d) we usually have a strong emotional reaction to the presence of values, (e) multiple types of values are to be distinguished: aesthetic, ethic, social, utility, economic, religious, vital, and so on (Ingarden 1969, 98–99; 2016, 240–242), (f ) an important distinction is to be made between unconditional values (which are ends in themselves) and conditional values (that are only valuable in that they can contribute to the apparition of an unconditional value) and, last but not least, (g) the whole cultural world is constituted of purely intentional objectualities (works of art, institutions, laws, etc.) created by humans and imbued with various values, and its first purpose is to allow their manifestation, thus fulfilling our deepest needs and allowing us to truly become human (Ingarden 1983, 30). As we can see, the purely intentional world projected by language can thus present objects in such a way that they exhibit the same (or at least similar) metaphysical qualities and values as the ones they bear in our cultural world.10 Therefore the layer of represented objectivities is able, with the help of the other strata, to present and load itself with another ‘meaning’, of a non-linguistic but seemingly axiological nature. Of course, one can see no reason to limit this potentiality to those values bearing metaphysical qualities alone. Language can of course present objectivities loaded with moral, cultural, aesthetic, cognitive values, etc. (Ingarden 1973, 294, note). In this sense, it seems reasonable to think that language

180 

O. Malherbe

itself—and the very words used to depict those valuable objectivities— also gains some values (and thus an emotional load, as a consequence of [d]) following their use in a concrete human community (or sub-­ community, or even individual use): to Romeo, the very sound of ‘Juliet’ probably bore a reflection of her beauty, ‘capitalism’ feels like an ignominious word-sound to left-inclined thinkers, ‘violin’ like an harmonious word (except maybe to my friend whose neighbour plays it poorly every night), and so on. This tentative relation we establish between the ‘second’ meaning and the question of values may then shed a new light on some of Ingarden’s thinking regarding the layer of phonic formations.

4.2

Word-Sounds and the Life of Language

The layer of phonic formations plays an important role in Ingarden’s account of irrational and contextual meaning elements (Ingarden 1965, 60), both through the word-sound itself and through other phonic meaningful components. As mentioned, word-sounds are Gestalt qualities that are impervious to some variations of their concrete phonic material. There is thus ‘room’ in those variations either to improve the representation by additional features, or to fulfil other functions than that of representation, without impairing it. Intonation is an example of additional representative elements, as it allows the speaker to indicate whether the sentence is a question, or an order, and thus to provoke according modifications in the existential position (5) of the purely intentional correlate of the sentence (see above). The concrete phonic material can also fulfil a function of expressing the speaker’s psychical states through manifestation-qualities: the word-sound and sentences tone utterance may disclose those inner states to the interlocutor in an ‘intuitive way, without any reasoning’ (Ingarden 1965, 41).11 Aside from those possibilities given by extra word-sound phonic qualities, the very word-sound is also extremely significant to understand one’s choice of word in a concrete use of language, which will bring us back to the question of values. It is only common sense to acknowledge that two words or sentences, even perfectly synonymous, may be totally

8  Meaning(s) in Roman Ingarden’s Philosophy of Language 

181

unsubstitutable in their use in various contexts. As Ingarden states, various qualities are often attached to the word-sounds themselves. As a result, some of them seem to be particularly ‘fit’ to express some meanings or to accomplish certain functions. Profanities offer here clear examples (Ingarden 1965, 41) as they seem perfectly convenient to us in some colloquial contexts and totally inappropriate in others where we would opt for less ‘graphic’ options. English language is of help here as ‘graphic’ exactly expresses Ingarden’s thinking: vivid word-sounds have a deep connection with the layer of aspects (d): they make us vividly ‘see’ (intuit) the object or state-of-affairs. On the other hand, some words may be used to minimise this vividness, be it to ensure the clarity of a new philosophical concept (better invent a new word with a clean slate than use an old one with countless implications), for political reasons (one speaks of ‘surgical strikes’ and not of ‘gruesome bombing’), and so on. Finally, some word-­ sounds (either due to their own qualitative sound or together with their meaning [Ingarden 1965, 42]) bear specific axiological potentialities: they are ‘beautiful’, ‘simple’, ‘pathetic’, ‘solemn’, etc. Ingarden explains these phenomena through language life and history: such qualities accrue to word-sounds during their continuous development, entangled with ‘the history of the whole life of a human community’ (Ingarden 1965, 39). By remembering that Ingarden sees human communities as the co-­ creation of a cultural world saturated with values, we can begin to shed a new light on this ‘fitting’ character of some words: we use them according to the axiological features they bear, as they sink back all the way from the layers of represented objectivities to the word-sound itself (this also explains why the axiological features of a word-sound often depend both on the meaning and on the sound itself ).

4.3

 oncrete Examples of Value-Shaped Language: C Art, Science and Ordinary Language

We have tried to sustain the thesis according to which (a) meaning2 is of an axiological nature and is brought back from the various objects entangled in our cultural world to the language that represents them, and (b) words are chosen in their concrete use according to the axiological

182 

O. Malherbe

features they bear. We shall now analyse a few examples of such value-­ shaped uses of language. Without getting here into the considerable analysis Ingarden devotes to artistic language (Ingarden 1965, 1973), we can state that the principle of organisation of such a language can be easily grasped: it is entirely organised in such a way as to favour the appearance of artistic/aesthetic values,12 including metaphysical qualities. Ingarden emphasises that, in a great work of art, aesthetic values are grounded in each strata’s qualities and together form a harmonious polyphony (Ingarden 1965, 395–400; 1991, 139). According to his material conception of values, three stages are to be distinguished in the qualitative set of a value: the value quality, grounded in aesthetically valent qualities, themselves grounded in aesthetically neutral qualities.13 Valent qualities are here the cornerstone as, depending on their co-appearance or succession, they may give birth to positive or negative values. This entails that not only are the aesthetic values of the literary work of art to be grounded in each stratum, but those very strata are also to be arranged down to their most intricate details in such a way as to ensure the presence of the myriad of founding qualities (valent and then neutral) required for the value qualities’ manifestation. A literary work of art, should it succeed, must thus articulate together and harmonise ‘beautiful’ (this is of course an oversimplification) words, well arranged meaning units, interesting represented objects, and skilfully prepared aspects. Correlatively, these visible marks on the language will lead the reader into the aesthetic attitude (Ingarden 1973, 196), that is to say, a form of consciousness in which they will prove to be particularly sensitive to these qualities and which will influence their understanding of speech (notably, interpretation of sentences as quasi-­ judgements, see above). Scientific language on its part is of a totally different kind and shape. These differences find their ratio in the scientific values this language embodies (Ingarden 1991, 132). The scientific core value is objectivity, which translates, in language, as truthfulness (Ingarden 1996, 458), that is, as we know, the adequacy of the judgement’s state of affairs and the state of affairs belonging to the objectuality to be investigated (be it real, ideal or even purely intentional as long it is independent from the very judgement). Other scientific values are also to be considered: certitude (is

8  Meaning(s) in Roman Ingarden’s Philosophy of Language 

183

this judgement true by mere chance?), completeness, etc. (Ingarden 1996, 458). The goal of a scientific work is to secure and transmit scientific knowledge (Ingarden 1965, 350; 1991, 132), that is, to allow the reader to reconstruct as clearly as possible the state of affairs of its layer of meaning units (and, thus, the represented objectivities it projects). The language of science should therefore be arranged accordingly: it should be straightforward, should avoid aesthetic effects and metaphysical qualities (we all know the dangers of rhetoric), be devoid of ambiguity; it should use to the fullest extent of its possibilities univocal words (even create new ones), hopefully not be too boring (though this is not a core value), and so on. Again, this particular treatment of language will trigger in the receiver a scientific (or investigative) attitude (Ingarden 1973, 233), which will make it particularly sensitive to the knowledge values to which the text claims and will command a completely different behaviour (critical distance, interpretation of sentences as real judgements, etc.) The same shaping of the language through values can also be reconstructed for ordinary speech in its crudest sense from Ingarden’s scattered reflections about it. Ingarden speaks of the latter as the one which ‘we use in practical affairs to achieve something with other men in our common real world’ (Ingarden 1973, 65). In this sense, it is the direct counterpart of the natural or practical attitude ‘in which a person sets outs to change or effect something in the world’ (Ingarden 1973, 173, 183). Language being used as a tool, as a mean to other ends (ends that should, on their side, bear some unconditional values), it feels safe to assume it would be organised according to various utility values.14 Ingarden sketches this utilitarian attitude as one where we only pay attention to properties of the objectualities that are ‘of use’ for our goals (Ingarden 1997, 153; 1996, 430), while the others fade away. In this attitude we are also, much like in the scientific one, fixed on real objects (Ingarden 1976, 60–62), meaning the intentional factor will be transparent (unless the goal requires to be pictured first). As we can see, these three examples share numerous features: (1) they are value-driven, (2) correlatively, the language crafter finds himself in a particular attitude that makes him more receptive to special types of values (3), as a result, language itself is shaped (and words are chosen) in a

184 

O. Malherbe

very different way in order to allow some values to be enshrined in it, either as an end, or a mere mean to other ends.

5

Conclusion

We began by presenting Ingarden’s theory of meaning, focusing on its rational elements. We were then led to conceive language as an intentional entity composed of several finely articulated strata and organised around the layer of meaning units. Each word was revealed to be a two-­ sided entity, both a Gestalt quality of sound nature and a bearer of meaning. This meaning could itself be understood as an interconnected network composed of various elements (intentional orientation factor, a material and formal content, etc.). We then saw how this first conception of meaning offers a particularly fine account of the modifications that a word’s meaning undergoes in contexts through its decompartimentalisation. This decompartimentalisation then leads to the projection, by any sentence (and this has major implications on the question of reference), of a purely intentional correlate: the state of affairs. Words and language formations also implied the creation of intentional represented objectivities. Though fruitful, this conception of meaning was however still insufficient to account for the irrational components of meaning and language. It is here that the layered conception of language defended by our author proved to be particularly favourable, in that it provides so many ‘places’ to accommodate multiple additional dimensions to the language. We therefore tried to tackle the question of this ‘other’, ‘irrational’ meaning2 conveyed by language, notably through metaphysical qualities. We proposed to see in these metaphysical qualities an example, certainly significant, but not unique, of the importation of the axiological ‘meaning’ of mundane objects in language and tried to sustain our hypothesis by focusing first on the layer of phonic formations and then by presenting some concrete examples of value-shaped languages. It thus appears that meaning2 is of an axiological nature. The cultural world is, in Ingarden’s conception, saturated with endless value-bearing objects—those are the very reason of its existence. As values—be they

8  Meaning(s) in Roman Ingarden’s Philosophy of Language 

185

aesthetic, cognitive, ethics, and so on—are (a) the very features that give meaning to things themselves (as they suddenly ought to [not] exist) and (b) grounded in their bearer’s qualities, it feels only normal to have the language, through some backlash effect on the words themselves, loading itself with some reflection of those values. Language formations themselves, as cultural products, are also potential bearers of values: aesthetic values, cognitive, utilitarian (and we could complete the list: pleasure, economics, social, etc.). And to manifest those multiple values, some words are better suited than others due to the intrinsic (yet often historic) axiologically significant qualities of any of their layers (to oversimplify: ‘beautiful’ words for art, ‘precise’ for science, etc.; but also ordinary words referring to a beautiful or value-endowed object), with infinite and ever-­ adjusting loops at play.

Notes 1. The 1965 version (Ingarden 1965) is a mere reedition. 2. This anchoring is itself of an intentional nature. See (Ingarden 2016, 625). 3. It should be noted that Ingarden mainly focuses on ‘linguistic formations’ (Sprachgebilde): words, sentences, texts or speeches. He does not analyse language as a system in great detail (some information may be found in [Ingarden 1962, 2016, 610–613]). We shall do the same here. 4. Whether writing may truly form a fifth layer in an Ingardenian sense is still an open question. Ingarden only makes a passing reference to the question of writing. He states that written words also have a Gestalt quality (Ingarden 1965, 31, note) and that they are deeply intertwined with the word-sound as there is some kind of oscillation between the two of them: to some extent we ‘see’ the written word while speaking or thinking, and ‘hear’ it while writing. Furthermore, writing connects with the other layers. 5. Unless in some artificial symbolic languages where the sound formations are optional or nonexistent (Ingarden 1991, 146). 6. On Gestalt qualities and their significance in the various fields of Ingarden’s thought, see (Malherbe 2016). 7. I will not present here the distinction made by Ingarden in the Literary Work of Art between meaning and ideal concept (Ingarden 1965, 88,

186 

O. Malherbe

386 f.), as I think he later abandoned it, as shown by a note from the Polish version of the book (Ingarden 1960, 14). Even though in (1965), Ingarden grounded the possibility of interpersonal understanding in those very ideal concepts, in one of his latest books (Ingarden 1973, 350), he proposes a more subtle possibility: the shared experience of real objects on the one hand, and eidetic relation to pure qualities and ideal objects on the other hand. 8. Verbs have a profoundly different meaning-structure, entangled with Ingarden’s ontological concept of ‘process’, but I do not have the luxury to elaborate on that here. 9. On the other hand, some linguistic formations are happy to create new intentional objects in the real world (e.g. the consecration of a church brings into being a purely intentional object anchored in a pile of bricks, see [Ingarden 1989: 259]). 10. Ontologically speaking, this phenomenon is grounded in the fact that the same ideal qualities (pure potentialities) are incarnated, resp. concretised, in real, resp. purely intentional, beings. 11. Both tone and intonation also form Gestalt qualities (Ingarden 1965, 53). 12. For the sake of simplicity, I disregard here Ingarden’s distinction between artistic and aesthetic values. 13. As I have tried to show elsewhere (Malherbe 2016), Gestalt qualities play a significant role in the building of each stage on top of the former one. 14. Ingarden gives as examples political pamphlets or newspapers articles aiming ‘to the realisation of determined technical, financial or economic tasks’ (Ingarden 1976, 64).

References Chrudzimski, Arkadiusz. 2016. Ingarden on Substance. In Forme(s) et modes d’être, l’ontologie de Roman Ingarden, ed. Olivier Malherbe and Sébastien Richard, 211–227. Brussels: Peter Lang. Falk, Eugene Hannes. 1981. The Poetics of Roman Ingarden. Chapel Hill: The University of North Carolina Press. Ingarden, Roman. 1960. O dziele literackim; badania z pogranicza ontologii, teorii jezyka i filozofii, Trans. Maria Turowicz, Warszawa: PWN. ———. 1962. Le mot comme élément d’une langue. Logique et Analyse 20: 212–216.

8  Meaning(s) in Roman Ingarden’s Philosophy of Language 

187

———. 1965. Das literarische Kunstwerk. Tübingen: Max Niemeyer. ———. 1969. Erlebnis, Kunstwerk und Wert. Tübingen: Max Niemeyer. ———. 1973. The Cognition of the Literary Work of Art Man and Value. Trans. Ruth A. Crowley et al. Evanston: Northwestern University Press. ———. 1976. Gegenstand und Aufgaben der Literaturwissenschaft. Tübingen: Max Niemeyer Verlag. ———. 1983. Man and Value. Trans. Arthur Szylewicz. Munich: Philosophia Verlag. ———. 1989. Ontology of the Work of Art. Trans. Raymond Meyer and John T. Goldthwaith. Athens: Ohio University Press. ———. 1991. On Translations. Trans. Jolanta Wawrzycka. Analecta Husserliana 33: 139–192. ———. 1996. Zur Grundlegung der Erkenntnistheorie. Teil I: Das Werk. In Gesammelte Werke, ed. Włodzimierz Galewicz, vol. 7. Tübingen: Max Niemeyer. ———. 1997. Zum Problem der Objektivität der äußeren Wahrnehmung. In Gesammelte Werke, Vol. 8, Ed. and Trans. Włodzimierz Galewicz, 38–166. Tübingen: Max Niemeyer. ———. 2016. Controversy Over the Existence of the World. Vol. 2. Trans. Arthur Szylewicz. Frankfurt am Main: Peter Lang. Malherbe, Olivier. 2016. Quelques avatars de la Gestalt dans la philosophie d’Ingarden. In Forme(s) et modes d’être, l’ontologie de Roman Ingarden, ed. Olivier Malherbe and Sébastien Richard, 163–196. Brussels: Peter Lang. Mitscherling, Jeff. 1997. Roman Ingarden’s Ontology and Aesthetics. Ottawa: University of Ottawa Press. Seron, Denis. 2016. La violation du tiers exclu comme critère d’intentionnalité. In Forme(s) et modes d’être, l’ontologie de Roman Ingarden, ed. Olivier Malherbe and Sébastien Richard, 197–210. Brussels: Peter Lang. Simons, Peter. 1944. Strata in Ingarden’s Ontology. In Kunst und Ontologie, ed. Włodzimierz Galewicz et al., 119–140. Amsterdam: Rodopi. Stróżewski, Władysłąw. 2012. Ingarden’s Ontology and its Role in his Aesthetics. In Roman Ingarden: ontologie, esthétique, fiction, ed. Jean-Marie Schaeffer and Christian Potocki, 17–35. Paris: Editions des archives contemporaines.

9 Overcoming Psychologism: Twardowski on Actions and Products Denis Fisette

This paper is about the topic of psychologism in the work of Kazimierz Twardowski and my aim is to revisit this important issue in light of recent publications from Twardowski and on his works.1 Twardowski’s effort to overcome psychologism constitutes one of the driving forces of his thought (see Schaar 2016, 87) and it is the main topic of a talk that he delivered in the Philosophical Society at the University of Vienna in 1914 under the title “Funktionen und Gebilde” and better known by the English version: “Actions and Products. Some Remarks from the Borderline of Psychology, Grammar and Logic” (Twardowski 1999b). This is confirmed by Roman Ingarden in his classical paper on Twardowski in which he points out that the true meaning of this writing lies in the problem of psychologism (Ingarden 1948, 28–29). According to Ingarden, Twardowski’s solution to the problem of psychologism rests precisely on the distinction between action and product. I propose to

D. Fisette (*) Département de philosophie, Université du Québec à Montréal, Montréal, QC, Canada e-mail: [email protected] © The Author(s) 2021 A. Dewalque et al. (eds.), Philosophy of Language in the Brentano School, History of Analytic Philosophy, https://doi.org/10.1007/978-3-030-52211-7_9

189

190 

D. Fisette

follow the path suggested by Ingarden and to carefully examine Twardowski’s “Actions and Products” and his way out of psychologism.

1

 he Genesis of Psychologism T in the Young Twardowski’s Work

Twardowski’s autobiography provides insights into the genesis of his treatment of the problem of psychologism in his early work and the impact of Husserl’s criticism in his Logical Investigations. Twardowski points out that his initial position on that issue was that of Brentano, a position which he also attributes to Marty, and that merely consisted in conceiving of psychology as the ‘fundamental philosophical science’ (die philosophische Grundwissenschaft). This is the position he explicitly advocated in “Psychology vs. Physiology and Philosophy” and implicitly in his earlier works, especially in his 1894 book On the Content and Object of Presentations. However, unlike Brentano who never took seriously Husserl’s objections, Twardowski seems to recognise that it was Husserl’s objections against logical psychologism which forced him to abandon what he considered his own psychologism. The most substantial writing in which Twardowski argues against psychologism dates back from his 1908–1909 lectures “Psychology of thinking” in which he examines three arguments against logical psychologism, including the one based on the distinction between functions and products, which he systematically works out in “Actions and Products”. Let’s first have a quick look at Twardowski’s paper “Psychology vs. Physiology and Philosophy” in which, according to his intellectual autobiography, he advocates a form of psychologism. The kind of psychologism that he opposes to what he calls ‘metaphysicism’ in § 7 of this paper boils down to the idea that psychology acquires the status of first philosophy, i.e. becomes the cornerstone of philosophy as a whole. He criticises the Neokantian philosophers from Bade, namely Wilhelm Windelband, because of the sharp division they maintain between psychology and philosophy, and because they assimilate psychology to natural sciences. He sees in descriptive psychology a form of reaction to ‘metaphysicism’, i.e.

9  Overcoming Psychologism: Twardowski on Actions… 

191

metaphysics as first philosophy in the traditional sense of the term (Twardowski 1999e, 57). Accordingly, the revolution that occurred in philosophy in the study of psychology as of ethics and aesthetics, for example, consists in the way of conceiving and approaching the subject matter of these disciplines. One can think of the definition of psychology as a science of the soul which Brentano criticises at the very beginning of his Psychology (Brentano 1995, 2 f.) and which he opposes to psychology understood as a science of mental phenomena whose task primarily consists in the analysis and description of thought and conscious experience in general. Twardowski justifies the privileged philosophical status granted to descriptive psychology by emphasising the role of internal experience as a source of our knowledge of psychical life and because it provides its method and subject matter to philosophy (Twardowski 1999a, 59). Worth mentioning is Twardowski’s reference to Stumpf ’s article ‘Psychology and Theory of Knowledge’ (Stumpf 1891) which represents the first systematic study on the topic of psychologism and which has been instrumental in Husserl’s views on psychologism in Prolegomena to pure logic (see Fisette 2015a). We shall see that Twardowski’s own definition of psychologism is not foreign to that of Stumpf (see Stumpf 1891, 468–469).

2

 wardowski’s Picture Theory of Meaning T and Husserl’s Criticism in the Logical Investigations

In his autobiography, Twardowski recognises the major influence he received from Husserl’s criticism in Logical Investigations, but he was unaware of the existence of Husserl’s earlier writings in which he examines critically Twardowski’s On the Content and Object of Presentations. The first is the well-known manuscript “Intentionale Gegenstände” written in 1894 in response to his reception of Twardowski’s work (Husserl 1990–1991); the second piece is Husserl’s review of Twardowski’s book which has been posthumously published (Husserl 1994a; see Cavallin 1997; Fisette 2003). The two main objections which are of particular

192 

D. Fisette

interest for this study relate to Twardowski’s theory of intentionality and his psychologizing conception of meaning and intentional content. These two objections are clearly formulated in § 45 of the fifth Investigation entitled “The Presentational Content” in which Husserl criticises Twardowski for not distinguishing, within the content of an act, logical meaning (intentional content of a judgement) and sensory content (i.e. sensations and images) (Husserl 1982b, 175; 1994a, 389). Husserl’s main target is Twardowski’s notion of content understood in his Hauptwerk as a picture, and Twardowski uses the case of painting (Bild) to illustrate the way in which the content and object of an act are articulated: As is well known, one says that the painter paints a picture [Bild], but also that he paints a landscape. One and the same activity of the painter is directed toward two objects; the result of the activity is only one. After the painter has finished the painting of the picture and of the landscape, respectively, he has before him a painted picture as well as a painted landscape. The picture is painted; it is neither engraved, nor etched, etc.; it is a painted, real picture. The landscape, too, is painted, but it is not a real landscape, only a ‘painted one’. The painted picture and the painted landscape are in truth only one; for the picture depicts a landscape, hence it is a painted landscape; the painted landscape is a picture [Bild] of the landscape. (Twardowski 1977, 12)

In order to explain the dual meaning of the word ‘painted’ in this excerpt both as an external object and as an internal content (or object), Twardowski uses the distinction between modifying and attributive determination which he generally applies to the meaning of a term, for example, to the meaning of ‘man’ in a ‘good’ man, which is attributive, whereas the determination ‘dead’ is modifying in the expression ‘a dead man’ (see Haller 1982, xiii). Thus, the word ‘painted’ has an attributive meaning when it is applied to the painting (to distinguish it from an engraving, for example), and it has a modifying meaning when applied to a landscape because, of course, the painted landscape is not itself a landscape but precisely a picture or an image (Twardowski 1977, 13). As far as the articulation between the act, its content, and its object is concerned, Twardowski (1977, 16) claims that the painter is conscious of

9  Overcoming Psychologism: Twardowski on Actions… 

193

both the landscape, as a transcendent object, and the painting as an internal object (Twardowski 1977, 14). The painting, like the content or the image, is the means by which the transcendent object, the landscape, is presented. This is apparently its primary function. But the painter could turn his attention toward the painted landscape and thus adopt a ‘modified’ attitude toward the content, and he would then have a secondary object before his mind. To quote Twardowski’s elegant formulation: We shall say of the content that it is thought, presented, in the presentation; we shall say of the object that it is presented through the content of the presentation (or through the presentation). What is presented in a presentation is its content; what is presented through a presentation is its object. (Twardowski 1977, 16)

In other words, the content has a mediating function between the act of presenting and the object presented, it is the medium through which an object is presented. Husserl’s main criticism of Twardowski in Logical Investigations, in addition to Twardowski’s theory of intentionality, pertains to the dual direction of an act, both toward its own content and toward an object. For Twardowski, when one thinks of a landscape, one has two objects before the mind, the real landscape and the mere picture. By recognising that a thought is primarily in contact with transcendent objects and indirectly with one’s own presentations, Twardowski takes a first step towards the overcoming of an immanentist theory of intentionality à la Brentano. But in arguing that one can think of these transcendent objects only by means of immanent internal objects, he keeps a foot firmly anchored in Brentano’s early theory of intentionality.

3

 wardowski’s Self-Criticism in his T Lectures on the Psychology of Thinking

Twardowski’s first public reaction to the objection of psychologism seems to occur in his lectures on the psychology of thinking (Twardowski 2014b, 134–136) in which logic is defined as ‘the science of correct

194 

D. Fisette

thinking’, and thus as ‘a component of psychology’ (Twardowski 2014b, 134). Twardowski explains why this position must be rejected and provides three arguments against it. The first, which he only mentions in these lectures, is that logic ‘emerged and developed independently from psychology’ (Twardowski 2014b, 134). The second argument, which has also been formulated both by Stumpf (1891, 499–500) and Husserl (1982a, 40 f.), is that the laws of psychology are mere ‘generalizations of experiential data in empirical sciences’ and are therefore only probable; on the other hand, scientific and logical laws are ‘apodictic and independent from experience, logic being an a priori science’ (Twardowski 2014b, 135). Now, since the laws of logic thus understood cannot be based on mere empirical generalisations and probable statements of an empirical science such as psychology, logic cannot therefore be based on psychology. Twardowski’s third argument rests on the distinction between mental function (thinking) and its content (thought). This argument is nicely summarised in this excerpt: This is because the object of logic is thought, not thinking; not a mental function but rather its product. […] The difference between thinking and its product is very clear when, among other things, one considers the relationship of both of them to speech; for instance, the meaning of the word ‘the Sun’ is identical with the thought of the Sun; making present or realizing the meaning of the word ‘the Sun’ is identical with thinking of the Sun to oneself. […] These are mental facts, whereas a thought, a conviction, or a judgment is a product of these facts. (Twardowski 2014b, 135–136)

To this distinction between function and product corresponds that between the field of psychology, i.e. thinking in general, and that of logic and the remaining philosophical sciences, i.e. products (Twardowski 2014b, 136). This last argument is further worked out in “Actions and products” which I shall now examine.

9  Overcoming Psychologism: Twardowski on Actions… 

4

195

 reliminary Remarks on “Actions P und Products”

In “Actions and products”, Twardowski resumes his discussion on the content and object of presentations where he left it in 1894, and his main purpose in this study is to revisit in depth the notion of content in light of the notion of Gebilde (product). Unlike his 1894 work in which his analysis focused more specifically on the relationship between content and object of presentations, this study pertains to an act’s relation to its content, i.e. the correlation between functions and their products. However, Twardowski’s analysis in this important paper is not limited to mental phenomena and psychology. The originality of Twardowski’s study consists among other things in extending the function-product correlation to the field of physical and psychophysical phenomena and to language in general. Like most of Brentano’s students, he proposes a new classification of functions and products which, while preserving the hierarchy between the classes of functions and mental products, emphasises what he calls the class of physical and psychophysical functions and products which he examines in the second part of this study. Moreover, according to Twardowski’s indications in § 10 regarding the historical origins of this distinction between functions and products, these concepts are to be understood in the sense that Stumpf uses them in his two 1906 Academy treatises (Stumpf 1906a, b). True, Twardowski associates this distinction with several other philosophers, but careful examination shows that only Stumpf offers a thorough analysis of this distinction, and for the reasons that I mentioned above, it makes no doubt that he is Twardowski’s major source of inspiration in this writing. In “Phenomena and Psychical Functions”, Stumpf argues that mental functions and sensory phenomena, though intimately linked (they can only be distinguished by abstraction), are irreducible to one another and belong to distinct domains. The notion of mental function corresponds to that of act or mental phenomenon, the study of which belongs to descriptive psychology understood in the narrow sense of the science of the psychical functions, while the other elements involved in the accomplishment of an intentional act, from sense perception to voluntary

196 

D. Fisette

actions, namely sensory phenomena, intentional content and relations, all belong to different domains and neutral sciences. This is particularly the case of Gebilde, which Stumpf defines as the necessary correlates of psychical functions (1906a, 28 f.) and which he conceives of as the specific contents of each classes and subclasses of mental functions. To designate the specific content or Gebilde of the functions belonging to the class of judgements, Stumpf uses the concept of state of affairs which is expressed linguistically in ‘subordinate clauses’ (daß-Sätzen) or in the form of ‘substantivised infinitive’. Gebilde also occur both in intellectual and emotional functions, and in the latter case, it is conceived as a value. The study of Gebilde belongs to this new ‘neutral’ science that he called ‘eidology’ (Stumpf 1906b, 32 f.) and it is closely related to Husserl’s pure logic in the Prolegomena.

5

 verview of Twardowski’s “Actions O and Products”

Twardowski’s paper is divided into three parts. In the first part (§§ 1–9), he examines the distinction between function and product through the grammatical distinction between verb and noun like, for example: walking-­walk, running-run, jumping-jump, speaking-speech, thinking-­ thought, judging-judgement, etc. In his analysis of the verbal forms of the type presenting-presentation and judging-judgement, he points out that, in most cases, the verb stands for an activity, a process, or more generally for a function. It marks the dynamic moment of the function while the corresponding name has a static moment. In terms of the grammatical form, the substantive is similar to what Twardowski calls the figura etymologica according to which the judged or represented names are the internal objects of the judging or presenting (Twardowski 2017b, 189). All substantives that are formed on the same root as the corresponding verb are internal objects. Moreover, there is a close relationship between the grammarians’ figura etymologica and the notion of correlation, for example that between the act of perceiving and the perceived, by which the intentionality of psychical experiences is generally

9  Overcoming Psychologism: Twardowski on Actions… 

197

characterised. In this work, Twardowski again uses the distinction between internal and external object by which he meant the immanent content and the transcendent object of an act of presentation. Moreover, this paper confirms that he still maintains the idea of a dual direction of an act which is in no way affected by the modifications brought about to the concept of immanent content through the introduction of that of product (Twardowski 2017b, 176). The notion of ‘geurteilt’ (judged) that he uses in this context refers to the content of an act of judgement that belongs to a class of products that he examines in the other two parts of the paper. This internal object is the result of an act or ‘that which arises owing to, as the result of some action, i.e. by means of that action’ (Twardowski 1999b, 108), and this something is also named a product to indicate that these contents are realised thanks to an act on which they depend and to designate the very process at the origin of the formation of these contents. The second part of the text bears on the classification of functions and products. Twardowski (2017b, 191) proposes in fact a dual classification: the first is based on Brentano’s distinction between physical and psychical phenomena which is applied both to functions and products. In the class of products and physical functions, he further distinguishes those which are purely physical from those which are psychophysical. The second classification only applies to products and it distinguishes the class of enduring from that of non-enduring products. Twardowski maintains that this function-formation correlation can be found in all cases, such as walking, running, or jumping, which involve a physical activity, i.e. in this case, a bodily movement of the agent. Hence the use of the terms ‘action’ and ‘product’ to designate, on the one hand, the action of running, walking and jumping, and, on the other hand, the product of these actions, i.e. the run, walk, and jump. Gebilde or products and physical functions are further distinguished into those which are purely physical and those which are psychophysical. For example, the pair crying-cry belongs to the class of purely physical functions and products when it is a bodily movement, i.e. a mere mechanical reflex, a non-voluntary, or non-intentional movement. But in this case, we are not dealing with an action as such, i.e. an intentional behavior. On the other hand, psychophysical functions and products are

198 

D. Fisette

characterised by their relation to psychical functions (Twardowski 1999b, 109). We shall see that all actions, such as crying or writing, speaking or painting, express something, for example an emotion, which are not to be confused with their meaning proper. The second classification, though simpler, is just as important as the first one for Twardowski’s project in this study. It only bears on products and concerns, at first sight, the (temporal) relationship that these maintain with psychical or physical functions. It is based on the distinction between enduring and non-enduring products. A psychophysical product like a cry belongs to the class of non-enduring formations because the duration of its existence coincides with that of the act that produces it. On the other hand, enduring products such as writing and painting, are products that last longer than their functions because a function like writing, unlike the verb crying, has in addition to an internal object, like the writing, an external object that serves as a support or material that exists independently of this function (Twardowski 2017b, 177). Moreover, Twardowski distinguishes internal objects, to which mental functions relate, from external objects such as the painter’s canvas and his instruments, which are prior to the formations and serve as material or support for the products, such as a painting, and which thus enable it to persist through time. This is the case of both a footprint left in the sand and a drawing on a sheet of paper, the latter being a psychophysical product which also presupposes a psychical function and a transcendent object (the presented landscape). Moreover, enduring products as such are distinguished from materials developed through the use of functions in that they constitute the configurations and groupings that result from the processing or the work of the function. In the third section, he evaluates the possible ‘crossover’ of the elements of these two classifications, in particular that of the class of enduring products with the subclass of psychophysical functions and products. In addition to the taxonomic issues which occupy much of this last section, one of Twardowski’s major concerns is the status of these products, which have a longer duration than the functions that produce them or the temporary products that are fixed by technical means. In § 39, Twardowski first distinguishes what a product expresses from what it means. A cry can express pain just as a work of art necessarily expresses

9  Overcoming Psychologism: Twardowski on Actions… 

199

the artist’s sentiments. But the meaning of such psychophysical products is not exhausted in the expression of emotions or feelings, for in addition to expressing, products mean something, and their meaning is comparable to what Stumpf calls ‘invariants of Gebilde’ (Twardowski 2017b, 184; 1999b, 127). Twardowski refers, in this context, to the beginning of Husserl’s Logical Investigations where he uses the notion of ideal meaning, and argues that the meaning thus understood is not something transcendent with respect to psychical functions, something that belongs to a third world, but an abstractum resulting from a concept formation process that operates on the very content of functions (Twardowski 2017b, 185, § 39). As a product, meaning forms a concrete whole with the function that actualises it and thus maintains a unilateral dependency relation with it. It cannot exist anywhere else than in the intellect that produced it. But we saw that Twardowski does not dispute the objective nature of the Gebilde/product in the field of logic and more generally in the field of the moral sciences. But he maintains that the products’ emancipation from their original functions and from psychology in general, and their independence with respect to functions, shall not to be understood as one form or another of Platonism. That is why he proposes an explanation of how a psychical product acquires its relative independence with respect to functions. The best example lies in substitutive products, i.e. products which do not emerge straightforwardly from their correlative functions, but by other artificial means. For example, a logical proposition deprived from its assertive force or, as he says, as a product that would simply be presented in the absence of the judgement’s function (or a ‘represented judgement’ in quotation marks), is considered a non-­ natural product or an ‘artifact’. This is how he conceives of Bolzano’s Sätze an sich as judgements considered independently from the act of judging (Twardowski 2017b, 187; 2014b, 135–136). Propositions are ‘hypostasised mental products’, conferring on them some sort of peculiar being, as if they were something that exists beyond the mind in which they originate’ (Twardowski 1999c, 136).

200 

6

D. Fisette

Psychologism and the Delineation of Psychology and the Humanities

Let us now return to psychologism which is the subject matter of the last section of Twardowski’s paper and which is further worked out in an article entitled “The Humanities and Psychology”. These two writings seem to corroborate Twardowski’s remark in his autobiography according to which the form of psychologism that he advocated before the publication of Husserl’s Logical Investigations and which he now tries to overcome pertains to the delineation of the field of psychology and that of the humanities, including logic, rather than to the problem of the ontological status of laws and propositions (Twardowski 1999d, 31). This is also his main argument in his lectures on the psychology of thinking in which the delimitation of these two domains corresponded rigorously to the distinction between mental functions, understood as the proper object of psychology, and logic whose objects of study are propositions understood as products or specific contents of judgement. Let’s examine more closely Twardowski’s argumentation. In the last section of “Actions and Products” (§ 45), Twardowski again addresses the issue of psychologism in emphasising this time the more general topic of the relationship between psychology and the humanities. He challenges Stumpf ’s position in his treatise ‘On the classification of sciences’ according to which the subject matter of the humanities are complex psychical functions. Against Stumpf, Twardowski proposes to repatriate Brentano’s three main normative philosophical sciences, i.e. logic, aesthetics and ethics, into the domain of the Geisteswissenschaften by entrusting each of these philosophical disciplines with the study of their respective products rather than functions. Twardowski claims that the objects of the humanities are not only products of the human mind (i.e. products of psychical functions as such) but also psychophysical products such as ‘various organizations of communal life, and even the very expression of human speech’ which are psychophysical products just like jumping or singing (Twardowski 1999d, 136). It also provides important details on the subject matter of descriptive psychology and its delineation with respect to the other human sciences. For contrary to

9  Overcoming Psychologism: Twardowski on Actions… 

201

what the 1914 paper might suggest, psychology deals not only with psychical functions but also with products of the human mind, i.e. mental products, whereas the subject matter of all the other philosophical sciences are merely psychophysical products. Twardowski argues that the latter abstract from this factual connection of mental products with mental functions which produce them, and treat these products independently of mental life, ‘in which alone they can truly exist’ (Twardowski 1999d, 136). Hence the definition of the Geisteswissenschaften ‘as the sciences whose objects [of study] are either mental products, considered independently of the mental functions that produce them, or psychophysical products, considered as such’ (Twardowski 1999c, 139). However, descriptive psychology remains a fundamental science among the main philosophical sciences and the humanities in general insofar as it is indispensable to explain the genesis of the mental and psychophysical products which depend in turn on psychical acts or functions. But they remain distinct from psychology which studies the primary psychical functions and its founding role for the humanities. This division of labor and this classification make it possible, according to Twardowski, to definitively dismiss the objection of psychologism (Twardowski 2017b, 188).

7

Final Remarks

To conclude, I would like to briefly assess the relevance of this solution to the problem of psychologism based on the function-product distinction and a sharper delineation of psychology with respect to logic and humanities. Several commentators such as Cavallin (1997) in his classical book on Twardowski, seem to be sceptical when they use the distinction between methodological and ontological psychologism to explain the evolution of Twardowski’s thought on that issue (see Woleński 1989; Schaar 2016, 61). Methodological psychologism is understood as the use of the methods of psychology in philosophy, perhaps introspection but most certainly the analysis and description of mental phenomena, whereas ontological psychologism consists in the claim that ‘objects of a certain type (e.g. values, meanings, judgements) are mental (psychical)

202 

D. Fisette

objects, and the sciences which deal with them (axiology and logic respectively) are parts of psychology’ (Cavallin 1997, 41). Cavallin, for instance, argues that Twardowski advocated these two forms of psychologism before 1902, but after the publication of Husserl’s Logical Investigations, he would have abandoned ontological psychologism while preserving methodological psychologism. The question then arises whether the importance granted to descriptive psychology in “Actions and Products” commits Twardowski to something like methodological psychologism. I doubt it, and I am rather of Stumpf ’s opinion that despite the importance of descriptive psychology in Husserl’s Logical Investigations, for example, he was certainly not targeted by his own criticism. I agree with Stumpf ’s claim in “On the classification of sciences” that the form of psychologism targeted by Husserl only concerns ultimately the relationship between pure logic, understood as a Wissenschaftslehre, and genetic psychology, but not descriptive psychology as such (Stumpf 1906a, 200). Stumpf ’s interpretation seems to be corroborated by Husserl’s remarks in the discussion he had with Brentano in Florence in 1907 and in the correspondence thereupon, which focused on logical psychologism (see Husserl 1994b, I, 26). Husserl explained to Brentano the distinction, in the Prolegomena, between two aspects of logic: logic understood as a Kunstlehre or practical discipline, and logic understood as a theoretical discipline, i.e. as a theory of science and pure logic. In response to Brentano’s apprehensions, Husserl clearly reminds him that to base logic understood as a Kunstlehre on psychology is not psychologism. However, as far as this conception of logic is attributable to Brentano and most of his students, including Twardowski, he is not directly targeted by this criticism. For the form of psychologism described by Husserl in his correspondence with Brentano is simply the ‘overestimation of psychology as an alleged fundamental discipline for the whole of philosophy, and therefore also for pure logic and the theory of knowledge’ (1994b, 27). This is what Husserl reminds Brentano in this correspondence and during their discussions in Florence in 1907, discussions which obviously did not convince Brentano (Brentano 1995, 238). Many philosophers and psychologists have reacted just like Brentano, who seems to have understood this objection as an all-out criticism of psychology. Husserl was therefore misunderstood when he said, in the

9  Overcoming Psychologism: Twardowski on Actions… 

203

Prolegomena, that he used the term psychologism in a sense entirely devoid of any pejorative colour. Among Brentano’s students, only Twardowski and Stumpf (see Fisette 2015b) seem to have correctly understood the very meaning of this objection and especially its bearing on descriptive psychology. There is therefore no need to worry about the supposed tensions that exist between Husserl’s arguments against psychologism and the place assigned to descriptive psychology in the philosophy of Brentano and his successors, provided that psychologism is only attributable to the foundation of logic on genetic or physiological psychology. For no student of Brentano, starting with Twardowski, has subscribed to this form of psychologism. However, the question remains whether Twardowski succeeded in overcoming logical psychologism in his paper “Actions and products” given the dependence that he maintains between products and the function of judgement. For this dependence between the propositional content of a judgement and the function of judging seems to be the target of Twardowski’s argument against psychologism in his lectures on the psychology of thinking according to which scientific and logical laws are ‘apodictic and independent from experience, logic being an a priori science’ (Twardowski 2014b, 135; see Schaar, 2016, 160 f.). This is another way of formulating Ingarden’s question whether Twardowski succeeded in finding his way between the Scylla of psychologism and the Charybdis of Platonism. This question still remains open.

Note 1. Twardowski 2017a; 2016; 2014a; Meinong and Twardowski 2016; Schaar 2016; Brożek et al. 2015.

References Brentano, Franz. 1995. Psychology from an Empirical Standpoint. Trans. Antos C. Rancurello, D.B. Terrell, and Linda L. McAlister. London: Routledge.

204 

D. Fisette

Brożek, Anna, et al., eds. 2015. Tradition of the Lvov-Warsaw School Ideas and Continuations. Amsterdam: Rodopi and Brill. Cavallin, Jens. 1997. Content and Object. Husserl, Twardowski, and Psychologism. Berlin: Springer. Fisette, Denis. 2003. Représentations. Husserl critique de Twardowski. In Aux origines de la phénoménologie. Husserl et le contexte des Recherches logiques, ed. Denis Fisette et al., 61–92. Paris: Vrin. ———. 2015a. Reception and Actuality of Carl Stumpf ’s Philosophy. In Philosophy from an Empirical Standpoint. Carl Stumpf as a Philosopher, ed. Denis Fisette et al., 11–53. Amsterdam: Rodopi. ———. 2015b. A Phenomenology without Phenomena? Stumpf ’s Criticism of Husserl’s Ideas I. In Philosophy from an Empirical Standpoint. Carl Stumpf as a philosopher, ed. Denis Fisette et al., 319–356. Amsterdam: Rodopi. Haller, Rudolf. 1982. Einleitung. In Zur Lehre vom Inhalt und Gegenstand der Vorstellungen. Eine Psychologische Untersuchung, I–XVIII.  Munich: Philosophia Verlag. Husserl, Edmund. 1982a. Logical Investigations, Vol. 1, 2nd ed. Trans. John N. Findlay. London: Routledge. ———. 1982b. Logical Investigations, vol. 2, 2nd ed. Trans. John N. Findlay. London: Routledge. ———. 1990–1991. Ursprüngliche Druckfassung der Abhandlung ‘Intentionale Gegenstände’ von Husserl. Brentano Studien 3: 142–176. ———. 1994a. Critical Discussion of K. Twardowski: Zur Lehre vom Inhalt und Gegenstand der Vorstellungen. Eine psychologische Untersuchung. In Early Writings in the Philosophy of Logic and Mathematics, 388–395. Berlin: Springer. ———. 1994b. In Briefwechsel, vol. 1 (Die Brentanoschule), ed. Karl Schuhmann and Elisabeth Schuhmann. Dordrecht: Kluwer Academic Publishers. Ingarden, Roman. 1948. The Scientific Activity of Kazimierz Twardowski. Studia philosophica 3: 17–30. Meinong, Alexius, and Kazimierz Twardowski. 2016. In Der Briefwechsel, ed. Venanzio Raspa. Berlin: De Gruyter. van der Schaar, Maria. 2016. Kazimierz Twardowski: A Grammar for Philosophy. Amsterdam: Rodopi and Brill. Stumpf, Carl. 1891. Psychologie und Erkenntnistheorie. In Abhandlungen der Königlich Bayerischen Akademie der Wissenschaften 19, zweite Abt, 465–516. Munich: Franz. ———. 1906a. Zur Einteilung der Wissenschaften. In Abhandlungen der Königlich-Preußischen Akademie der Wissenschaften, Philosophish-historische Classe, 1–94. Berlin: V. der Königliche Akademie der Wissenschaften.

9  Overcoming Psychologism: Twardowski on Actions… 

205

———. 1906b. Erscheinungen und psychische Funktionen. In Abhandlungen der Königlich-Preußischen Akademie der Wissenschaften, Philosophish-historische Classe, 3–40. Berlin: V. der Königliche Akademie der Wissenschaften. Twardowski, Kazimierz. 1894. Zur Lehre vom Inhalt und Gegenstand der Vorstellungen: Eine psychologische Untersuchung. In Gesammelte deutsche Werke, ed. Anna Brożek, Jacek Jadacki, and Friedrich Stadler, 39–122. Berlin: Springer. ———. 1977. On the Content and Object of Presentations. A Psychological Investigation. Trans. Reinhardt Grossmann. The Hague: Martinus Nijhoff. ———. 1999a. In On Actions, Products and Other Topics in Philosophy, ed. Johannes Brandl and Jan Woleński. Amsterdam: Rodopi. ———. 1999b. Actions and Products. Some Remarks from the Borderline of Psychology, Grammar and Logic. In On Actions, Products and other Topics in Philosophy, ed. Johannes Brandl and Jan Woleński, 103–132. Amsterdam: Rodopi. ———. 1999c. The Humanities and Psychology. In On Actions, Products and other Topics in Philosophy, ed. Johannes Brandl and Jan Woleński, 133–140. Amsterdam: Rodopi. ———. 1999d. Self-Portrait. In On Actions, Products and Other Topics in Philosophy, ed. Johannes Brandl and Jan Woleński, 17–32. Amsterdam: Rodopi. ———. 1999e. Psychology vs. Physiology and Philosophy. In On Actions, Products and other Topics in Philosophy, ed. Johannes Brandl and Jan Woleński, 41–64. Amsterdam: Rodopi. ———. 2014a. In On Prejudices, Judgments and Other Topics in Philosophy, ed. Anna Brożek and Jacek Jadacki. Amsterdam: Rodopi and Brill. ———. 2014b. Psychology of Thinking. In On Prejudices, Judgments and Other Topics in Philosophy, ed. Anna Brożek and Jacek Jadacki, 133–160. Amsterdam: Rodopi and Brill. ———. 2016. In Logik. Wiener Logikkolleg 1894/95, ed. Ariana Betti and Venanzio Raspa. Berlin: De Gruyter. ———. 2017a. In Gesammelte deutsche Werke, ed. Anna Brożek, Jacek Jadacki, and Friedrich Stadler. Berlin: Springer. ———. 2017b. Über Gebilde und Funktionen. In Gesammelte deutsche Werke, ed. Anna Brożek, Jacek Jadacki, and Friedrich Stadler, 165–191. Berlin: Springer. Woleński, Jan. 1989. Logic and Philosophy in the Lvov-Warsaw School. Dordrecht: Kluwer.

10 Is the Content-Object Distinction Universally Valid? Meaning and Reference in Twardowski and Meinong Bruno Leclercq

1

Twardowski’s Content-Object Distinction

In Zur Lehre vom Inhalt und Gegenstand der Vorstellungen, Kazimierz Twardowski challenges the claim made by Bernard Bolzano in § 67 of his Wissenschaftslehre (Bolzano 1972 [1837], 88–89), according to which some ideas or presentations (Vorstellungen) such as ‘nothing’, ‘green virtue’, ‘round square’ or even ‘golden mountain’ have a definite content (Inhalt) and yet lack objectivity (Gegenständlichkeit). For Twardowski (1977 [1894], 1–31), all presentations both have a content and an object. When I think about a golden mountain or a round square, my mental act of thinking is directed by some content towards some object. Twardowski’s theory of contents and objects of presentation therefore involves two main claims: first, the claim that the content and the object of a presentation are two different things, and second, (against Bolzano) B. Leclercq (*) Department of Philosophy, University of Liège, Liège, Belgium e-mail: [email protected] © The Author(s) 2021 A. Dewalque et al. (eds.), Philosophy of Language in the Brentano School, History of Analytic Philosophy, https://doi.org/10.1007/978-3-030-52211-7_10

207

208 

B. Leclercq

the claim that all presentations have both a content and an object, including presentations whose object does not exist.1 That the content and the object of a presentation should always be strictly distinguished is justified by Twardowski for four reasons. The first two are related to the fact that the content and the object of an idea do not share the same properties: first, when I think of a unicorn, the object of this idea does not exist but the content does, since it directs my thought so as to allow it to present a unicorn (1977 [1894], 22, 27, 61); and secondly, it is not the (mental) content but the object of this idea that has a horn on its forehead (1977 [1894], 21, 28–29, 41). The last two reasons of the content-object distinction relate to the fact that the content and the object of an idea do not coincide: on the one hand, one and the same object can be presented through different contents, e.g. Salzburg can be thought of as Juvavum or as Mozart’s birthplace (1977 [1894], 29–30);2 on the other hand, and conversely, two different objects can be presented through one and the same content, e.g. the content of a triangle can direct the mind towards lots of different actual and possible triangles (1977 [1894], 31).3

2

Mental Content—Ideal Content—Object

As appears in the previous arguments as well as in the whole treatise, Twardowski’s notion of ‘content’ is somewhat ambiguous. On the one hand, ‘content’ is a psychological notion; the content is some part of the mental act that directs it towards the object and, being mental, it exists (as a real thing) as much as the act even if the object itself does not exist. On the other hand, ‘content’ seems to be a logical notion, more akin to Gottlob Frege’s Sinn, since contents seem to be ideal ways of considering the object that can be shared by different minds (e.g. ‘Mozart’s birthplace’ for Salzburg). If the content is something mental, which is subjectively and qualitatively felt—some of Twardowski’s developments present it in terms of a mental picture –, then, of course, it is of a very different nature than the (real or ideal) objects that are presented by its means. But this would confine Twardowski to a psychologist’s view of contents that could hardly

10  Is the Content-Object Distinction Universally Valid? Meaning… 

209

support a theory of meaning and would at least fall under the criticism of authors such as Gottlob Frege or Edmund Husserl, who insisted on the ideal and intersubjective nature of meaning contents. Husserl, who indeed criticised Twardowski’s text on this point, will later on distinguish two notions of ‘content’: the content which is presented, and the corresponding orientation of the mental act of presentation. The latter, the ‘noesis’, is part of the mental act and exists at the same moment as this act.4 The former, the ‘noema’, is not mental; it is an ideal content that can be shared by several mental acts, including mental acts of several subjects (Husserl 1982 [1913], 213–221, 236–240). It seems that Twardowski used the word ‘content’ for what Husserl will name ‘noesis’. However, Twardowski’s theory of content, and his claim of the universal validity of the content-object distinction, do not necessarily boil down to such a psychologist’s view.5 While the first of Twardowski’s arguments—as part of the mind, the content always exists even if the object, which is not part of the mind, does not—is linked to this psychologist’s view, the other three arguments seem to hold even if the content is not mental but something like Frege’s Sinn6 or Husserl’s Noema. Whether or not they are in accordance with Twardowski’s own historical views, we will here investigate the claims that the object of a presentation should always be distinguished from its ideal content and that all presentations have an object yet perhaps inexistent. Such claims are interesting because they clearly run counter to Bolzano’s view, since Bolzano himself hold some kind of semantic objectivism (with the notion of Vorstellung an sich playing the role of the ideal content) and yet considered that there were objectless presentations, i.e. presentations that have an ideal content but are devoid of any object. Such claims would also (and foremost) challenge what has become the standard theory of meaning according to which having an ideal meaning content does not always imply having an object. From Bolzano to Frege, Russell and Carnap, extensional semantics has been claiming that words such as ‘unicorn’ can have a definite meaning (made of characteristic features) and yet lack objects corresponding to this meaning; their ‘range’ (Umfang) or ‘extension’ is empty. But is the distinction between ideal content and object universally valid? Is there, in every presentation, an object on top of the ideal content?

210 

B. Leclercq

Things seem to be very clear as long as what is presented is a real object. In that case, the object, e.g. Salzburg, is clearly different from the content(s) through which it is presented, e.g. as ‘Mozart’s birthplace’ or as ‘the city that once was Jovavum’. First of all, the object does not belong to the same ontological category as the contents; it is a reality while contents are ideal. Secondly, it is easy to see, in this case, that two different contents can aim at the same real object, and conversely that one and the same content, e.g. ‘Mozart’s child’ can aim at different real objects. But what if the content of the presentation does not aim at a real object? Should we say, like Bolzano, that the presentation is objectless or, like Twardowski, that it aims at an object, distinct from its content, and yet that this object does not exist?

3

Meinong’s Theory of Objects

Twardowski notoriously paved the way to Alexius Meinong’s theory of objects, which will work out on the thesis that all presentations have an object next to their content,7 but that some objects do not exist or even are impossible. Starting from Brentano’s theory of intentional reference and from the notion of objects of mental acts, Meinong (1960 [1904], 77) developed a general theory of ‘objects’, by which he did not only mean objects of actual presentations, but also objects of possible presentations, most of them never actually occurring in any mind.8 In Über Gegenstandstheorie, Meinong (1960 [1904], 83) shows that, despite a traditional ‘prejudice in favour of the actual’, ‘there are objects of which it can be said that they do not exist’. Objects indeed enjoy different forms of being: the existence (Existenz) of real (wirklich) objects is another kind of being than the subsistence (Bestand) of ideal objects such as mathematical entities or such as states of affairs (Objektive) (Meinong 1913 [1899], 382, 394–395; 1910 [1902], 73; 1960 [1904], 79–81).9 Furthermore, some objects, especially impossible objects like ‘round squares’, neither exist nor subsist, while they nonetheless have properties and are therefore the subjects of true judgements—they have a ‘Sosein’. Meinong’s most significant claim is thus that objects need not exist, or even be possible, to be objects of thoughts and judgements, that they can

10  Is the Content-Object Distinction Universally Valid? Meaning… 

211

have definite properties without being—a Sosein without a Sein (Meinong 1913 [1899], 489; 1960 [1904], 84–85). As Meinong’s student, Ernst Mally, states, there is an ‘independence of so-being from being’ (Mally 1904, 127) (Meinong 1960 [1904], 82).10 Such a theory not only contradicts Bolzano’s conception of ‘objectivity’ (Gegenständlichkeit), but also all subsequent stands which are constitutive of standard extensional logic (from Frege and Russell to Quine11), according to which names of objects involve referential presuppositions (Frege 1960b [1892], 68–69)—so that objects make the ontology— while conceptual words can have a meaning (Sinn) made of definite characteristic features12—in Quine’s terms, they make the ideology (Quine 1951, 11–15)—and yet not be satisfied by any object with the consequence that their extension is empty.13 At the crossroads of philosophy of mind and language, Twardowski and Meinong obviously offer a theory of meaning and reference, based on the notions of content and object, that diverges from standard extensional logic in a very interesting way.

4

 Few Principles of Meinongian A (Non-­Standard) Formal Logic

Leaning on Meinong’s views, some non-standard formal systems have been developed to account for the logic of the judgements and reasoning about such objects. For the last forty years, Hector-Neri Castañeda (1974), William Rapaport (1978), Terence Parsons (1978, 1980), Richard Routley (1980), Edward Zalta (1988), Dale Jacquette (1996, 2001), Jacek Paśniczek (1998) and a few others offered several systems based on some of Meinong’s insights.14 While differing from each other on significant points, they all share a few basic ‘Meinongian’ principles. Already accepted in Jaakko Hintikka (1959) or Karel Lambert (1967)’s free logic, a first one is the principle according to which names of objects are not ontologically loaded (Rapaport 1978, 155–156; Routley 1980, 3–4, 46–47; Jacquette 1996, 101, 108, 120–121, 136–137):15 names can refer to inexistent objects, so that (strong)16 existential quantification (Fa

212 

B. Leclercq

→ ∃x Fx) is not valid and only turns valid if completed by an existential statement (which requires a first-order existential predicate: Ea for a exists): Fa ∧ Ea → ∃xFx. Often named ‘characterisation principle’ or ‘free assumption principle’,17 another basic Meinongian principle is the principle according to which objects have all the (nuclear) properties by which they are characterised (in mental acts of presentation or verbal descriptions): the golden mountain is a mountain and it is made of gold; the round square is both round and square (Chisholm 1982b, 59–61; Routley 1980, 3–4, 46–47, 450; Jacquette 1996, 101, 108, 120–121, 136–137). Since, according to Meinong, having properties does not require existing or even being possible, nothing prevents objects to have the properties through which they are thought even if these properties happen not to be satisfied by any actual entity or even if these properties are incompatible in such a way that it is impossible that they are/be satisfied by an actual entity. A third basic Meinongian principle states that, unlike existent objects, some non-existent objects are incomplete, i.e. that they neither have a property nor its complementary for at least some pairs of complementary properties (Meinong (1968–1978) 1915, 181; Routley 1980, 93): the golden mountain, taken as such, neither has its top covered by snow nor has its top clear of snow; since no information about snow on its top is involved in the thought of the golden mountain, the latter is not determinate as far as snow on its top is concerned, it is thus incomplete and violates the law of excluded middle on this point (Meinong (1968–1978) 1915, 171–174; Findlay 1933, 159–162) (Chisholm 1967, 261; Chisholm 1982a, 43). In the same way, the round square neither is blue nor not blue.

5

Twardowski’s Content-Object Distinction Threatened

Even though they account for important Meinong’s insights, the last two principles, however, have to be carefully considered since they seem to lead to the loss of Twardowski’s content-object distinction.

10  Is the Content-Object Distinction Universally Valid? Meaning… 

213

The principle according to which some objects are incomplete, i.e. not determinate as regards properties which do not explicitly take part to their characterisation, could indeed be taken as the exact reverse of the characterisation principle: while the latter states that objects have all properties through which they are characterised, the former would state that objects only have the properties through which they are characterised (and no other one, except perhaps for properties which would logically follow from the ones by which they are characterised). According to such a view, Meinongian logic would be bound to the thesis that an object has all and only the properties through which it is characterised, i.e. an object boils down to its explicit (mental or verbal) characterisation and therefore coincides with this characterisation, which seems to be nothing else than the content that directs mental acts towards this object: the golden mountain is a mountain and it is made of gold, but has no other properties, except perhaps for some properties following from the former such as being a landform or being made of metal; the round square is round and square and is nothing else except perhaps for being elliptic or rectangular, which are analytic consequences of being round and square. In this case, the golden mountain would thus be a different object than the snow-covered golden mountain, who, unlike the former, is determinate as regards snow on its top; and the round square would be a different object than the blue round square, which, unlike the former, is determinate as regards colour. Each variation in the characterisation constitutes a new object (Meinong 1968–1978 [1915], 170–171; Findlay 1933, 158–159) with, as a consequence, some threat on the content-object distinction. The golden mountain boils down to these two properties (plus a few of their analytic consequences) and it can therefore not be aimed at through any other characterisation; conversely, since golden mountains do not exist, only one single (inexistent) object can be aimed at by the characterisation of the golden mountain. Content and object would here seem to coincide.

214 

6

B. Leclercq

 he Content-Object Distinction T as Regards Inexistent Objects

However, even for some inexistent objects, we would, with Twardowski, like to separate objects from the contents through which they are thought (or verbally described) and aimed at. The detective living at 221b Baker Street is the same (inexistent) object as Doctor Watson’s friend; both characterisations aim at—direct the mind towards—one and the same object; and this is because this object, even though it does not exist, does not boil down to his characterisation as the detective living at 221b Baker Street but has many other properties that go over this characterisation, among which the property of being Doctor Watson’s friend. Conversely, one and the same characterisation could aim at different inexistent objects: according to the Grimm brothers’ tale, Cinderella’s stepsister characterises two different (fictional) objects. Even though they are incomplete, fictional characters and other individual inexistent objects also seem to require Twardowski’s content-object distinction. The characterisation of the detective living at 221b Baker Street directs the mind towards an (inexistent) object, Sherlock Holmes, who does not boil down to this characterisation. This, Terence Parsons (1980, 55) explains, is due to the fact that the detective living at 221b Baker Street is only part of the complete characterisation of Sherlock Holmes, which extends to all Conan Doyle says about Holmes in his work. Yet, Parsons and Routley maintain, Sherlock Holmes is incomplete over and above this long characterisation and he is, for instance, not determinate as regards the presence or absence of a mole on his right shoulder—question which Doyle nowhere settles and therefore has no definite answer (Routley 1980, 93, 279, 354). Such an account of inexistent objects somehow seems to respect both Twardowski’s content-object distinction and Meinong’s claim that inexistent objects are incomplete: each reference to Sherlock Holmes is just a part of his complete characterisation in Doyle’s work, and this is why Holmes does not boil down to any of these partial characterisations; at the same time, Sherlock Holmes’ complete characterisation, however long, is finite and this is why the inexistent object Sherlock Holmes is incomplete.

10  Is the Content-Object Distinction Universally Valid? Meaning… 

215

Such a theory, however, still tends to equate the inexistent object Sherlock Holmes to some content, namely to the (long) characterisation provided by Doyle at the length of his work. Just like to the golden mountain, Parsons (1978, 82, 140) says, corresponds to Holmes a finite set of properties, only much larger.18 But, again, this is problematic; once again Twardowski’s content-object distinction is here jeopardised. Indeed, it seems that the case of the individual object Sherlock Holmes is somewhat different from the case of the golden mountain as such. As an individual object, Sherlock Holmes does not seem to boil down to the way he is explicitly characterised in Doyle’s work; Holmes seems to be an object that goes over and above its explicit characterisation, however long it is; we suppose that Holmes has properties Doyle does not explicitly state, and that things happen to him between the events that Doyle explicitly recounts. What makes Holmes an object, inexistent though it may be, lies in the fact that it does not reduce to any content, however long, directed towards it. For its part, the golden mountain as such, i.e. taken (de dicto) as whatever would be a mountain and be made of gold, boils down to the set of theses two properties (plus some of their analytic consequences).19 But this is because, taken as such, the golden mountain is not genuinely an individual object; it is some abstract object endowed with a general nature20 somehow corresponding to Carnap’s individual concept, and thus to some intensional entity21 (Carnap 1947, 13–16). The case would of course be very different if ‘the golden mountain’ would refer to some individual mountain made of gold, or full of gold, as is the case of Erebor in Tolkien’s The Hobbit. In that case, just like Holmes, the golden mountain would have a lot of other properties than just be a mountain and be made of gold;22 and therefore the object, Erebor, would not coincide with the content through which it is aimed at, ‘the golden mountain’, because it would exceed it by far. Twardowski’s content-object distinction thus seems to be valuable even in the case of individual inexistent objects like Holmes or Erebor. But what about abstract objects such as the golden mountain as such, which seems to boil down to a set of properties, which are also constitutive of the (ideal) content? Taken de dicto, i.e. as the abstract object made of all and only the properties that take part to its characterisation, the golden mountain tends to coincide with this content since it does not

216 

B. Leclercq

seem that the same abstract object could be aimed at/meant through some other content, or that, conversely, the same content could aim at another object. So, the third and fourth arguments raised by Twardowski in favour of his content-object distinction here seem to vanish.

7

 he Content-Object Distinction T as Regards Abstract Objects

Still, says Twardowski’s second argument, even in this case, there is a difference between the object and the content inasmuch as they don’t share some properties; it is the object and not the content that is a mountain and is made of gold. We’ll claim, however, that this is not obvious. As some Meinongian logical systems do, in the line of some of Meinong’s or Mally’s insights, we should distinguish between two ways of having properties: encoding (or constituency) and exemplifying (Mally 1912, 64, 76; Findlay 1933, 110–112; Rapaport 1978, 160–161; Zalta 1988, 17–21). The King of Belgians, who exists, exemplifies the property of being a king and he also exemplifies a lot of other properties. The King of France, for his part, does not exist, and therefore he does not exemplify the property of being a king; he only encodes it as part of his characterisation. The King of France is a king by definition; he analytically is a king. The King of Belgians, Philippe I, happens to be a king just like he happens to be the husband of Mathilde d’Udekem d’Acoz and the father of four children; he synthetically is a king.23 The King of France and the King of Belgians are not a king in the same way; the former encodes this property while the latter exemplifies it. Now, if we have this distinction in mind, it is not at all obvious any more than the golden mountain as an abstract object should be distinguished from the golden mountain as an ideal content in as much as only the former but not the latter could be a mountain and be made of gold. In any case, the golden mountain taken de dicto could only encode, but not exemplify, such properties. But this is the case of the golden mountain taken as an ideal content.24 Ideal contents typically encode

10  Is the Content-Object Distinction Universally Valid? Meaning… 

217

properties; they are characterised by defining features and therefore analytically have them. It is an analytical part of the (ideal) content of the golden mountain that it is a mountain and that it is made of gold; this content encodes the properties of being a mountain and of being made of gold; it is a mountain and it is made of gold in this (encoding) sense of being; it has these two properties in this (encoding) sense of having properties. And the golden mountain taken as an abstract object can’t do better… So, it is not at all obvious that Twardowski’s content-object distinction works as far as abstract objects such as the golden mountain as such are concerned; such objects seem to boil down to the contents that characterise them and through which they are presented. Mixing (individual) fictional objects, such as Erebor, with (general) abstract objects, such as the golden mountain as such, Meinong’s category of inexistent objects seems to be misleading. Unless it boils down to the long explicit characterisation of it, the former seems to respect the content-object distinction; but this is less clear in the case of the latter.

8

 re all Inexistent Objects Equal A as Regards the Content-Object Distinction?

Now, the question arises whether individual inexistent objects, which exceed the contents through which they are presented, encode or exemplify their properties. On one side, it could be said that they cannot exemplify their properties because only existent objects exemplify properties. On the other side, they seem to behave like existent objects as regards their properties: Sherlock Holmes seems to happen to live at 221b Baker Street and Erebor seems to happen to be full of gold; they seem to have these properties synthetically rather than analytically. Of course, it could be said that this is just seemingly, that, since they do not really have these properties (for lack of existing), they can only encode them. Yet, there seems to be a difference in the way the golden mountain as such and Erebor are full of gold, a difference which is strongly linked to the reason

218 

B. Leclercq

why the former is an abstract object that boils down to the characterisation of being a golden mountain and the latter is an individual object that goes over such a characterisation: even though Erebor is a golden mountain by (Tolkien’s) stipulation, it is presented as having this property among a lot of other ones and as having them synthetically, as happening to have them; while the golden mountain as such, i.e. as the abstract object only made of these two properties (and some of their analytic consequences), is presented as having them analytically, by definition, by constituency. Even though it does not really exemplify the property of being full of gold, Erebor does not merely encode it either; somehow it exemplifies this property in another possible world; next to (actual) exemplification and to encoding, there seems to be a third, intermediate, kind of having properties, namely unreal exemplification or exemplification in another possible world. And this way of having properties, which is specific to inexistent individual objects, is deeply related to the reason why Twardowski’s content-object distinction seems to be as relevant in their case as it is for existent objects and more relevant in their case than it is for abstract objects such as the golden mountain as such. As far as existent objects are concerned, Meinong says that they can only be presented by means of incomplete objects: Philippe I cannot be presented in a single mental act with all his properties, but can only appear as the King of Belgians or as Mathilde’s husband. In this case, Meinong (1968–1978 [1915], 196) says, the incomplete object is an auxiliary object for the ultimate object that is presented by means of it. This, again, seems to show that an abstract object such as the King of Belgians as such (i.e. taken de dicto) plays the role of a content; it is used to present an object (Philippe I) in some specific way. Yet Meinong claims that the King of Belgians as such and the existent King of Belgians are two different objects presented through the same content. For the latter, the (extranuclear) characteristic ‘determinate’ is added to the content in such a way that what is presented is the complete object, Philippe I (Meinong 1968–1978 [1915], 176, 189). Meinong goes on by saying that whether judgements are analytic or synthetic then depends on the auxiliary object by means of which it is presented: as Mathilde’s husband, Philippe I analytically is married; but, as the King of Belgians, he synthetically is married (Meinong 1968–1978 [1915], 205–207).25

10  Is the Content-Object Distinction Universally Valid? Meaning… 

219

But the same kind of relations seem to relate fictional characters or objects to abstract objects: Erebor is always presented by means of incomplete objects, i.e. as the golden mountain or as the mountain hanging over the town of Dale or as Smaug’s lair, and so on. It is made clear that what is meant by these definite descriptions is not the abstract object corresponding to their de dicto reading but an individual object corresponding to some de altero mundo reading, i.e. some de re reading yet in another possible world. And Erebor can be the subject of synthetic judgements because it can be presented by means of different characterisations: it is, for instance, synthetic that the mountain hanging over the town of Dale is full of gold (Leclercq 2018). Leaning on Meinong’s own suggestion, Meinongian logics often deem that, between the King of Belgians as such and Philippe I, the actual King of Belgians, or similarly between the golden mountain as such and Erebor, the golden mountain in Tolkien’s The Hobbit, there is some objective relation named ‘implexion’ (Meinong 1968–1978 [1915], 209–212; Findlay 1933, 167–170): the King of Belgians as such ‘is implected in’ Philippe I since the latter has all the properties constitutive of the former. In “Homeless objects”, Roderick Chisholm explains this theory by relating it to (Plato’s inspired) theory held by Guillaume de Champeaux according to which ‘the incompletely determined bed exists in every actual bed, just as the abstract object triangle (note we say ‘triangle’ and not ‘triangularity’) exists in determinate triangles, and just as the species man (‘man’ not ‘humanity’) exists in each particular man’ (Chisholm 1982a, 50). Implexion of an abstract object in a concrete (and more complete) object would be something like Plato’s participation (μέθεξις). Even more radically, Héctor-Neri Castañeda (1974, 24) deems that objects are ‘guises’, i.e. small bundles of properties—‘the golden mountain’, ‘the married man’, ‘the father of four children’, ‘the King of Belgians’, and so on –, while existent objects only arise when several of such guises happen to be involved in one and the same substance (or similarly happen to be involved in one and the same fictional individual object [1974, 36–37]): in the actual world, the guises ‘the married man’, ‘the father of four children’ and ‘the King of Belgians’ happen to co-substantiate in Philippe I; similarly, in The Hobbit, ‘the golden mountain’, ‘the mountain hanging over the town of Dale’ and ‘Smaug’s lair’ happen to co-sociate.

220 

B. Leclercq

According to such a view, guises come first and individual objects only arise as complex relations between guises (Castañeda 1974, 23). In such views, abstract objects or guises (which Castañeda takes to be concrete) such as the golden mountain as such seem to be very close to contents in the sense of what directs the mind towards an object and presents it in some way or under some aspect. But, then, it is not obvious that such contents should themselves be taken as objects and distinguished from the contents through which they are presented.

9

Conclusions

If Twardowski’s content-object distinction just lies on a psychologist’s view of contents, it is of course universally valid since the qualitative parts of the mental act that direct it towards what is presented are of a very different nature than what is presented through them. If, however, we are talking about the ideal content, it is not obvious that the content-object distinction works for abstract objects the same way as it works for individual objects, be they real or fictional. Yet it is quite interesting that the content-object distinction seems to be valid for fictional objects as much as for real objects. In this respect, the way the notion of ‘inexistent objects’ has been developed in some Meinongian formal systems by mixing abstract meanings with fictional individuals, however seems to be rather misleading.

Notes 1. On the notions of object and content in Brentano, Höfler (1890), Twardowski and Meinong, see (Marek 2001). 2. For similar claims in Meinong, see for instance (Meinong 1968–1978 [1915], 187). 3. Similarly, for Meinong, see for instance (Meinong 1910, 237, 276). 4. As Findlay (1933, 3) points out, the early Husserl used to name ‘matter’ (Materie) of the mental act what Twardowski and Meinong for their part

10  Is the Content-Object Distinction Universally Valid? Meaning… 

221

named ‘content’ (Inhalt) of the act. See (Husserl 1970 [1901], V, 119–122). 5. Twardowski’s own criticism of psychologism will later on appear in his general distinction between acts and their ‘products’ (see Twardowski 1912; Fisette, this volume). 6. Frege insisted on the fact that he uses the words ‘concepts’ (Begriffe) and ‘thoughts’ (Gedanken) for the ideal contents of presentation and judgement rather than for parts of the mental act by which they are intended. 7. As Findlay (1933, 6–8) pointed, the early Meinong, who co-authored Höfler’s Logik, did not clearly distinguish content and object either; he only started to do it under the influence of Twardowski. For an acknowledgement of this influence, see for instance (Meinong 1913 [1899], 384). Just like Twardowski, Meinong then held a psychologist’s view of the content. Even though it is most often described in the words used for the object it points to, the content, Meinong says, is something mental; it is lived through and qualitatively felt. See for instance (Meinong 1913 [1899], 385) or (Meinong 1906, 59). 8. See (Findlay 1933, 44–45) or (Chisholm 1967, 262). 9. Brentano himself later on clarified his own theories so as to sanitise them from this kind of consequences (Brentano 1924, 287–288). 10. This thesis will later on lead to the assertion that, unlike redness, existence is an ‘extranuclear property’. On this theory, as well as on its relation to Kant’s assertion that existence is not a real predicate or Frege’s assertion that existence is a second-level predicate, see (Jacquette 2001, 397–426). 11. From Russell’s “On denoting” (Russell 1905) to Quine’s “On what there is” (Quine 1948), going through Ryle’s review of John N. Findlay’s Meinong’s theory of objects or his comments in ‘Intentionality theory and the nature of thinking’, Meinong has indeed personified, for Frege’s school, the ontological overindulgence of Brentano’s school. That, however, the early Russell (1903) shared Meinong’s views and even defended a more naïve version of these views which will yet radicalise his later fight against them, has often been highlighted. 12. See in particular the preliminary draft of “Über Begriff und Gegenstand” (Frege 1960a [1892], 111). 13. The nature of concepts is not referential but only functional. On this, see (Leclercq 2011).

222 

B. Leclercq

14. On a few interesting properties of these formal systems, see (Leclercq 2012). 15. Frege (1979 [1883]; 1960b [1892]) defended the opposite view. 16. By using a non-ontological (yet objectual) ‘existential’ quantifier, Routley’s noneist logic gets back normal existential quantification (Routley 1980, 204–206). 17. This, of course, is a consequence of the independence of the so-being from the being (Chisholm 1967, 261; 1968, 374; 1982b, 57; 1982a, 39). 18. Routley (1980, 353, 434, 441, 478, 874–876, 885–887) makes a similar suggestion, but he refuses to see there any possibility of reducing objects to something else (Routley 1980, 485, 514–515). 19. This is what Meinong (1910, 271) names ‘closed’ objects. See (Findlay 1933, 157). 20. What makes the singularity of the golden mountain as such? Is it really different to think about this golden mountain and to think about golden mountains? It is, for instant, quite significant that Roderick Chisholm (1982a, 49) tackles the problem of the incompleteness of inexistent objects through the angle of universals: Plato’s ideal bed is incomplete as regards length; it neither is six feet long nor has any other dimension. Similarly, Locke’s triangle in general is incomplete in as much as it is neither oblique nor rectangle, neither equilateral, equicrural nor scalene (Chisholm 1982a, 50). In his lemma on Meinong for the Encyclopedia of Philosophy, it is not even of the golden mountain, but of golden mountains, that Chisholm (1967, 261) says that it is neither true nor false of them that they are higher that Mount Monadnock. See (Leclercq 2014). 21. On Carnap’s ‘functionalisation’ of intentional objects, see (Leclercq 2015). 22. It is not the same to think (conceptually) about the golden mountain as such and to think about—e.g. imagine—some specific golden mountain. In the latter case, it makes sense to ask whether it is snow-covered or not. Imagination, Husserl says, fulfils the meaning intentions quite the way perception does; it therefore provides thought with objects while meaning intentions went no further than conceptual contents. 23. Meinong himself took it as important to distinguish between analytic and synthetic relations. On this, see (Findlay 1933, 144–145). 24. Chisholm (1960, 13) himself seems to admit that talking about properties of abstract objects is not very different from talking about constitutive attributes of concepts.

10  Is the Content-Object Distinction Universally Valid? Meaning… 

223

25. Conversely, Meinong (1968–1978 [1915], 216–217) says, modal judgements about an object such as the King of Belgians depend on the complete objects that can be presented by means of it: the King of Belgians can be married since at least one complete King of Belgians is married; but the King of Belgians cannot be a dog since no complete King of Belgians is a dog. On all this, see (Findlay 1933, 162–217).

References Bolzano, Bernard. 1972 (1837). Theory of Science. Trans. Rolf George. Oxford: Basil Blackwell. Brentano, Franz. 1924. Psychologie vom empirischen Standpunkt, 2nd ed. Leipzig: Meiner. Trans. as Psychology from an empirical standpoint. London: Routledge, 1995. Carnap, Rudolf. 1947. Meaning and Necessity: A Study in Semantics and Modal Logic. 2nd ed, 1956. Chicago: The University of Chicago Press. Castañeda, Héctor-Neri. 1974. Thinking and the Structure of the World. Discours d’ontologie. Philosophia 4 (1): 3–40. Chisholm, Roderick M., ed. 1960. Realism and the Background of Phenomenology. Glencoe: Free Press. ———. 1967. Alexius Meinong. In Encyclopedia of Philosophy, ed. P. Edwards, vol. 5, 261–263. New York: Macmillan. ———. 1968. Critical Review of R.  Kindinger, Philosophenbriefe: aus der Wissenschaftlichen Korrespondenz von Alexius Meinong. Philosophical Review 77: 374. ———. 1982a. Homeless Objects. In Brentano and Meinong Studies, 37–52. Amsterdam: Rodopi. ———. 1982b. Beyond Being and Non-Being. In Brentano and Meinong Studies, 53–67. Amsterdam: Rodopi. Findlay, John N. 1933. Meinong’s Theory of Objects. Oxford: Oxford University Press. Frege, Gottlob. 1960a (1892). On Concept and Object. Trans. Peter Geach and Max Black. In Translations from the Philosophical Writings of Gottlob Frege, 42–55. Oxford: Basil Blackwell. ———. 1960b (1892). On Sense and Reference. Trans. Peter Geach and Max Black. In Translations from the Philosophical Writings of Gottlob Frege, 56–78. Oxford: Basil Blackwell.

224 

B. Leclercq

———. 1979 (1883). Dialogue with Pünjer on Existence. Trans. Peter Long and Roger White. In Posthumous Writings. Oxford: Basil Blackwell, 53–67 Hintikka, Jaakko. 1959. Existential Presuppositions and Existential Commitments. Journal of Philosophy 56 (3): 125–137. Höfler, Aloïs. 1890. Philosophische Propädeutik. Erste Teil: Logik. Vienna: Tempsky. Husserl, Edmund. 1970 (1901). Logical Investigations, Vol. 2. Trans. John N. Findlay. London: Routledge. ———. 1982 (1913). Ideas Pertaining to a Pure Phenomenology and to a Phenomenological Philosophy—First Book: General Introduction to a Pure Phenomenology. Trans. Fred Kersten . The Hague: Nijhoff. Jacquette, Dale. 1996. Meinongian Logic. The Semantics of Existence and Nonexistence. Berlin: de Gruyter. ———. 2001. Nuclear and Extranuclear Properties. In The School of Alexius Meinong, ed. Liliana Albertazzi, Dale Jacquette, and Roberto Poli, 397–426. London: Routledge. Lambert, Karel. 1967. Free Logic and the Concept of Existence. Notre Dame Journal of Formal Logic 8 (1–2): 133–144. Leclercq, Bruno. 2011. Logical Analysis and its Ontological Consequences: Rise, Fall and Resurgence of Intensional Objects in Contemporary Philosophy. In Ontological Landscapes. Recent Thought on Conceptual Interfaces between Science and Philosophy, ed. Vesselin Petrov, 53–96. Frankfurt: Ontos Verlag. ———. 2012. En matière d’ontologie, l’important, ce ne sont pas les gains, mais la participation. Igitur 4 (2): 1–24. ———. 2014. Faire cohabiter les objets sans domicile fixe (homeless objects). Chisholm et les logiques meinongiennes. Bulletin d'Analyse Phénoménologique 10 (6): 85–111. ———. 2015. The Object and Its Concept are (not) One and the Same. In Objects and Pseudo-objects. Ontological Deserts and Jungles from Brentano to Carnap, ed. Bruno Leclercq, Denis Seron, and Sébastien Richard, 63–81. Berlin: de Gruyter. ———. 2018. Looking for Smaug and the Golden Mountain. In Laat ons niet ernstig blijven, Huldeboek voor Jean Paul Van Bendegem, ed. Bart Van Kerkhove, Karen François, Steffen Ducheyne, and Patrick Allo, 327–335. Cambridge: Academia Press. Mally, Ernst. 1904. Untersuchungen zur Gegenstandstheorie des Messens. In Untersuchungen zur Gegenstandstheorie und Psychologie, ed. Alexius Meinong, 121–262. Leipzig: J.A. Barth.

10  Is the Content-Object Distinction Universally Valid? Meaning… 

225

———. 1912. Gegenstandstheoretische Grundlagen der Logik und Logistik. Leipzig: Barth. Marek, Johann Christian. 2001. Meinong on Psychological Content. In The school of Alexius Meinong, ed. Liliana Albertazzi, Dale Jacquette, and Roberto Poli, 261–286. Aldershot: Ashgate. Meinong, Alexius. 1906. Über die Erfahrungsgrundlagen unseres Wissens. Berlin: Springer. ———. 1910. Über Annahmen. 2nd ed. Leipzig: Barth. ———. 1913 (1899). Über Gegenstände höherer Ordnung und deren Verhältnis zur inneren Wahrnehmung. In Gesammelte Abhandlungen, Vol. 2, edited by Rudolf Haller, Rudolf Kindinger and Roderick M.  Chisholm. Leipzig: Barth. 379–469. ———. 1960 (1904). On the Theory of Objects. Trans. Isaac Levi, D.B. Terell and Roderick Chisholm. In Realism and the background of phenomenology, ed. Roderick M. Chisholm, Glencoe: Free Press, 79–117. ———. 1968–1978 (1915). Über Möglichkeit und Wahrscheinlichkeit. Beiträge zur Gegenstandstheorie und Erkenntnistheorie. In Gesamtausgabe, Vol. 4, ed. Rudolf Haller and Rudolf Kindinger, Graz: Akademische Druck. Parsons, Terence. 1978. Nuclear and Extranuclear Properties. Noûs 12 (2): 137–151. ———. 1980. Nonexistent Objects. New Haven and London: Yale University Press. Paśniczek, Jacek. 1998. The Logic of Intentional Objects. Dordrecht: Kluwer. Quine, Willard Van Orman. 1948. On What There Is. Review of Metaphysics 2 (5): 21–38. ———. 1951. Ontology and Ideology. Philosophical Studies 2: 11–15. Rapaport, William J. 1978. Meinongian Theories and a Russellian Paradox. Noûs 12 (2): 153–180. Routley, Richard. 1980. Exploring Meinong’s Jungle and Beyond. Canberra: Department Monograph #3 of the Philosophy Department of the Australian National University. Russell, Bertrand. 1903. Principles of Mathematics. London: Allen and Unwin. ———. 1905. On denoting. Mind 14 (56): 479–493. Twardowski, Kazimierz. 1977 (1894). On the Content and Object of Presentation. Trans. Reinhardt Grossmann. The Hague: Martinus Nijhoff. Zalta, Edward. 1988. Intentional Logic and the Metaphysics of Intentionality. Cambridge MA: MIT Press.

11 Extensionality/Intensionality in Polish Philosophy of Language: From Twardowski to Ajdukiewicz Jan Woleński

1

Introduction

It may be surprising that many major general1 encyclopaedias and dictionaries of philosophy (for example, Runes 1942; Edwards 1967; Ritter et al. 1971–2007; Craig 1998) have no separate entry on extensionality or include a very small one. Intensionality is somehow better treated, probably due to problems with so-called intensional contexts. However, there is one exception.2 Zygmunt Zawirski, a Polish philosopher and a member of the Lvov-Warsaw School, prepared an extensive philosophical dictionary. He began this work during World War II and finished it about 1947. The book was even announced by a publishing house in Krakow, but never published, probably due to political circumstances (non-­ Marxist works were ‘politically incorrect’ at that time and their publication was not easy). Zawirski popularised ‘normal’ philosophy, ideologically neutral, but this attitude was criticised by the ruling authorities and their

J. Woleński (*) University of Information, Technology and Management, Rzeszów, Poland e-mail: [email protected] © The Author(s) 2021 A. Dewalque et al. (eds.), Philosophy of Language in the Brentano School, History of Analytic Philosophy, https://doi.org/10.1007/978-3-030-52211-7_11

227

228 

J. Woleński

propagandists. Fortunately, the manuscript of Zawirski’s work has been preserved.3 It has an ‘Extensionality’ entry (two pages in length— ekstensjonalność in Polish). Zawirski explains that the term is derived from Latin extensio which had two meanings: (a) spatial extension, and (b) scope as in the expression ‘scope of a concept’. The sense (b) was adopted by traditional logic. Zawirski says that contemporary logic is extensional in the sense that it deals with extensions, not intensions. Consequently, the concept of extensional propositional function is fundamental for logic. Roughly speaking, two propositional functions are extensional if and only if they are equivalent (have the same logical value, that is, truth or falsity) for arguments having the same values. Thus, if we substitute truth for truth or falsehood for falsehood, the operation in question does not change the logical value of an expression in which this substitution is performed. All functors of propositional calculus are extensional functions, as well as identity, as defined by Leibniz. Zawirski adds that not all propositional functions are extensional, because some are intentional (not ‘intensional’—see below). Zawirski’s treatment of extensionality agrees with more specialised accounts occurring in more or less advanced textbooks and monographs in logic, published in the 1930s and 1940s. It indirectly documents that Polish logicians were interested in the nature of extensionality. To anticipate my further considerations, the principle of extensionality was one of the fundamental dogmas of the Polish School of Logic. Perhaps it was the reason why Zawirski decided to include a relatively advanced entry on extensionality in his general philosophical dictionary. My aim is to report on the main contributions to the extensionality/intensionality problem in Polish logical literature, their genesis and significance,

2

General Historical Remarks

Traditional (pre-mathematical) logic was mostly concerned with terms (names) as main logical items. Consequently, the distinction between content and scope (Inhalt and Umfang in German) of nominal expressions was basic for the logic of terms. Today, we speak about intensional

11  Extensionality/Intensionality in Polish Philosophy… 

229

(dealing with intensions) and extensional (dealing with extensions) logic. However, it is difficult to establish who introduced the terms ‘extensional and ‘intensional’ in their semantic meaning. Goclenius (1613, 201) explains that ‘in logicis extensio est ampliatio’ (‘in logic, extension is the enlarging of something’). One of the applications of ampliatio consists in extendere legem, that is, enlarging the scope of legal category, for instance by so-called interpretatio extensiva (extensive interpretation). Although this example may be interesting from the point of view of the development of legal logic, it would be difficult to see in ampliatio an anticipation of the contemporary meaning of the term ‘extension’ (the name ‘intensio’ does not occur in Goclenius’ Lexicon philosopophicum). Wilhelm Krug (Krug 1832–1838 I, 871, entry ‘Extension’) speaks about extensive Größe (extensive magnitudes). The entry ‘Intension’ (II, 530–531) associates intension with intensiveness. Krug’s dictionary also has (IV, 289) the entry ‘Umfang’. The scope can be attributed to concepts referring to a collection of objects as well as to propositions as being general or particular. Sometimes it happens that it is possible to attribute to concepts extensive magnitudes. If concepts C and C’ have such scopes, they can be compared by means of relations between magnitudes (cardinalities in the contemporary logical nomenclature) characterising scopes and allowing to compare them. For example, the scope of lions is subordinated to the scope of animals. Moreover, Krug mentions the traditional (already in his time) principle that increasing the content results in decreasing the scope. Although he did not introduce a clear semantic idea of extension, some of his remarks anticipate later views, even if the meaning of ‘extensive magnitude’ that he uses is weaker than the present account of extension, which is not necessarily associated with calculations. Rudolf Eisler in his dictionary (Eisler 1930 I, 427, 763) links extension with being extensive in the spatial (geometrical) sense, but intension with intensiveness. No semantic meaning is attributed to these concepts and they are not employed in the entry ‘Umfang’ (1930, III, 294–295). In the nineteenth and twentieth centuries, several terminological novelties appeared, like connotation/denotation (Mill), sense/reference (Frege; originally Sinn/Bedeutung—the latter can be translated as ‘meaning’ according to its uses in linguistics) or meaning/denotation (Russell).

230 

J. Woleński

It seems that it was Bertrand Russell who first introduced a clear distinction between extension and intension: It has been customary, in works of logic, to distinguish two standpoints, that of extension and that of intension. Philosophers have regarded the latter as more fundamental, while Mathematics has been held to deal especially with the former. (Russell 1903, 66)

Russell illustrated the distinction in question by pointing out that classes can be characterised extensionally by referring to their scope or intensionally by referring to a predicate which is satisfied by a certain collection of objects. More technically (and restricting ourselves to monadic predicates), if P(x) is a predicate, then all objects which satisfied this predicate form a class determined by it. Since the former way (pointing out the element of a class by listing them) is not realisable in the case of infinite classes, logic cannot be based on a purely extensional mode of thinking. On the other hand, if we fix the scope of a class, extensionally or intensionally, the former account remains as proper as the latter. Consequently, relations between classes can be established according to their extensions—this is a characteristic feature of set theory. The algebra of logic as developed by George Boole, Ernst Schröder or Charles S. Peirce was extensional. The same concerns logic as codified by Frege and Russell himself, although they observed several problems with intensional contexts, descriptions, substitutivity of co-referential terms, intensionality or extensionality of the theory of logical types, etc.4 Propositional calculus—the simplest logical theory—is a paradigm of a perfect extensional theory. Let A be a formula of propositional calculus having A1, A2, …, An as its constituents (subformulas). Thus, the logical value of A is entirely determined by the logical values of A1, A2, …, An . For example, the value of A = A1 ∧ An depends only on the logical values of both conjuncts—A is true if A1, A2 are true, and false in other cases. It seems that the general principle of extensionality was introduced by Rudolf Carnap (1929, 22) and says that all propositional functions are extensional, that is, if A and B are equivalent propositional functions, they remain such after substitutions of their constituents by equivalent formulas—if two formulas are

11  Extensionality/Intensionality in Polish Philosophy… 

231

co-true (both true) or co-false (both false) they are equivalent (it is material not logical equivalence).

3

Twardowski and Extensionality

Alfred Tarski made a remarkable historical remark about the role of Kazimierz Twardowski in the development of the Polish School of Logic: Almost all researchers who pursue the philosophy of exact sciences in Poland are indirectly or directly the disciples of Twardowski, although his own work could hardly be counted within this domain. (Tarski 1992, 20)

As far as the problem of extensionality is concerned, Twardowski was not directly involved in it. On the other hand, some of Twardowski’s views influenced later Polish logicians and philosophers. Twardowski followed Brentano in several of his views about language, particularly on names. According to the latter (see Brentano 1955, § 18), names have three functions: (i) they express some act of presentation; (ii) they mean some content of presentation; (iii) they name some object of presentation. In particular, there are concepts with the same objects belonging to their scope (Umfang), but having different meanings (Bedeutungen). In general, Brentano’s theory of names had obvious links with his theory of intentionality. Due to this background, Twardowski and his students could consider referential relations of nominal expressions as grounded in the intentionality of mental acts. Twardowski adopted Brentano’s account of the three functions of names. On the other hand, he precisely distinguished object and content of presentation (see Twardowski 1894). His novelty consisted in developing a much more linguistic approach to mental phenomena studied by descriptive psychology than the one that can be found in Brentano’s work. In particular, Twardowski analysed names as sui generis entities independently of mental acts. Consequently, intentional aspects of presentations as mental acts were analysed via semantic features of corresponding names. In other words, he proposed to analyse properties of names instead of dealing with features of mental acts. In fact, Twardowski

232 

J. Woleński

introduced a new paradigm into the Brentanian tradition. Of course, he did not neglect mental acts, but he considered the realm of language as independent of the psyche, although related to mental phenomena. Twardowski gives the example of the phrases ‘the city located at the place of the Roman Juvanum’ and ‘the birthplace of Mozart’ which have different meanings, although they refer to the same object. This is reminiscent of Frege’s famous case of ‘the Morning Star’ and ‘the Evening Star’, but without deriving consequences as important as Frege did. For him sense and reference of expressions (more precisely, of names—although Twardowski worked on objects of propositions, his ideas about this topic are rather vague and not continued), are properties of linguistic items, not of mental acts. Consequently, if referential relations hold between names and objects, this circumstance allows considering the extension (Twardowski did not use this word) as somehow autonomous. Using Twardowski’s later distinction between acts and products (see 1912), language, even as generated by mental acts, can be considered as a product having special properties (more on that in Fisette, this volume). This view helped establish semantics as independent of psychology. There is also an interesting terminological peculiarity. Many Polish philosophers (including Zawirski, see Sect. 1) used the term ‘intentional’ instead of ‘intensional’ (this way of speaking was abandoned after 1945). The reason was that context intensionality—which is exhibited, for example, by sentences such as ‘John thinks that p’—is derived from the intentionality of the corresponding mental act (thinking that p), which is intentional by definition. The connection between this terminological tendency and Brentano’s tradition is obvious.

4

Łukasiewicz and Leśniewski on the Extensionality of Logic

Jan Łukasiewicz’s first approach to two-valued logic was Fregean (see Łukasiewicz 1920a). Sentences are (as in Frege) names of logical values. Hence, all true sentences refer to the same object, namely the True, and all false sentences refer to the False. The principle of extensionality is

11  Extensionality/Intensionality in Polish Philosophy… 

233

satisfied because propositional connectives are univocally characterised by equations acting on logical values—for instance: 1 ∧ 1 = 1, 1 ∧ 0 = 0, 0 ∧ 1 = 0, 0 ∧ 0 = 0. Later (see Łukasiewicz 1929), Łukasiewicz rejected his early construction of propositional logic and worked with semantics based on the concept of logical matrix, earlier employed by Paul Bernays, Emil Post, Ludwig Wittgenstein and other logicians. The semantics associated with this approach became standard in contemporary logic, since it also satisfies the principle of extensionality. Clearly, both approaches, the Fregean and the one using matrices (truth-tables), are perfectly equivalent. In fact, the extensionality of (classical) propositional logic can be considered as the model case of extensionality at all. If A is a compound formula, that is, with propositional variables and logical constants as constituents, the logical value of A becomes completely determined by the logical values of its propositional constituents. Although Łukasiewicz’s approach to the propositional calculus (subsequently elaborated on by many of his students) is very precise, comprehensive (for instance on the metalogical side, in particular regarding its results on matrices), using a special free-bracket notation (Polish or Łukasiewicz’s notation) and supplemented by various axiomatisations, it is a codification of well-known or at least expected facts and intuitions. The actual significance and power of the principle of extensionality can be well illustrated by many-valued and modal logic as construed by Łukasiewicz (see Łukasiewicz 1920b, c, 1930). Consider Łukasiewicz’s three valued logic (the system Ł3). It has three logical values, namely 1 (true), ½ (undetermined) and 0 (false). All functors obtainable in Ł3 are extensional. Negation is defined by the equations ¬1 = 0, ¬½ = ½, ¬0 = 1, conjunction by 1 ∧ 1 = 1, 1 ∧ 0 = 0, 0 ∧ 1 = 0, 0 ∧ 0 = 0, 1∧ ½ = ½, ½ ∧ 1 = ½, ½ ∧ ½ = ½, 0 ∧ ½ = 0. Ł3 can be considered an extensional generalisation of two-valued propositional logic by adding the value ½. The same holds true of any n-valued Ł-logic. Ł-systems are not exceptions in the variety of extensional logic, because most many-valued logics (Post-logic, Bochvar-logic, Kleene-logic) share the extensionality property. One of Łukasiewicz’s philosophical motivations in constructing three-­ valued logic was to explain the concept of possibility. In his first works on this system (see Łukasiewicz 1920b, c), he interpreted the value

234 

J. Woleński

‘undetermined’ as a mere possibility and argued that the operator ‘it is possible that’ could not be defined in two-valued logic. If M(A) (it is possible that A) is to be extensional, the value of this sentence must be determined by the value of A. In Ł3, we have: M(1) = 1, M(½) = 1 and M(0) = 0. These equations show that the following laws hold: if A is necessary, A is possible; if A is true, A is possible; if A is true, A is necessary; if A is false, A is impossible; these settings are consequences of extensionality. Łukasiewicz considered these principles as characteristic for modalities and claimed that they should be valid in any reasonable modal logic. Later on, Tarski proposed to define M(A) as ¬A ⇒ A. This definition also works for any n-valued logic and satisfies the principle of extensionality. Thus, modal logic became extensional. One of the controversial problems related to Łukasiewicz’s approach to modal logic concerned the formula: M(A) ∧ M(¬A) ⇒ M(A ∧ ¬A). It is valid in Ł-systems, but intuitively not acceptable, due to a natural view that contradictions are impossible (false by necessity). Łukasiewicz tried (after 1945, see Łukasiewicz 1951) to improve his modal logic by adopting four-valued basic system (I skip the details) with rejection rules—this new modal logic was extensional. Anyway, Ł-modal logic is not an extension of two-valued logic, contrary to Lewis’ logic, which is two-valued but intensional. In fact, Polish logicians were not particularly interested (at least until the 1950s) in Lewis’ logic, as well as in the relations between probability and many-valued logic (see Zawirski 1934 for an attempt in this direction), because of its intensionality. It is a very instructive example of the rejection of intensionalities (even alethic) as suitable for logical analysis—the extensionality dogma really had far-reaching consequences for the Polish School of Logic. Stanisław Leśniewski constructed (see Leśniewski 1929) a powerful propositional calculus called protothetic, that is, a system with quantifiers binding propositional variables and functorial variables. This logic has sufficient resources to formalise the principle of extensionality as a logical theorem—in the cases mentioned earlier, the principle of extensionality is a metalogical rule. The principle in question is captured in protothetic by the formula (A ⇔ B) ⇒ (fA ⇒ fB), (if A and B are equivalent, whatever can be said, in terms of logic, about A, can also be said about B). Łukasiewicz achieved the same result in his system of propositional logic

11  Extensionality/Intensionality in Polish Philosophy… 

235

with variable functors (see Łukasiewicz 1951). Other systems developed by Leśniewski, namely ontology (the calculus of names; see Leśniewski 1930) and mereology (the theory of collective classes with parthood relations; see Leśniewski 1916; 1927–1931) are also extensional. Leśniewski’s ontology can be considered a theory of objects. It is radically nominalistic. Hence, general (common) names are characterised via their scopes, but not understood as sets. In other words, if N is a common name, for instance ‘horse’, it refers to at least one of the horses (it is related to the concept of general names as referring to more than one designatum). Combining this idea with the collective approach to classes, Leśniewski understood extensions of general names as not sets in the standard (set-­ theoretical) understanding. Extensionality in mereology is manifested by the fact that arbitrary objects can be linked by a mereological sum composed of them. Leśniewski’s radical extensionalism led him to formulate a general method of elimination of intensional contexts, but the details are unknown. He probably used in-quoting (putting expressions between quotes, like ‘in-quoting’ as a device and considered intensional contexts as individual objects.

5

 xtensionality and the Semantic Theory E of Truth

Tarski declared (Tarski 1933, 1944; see Woleński 2019 for a detailed analysis) that he had defined the concept of true. This might suggest that his definition explains the meaning of the predicate ‘is true’, that is, its intension. An additional support for this thesis is provided by Tarski’s explicit and numerous declarations that he follows Aristotelian intuitions. However, the definition itself in its formal setting concerns a set of true sentences in a language L, that is, the extension of the predicate ‘is true’. One and the most popular version of the semantic definition of truth runs as follows: a sentence is true if and only if it is satisfied by every sequence of objects. This formula appeals to sequences of objects, or items understood extensionally, that is, as ordered sets. The truth-­ undefinability theorem can be viewed as another manifestation of

236 

J. Woleński

extensionality in Tarski’s theory of truth. This theorem says that the set of true sentences of rich theories (for instance, Peano arithmetic and every theory which contains it) cannot be defined with the resources of such theories—this formulation explicitly refers to the set of truths. If we take the famous T-scheme, that is the formula T(‘A’) ⇔ A*,5 it establishes an extensional equivalence (in the semantic truth-theory) between every concrete instance of T(‘A’) and A*6—for example, the sentence ‘the sentence ’ is equivalent with the sentence ‘snow is white’. It is not a logical equivalence, but an equivalence that holds due to principles of the semantic theory of truth. Typical explanations of the semantic theory of truth say that A* in the T-scheme is a translation of the sentence A into ML—this fact is not evident in our example with snow, unless we consider the sentence ‘snow is white’ as self-translated into English. However, this strategy involves intensionality through the concept of translation which depends on the meanings of the translated expressions. In order to avoid intensionality, Tarski did not define the embedding into ML of the sentences occurring in T-schemes as a translation, but as their inclusion in the metalanguage; thus, the symbol A* means that A ∈ ML. Although Tarski assumed that linguistic expressions have intelligible meanings, the entire semantic job is made by valuation function which acts extensionally. Tarski’s extensionalism in the semantic theory of truth is misinterpreted very frequently. Firstly, what would happen if the sentence ‘snow is white’ meant that grass is green (see Putnam 1985-1986)? The simple answer is: nothing, except that the valuation function would be different, namely, it would attribute the predicate ‘is green’ to white objects or the predicate ‘is white’ to green objects. Secondly, Davidson’s (see Davidson 1967) attempts to explain that the definition of the concept of meaning as truth fails because ‘it means that’ is intensional, whereas ‘it is true that’ (at least on Tarski’s account) is extensional. However, some problems remain (see Kokoszyńska 1936). Since Tarski explicitly says that the problem of truth is meaningless for purely formal (not meaningful) languages, one might claim that we need to speak about meanings. Tarski replied (see Tarski 1936b) that we should rather appeal to meaningful languages, not to meanings as such. Tarski’s extensionalism is also obvious in his definition

11  Extensionality/Intensionality in Polish Philosophy… 

237

of logical consequence (see Tarski 1936a). We say that B is a logical consequence of A if and only if every model of A is a model of B. Tarski, due to his extensionalism, did not like to formulate this definition via the concept of necessity or analyticity because of their intensional character.

6

 ome Polish Attempts S to Eliminate Intensionality

6.1

Kotarbiński

Consider the sentence (see Kotarbiński 1929, transl., 349–353): John thinks that p. Kotarbiński called the sentences falling under (1) ‘psychological’ because of their connections with mental states (or with propositional attitudes, according to the contemporary terminology). The well-known problem is that we cannot substitute other equivalent sentence for A, because such a substitution does not preserve truth. For instance, ‘John thinks that Warsaw is the capital of Poland’ is not equivalent with the sentence ‘John thinks that the largest Polish city is the capital of Poland’. Kotarbiński proposed to replace (1) by: John thinks so: p. To use the above example, we obtain ‘John thinks so: Warsaw is the capital of Poland’. According to Kotarbiński the problem disappears, because the entire sentence is about John, not about Warsaw. That John thinks about Warsaw as the capital of Poland is the concrete item and cannot be replaced by John’s thinking about the largest city in Poland. We are dealing here with two different particulars. This circumstance prevents to make substitution of ‘Warsaw is the largest city of Poland’ for ‘Warsaw is the capital of Poland’ in the related particularisation of (2). It seems that Kotarbiński’s account of psychological sentences was influenced by Leśniewski.

238 

J. Woleński

6.2

Ajdukiewicz

Ajdukiewicz’s attempt to eliminate extensionality (see Ajdukiewicz 1967) is perhaps the most detailed in Polish philosophy during the reported period. Consider the sentences (I repeat original examples): Caesar knew that Rome is located on the Tiber; Caesar knew that the capital of papacy is located on the Tiber; Caesar knew that the capital of papacy is located on the Tiber; Caesar knew about Rome, the Tiber and the relation of being-located-on that Rome is located on the Tiber; Caesar knew about the capital of papacy, the Tiber and the relation of being-located-on that the capital of papacy is located on the Tiber. How to avoid the paradox arising from substituting ‘the capital of papacy’ for ‘Rome’? According to Ajdukiewicz, the form of (4) is: Caesar knew about f(papacy), the Tiber and being located on the Tiber, that f(papacy) is located on the Tiber. If ‘the capital of ’ is taken as f, the substitution is impossible. Moreover, this formula is extensional because substitution is possible if Caesar knew that f ’(papacy) = f ”(papacy)—both are false, but (7) is extensionally equivalent to Caesar knew that the capital of the Roman Republic is located on the Tiber. Ajdukiewicz’s paper was published posthumously and it is difficult to say whether his method is general, for instance, whether it applies to cases with two different proper names and examples, like ‘Mark Twain was the author of The Adventures of Tom Sawyer’ and ‘Samuel Clement was the author of The Adventures of Tom Sawyer’.

11  Extensionality/Intensionality in Polish Philosophy… 

7

239

Final Remarks

As I already noticed, Polish logicians shared the thesis of extensionality in a very strong sense. Carnap (see Carnap 1947) distinguished two kinds of intensional contexts. Let Int(A) be a general form of intensional contexts. Consider M(A) and M(B). If A and B are logically equivalent, the same holds true of M(A) and M(B). Now consider (i) ‘a knows that A’ and (b) ‘a knows that B’. It can happen that A and B are logically equivalent, but (i) and (ii) are not. Carnap proposed intensional isomorphism as a relation helpful for analysing intensional contexts of type (ii). Polish logicians, like Łukasiewicz, agreed that M(A) and M(B) are logically equivalent, if A and B are such, but claimed that modalities should be extensionally equivalent and many-valued logic was a price to pay in order to go in that direction. Ajdukiewicz’s analysis is perhaps a step toward the extensionalisation of intensional isomorphism, but this issue requires further investigations. How was ‘Polish’ thinking on intensionality related to Brentano? One can say that since intensionality is a by-product of intentionality of mental acts, the latter should be analysed as a direct relation to objects via linguistic expressions. A terminological mark of this idea is documented by a frequent usage (see above) of the expression ‘intentional function’ instead of ‘intensional function’. Moreover, most Polish analytic philosophers had doubts, similarly to those raised by Brentano, concerning entia rationis. Hence, the treatment of meaning as an abstract object was unpopular in Poland and reservations toward intensionality were also motivated by this issue; the meaning of expressions was considered as an internal property of expressions. Leśniewski and Kotarbiński explicitly, but Tarski implicitly, accepted nominalism and this standpoint motivated their extensionalism. In the case of Leśniewski (Kotarbiński followed him), his ontology (conceived of as a logical system) was objects-oriented—this fact can be considered as rooted in Brentano’s tradition, although extensionalism is not a consequence of an objectual attitude (in the early Brentano it was not, contrary to the later views of this philosopher). As far as the matter concerns pure logic, the extensional point of view was considered as the best background for formal logical work.

240 

J. Woleński

Notes 1. I stress the word ‘general’ as contrasted with ‘specialised’ as applied to monographs or textbooks. 2. Of course, I do not claim that my knowledge of philosophical encyclopaedias and dictionaries is extraordinary. However, I am inclined to think that examples which I mention are typical in the discussed context at least until 1980. The later general literature more frequently mentions extensionality. 3. I discovered it by chance in the 1980s. 4. I omit a discussion of these questions in the works of these authors as well as subsequent developments. 5. T refers to the truth-predicate, A* is an embedding of A into the metalanguage ML—the entire definition is expressed in ML. 6. A T-schema is not a definition of truth, but a condition of adequacy—this condition is entailed by the definition of truth.

References Ajdukiewicz, Kazimierz. 1967. Intensional Expressions. Studia Logica 20: 63–86. Brentano, Franz. 1955. Die Lehre vom richtigen Urteil. Hamburg: Meiner. Carnap, Rudolf. 1929. Abriss der Logistik. Mit besonderer Berücksichtigung der Relationstheorie und Ihrer Anwendungen. Vienna: Julius Springer. ———. 1947. Meaning and Necessity: A Study in Semantics and Modal Logic. Chicago: Chicago University Press. Craig, Edward, ed. 1998. Routledge Encyclopedia of Philosophy, 12 vols. London: Routledge. Davidson, Donald. 1967. Truth and Meaning. Synthese 17: 304–323. Goclenius, Rudolf. 1613. Lexicon Philosophicum Quo Tanquam Clave Philosophiae Fores Aperiuntur. Francofurti. Edwards, Paul, ed. 1967. The Encyclopedia of Philosophy, 8 vols. New  York: Macmillan. Eisler, Rudolf. 1930. Wörterbuch der philosophischen Begriffe. Vol. 3 vols. Berlin: Mittler and Sohn. Kokoszyńska, Maria. 1936. “W sprawie względności i bezwzględności prawdy” (Concerning Relativity and Absoluteness of Truth). Przegląd Filozoficzny 39: 424–425.

11  Extensionality/Intensionality in Polish Philosophy… 

241

Kotarbiński, Tadeusz. 1929. Elementy teorii poznania, logiki formalnej i metodologii nauk (Elements of Theory of Knowledge, Formal Logic and Methodology of Science). Lvov: Osslineum. Trans. Olgierd Wojtasiewicz as Gnosiology. The Scientific Approach tot he Theory of Knowledge, Oxford: Pergamon Press, 1966. Krug, Wilhelm Traugott. 1832–1838. Allgemeines Handwörterbuch der philosophischen Wissenschaften nebst ihrer Literatur und Geschichte, 5. Leipzig: Brodhaus. Leśniewski, Stanisław. 1927–1931. O podstawach matematyki. Przegląd Filozoficzny 30: 164–206; 31: 261–291; 32: 60–101; 33: 77–105, 142–170. Trans. D.I. Barnett as “On the Foundations of Mathematics”. In Collected Works, 174–382. Dordrecht: Kluwer, 1992. ———. 1929. Grundzüge eines neuen System der Grundlagen der Mathematik. Fundamenta Mathematicae 14: 1–81. Trans. M.P. O’Neil as “Fundamentals of a New System of the Foundations of Mathematics”. In Collected Works, 410–605. Dordrecht: Kluwer, 1992. ———. 1930. Über die Grundlagen der Ontologie. Comptes Rendus des Séances de la Société des Sciences et de Lettres de Varsovie, Classe III 23: 111–132. Trans. M.P.  O’Neil as “On the Foundations of Ontology”. In Collected Works, 606–628. Dordrecht: Kluwer, 1992. Łukasiewicz, Jan. 1920a. Logika dwuwartościowa. Przegląd Filozoficzny 23: 189–205. Trans. L.  Borowski as “Two-valued logic”. In Selected Works, 89–109. Amsterdam: North-Holland, 1970. ———. 1920b. O pojeciu możliwości, Ruch Filozoficzny 2: 229–230. Trans. L.  Borowski as “On the concept of possibility”. In Selected Works, 15–16. Amsterdam: North-Holland, 1970. ———. 1920c. O logice trójwartościowej”, Ruch Filozoficzny, 2, 230–231. Trans. L. Borowski as “On Three-Valued Logic”. In Selected Works, 87–88. Amsterdam: North-Holland, 1970. ———. 1929. Elementy logiki matematycznej. Warsaw: Wydawnictwa Koła Matematyczno-Fizycznego Uniwersytetu Warszawskiego. Trans. Olgierd Wojtasiewicz as Element of Mathematical Logic. Oxford: Pergamon Press 1966. ———. 1930. “Philosophische Bemerkungen zu mehrwertigen Systemen der Aussagenkalküls”. Comptes Rendus des Séances de la Société des Sciences et de Lettres de Varsovie, Classe III 23: 51–77. Trans. L. Borowski as “Philosophical Remarks on Many-Valued Systems of Propositional Logic”. In Selected Works, 153–178. Amsterdam: North-Holland, 1970. ———. 1951. On Variable Functors of Propositional Arguments. Proceeding of the Royal Irish Academy (Section A) 54: 23–35.

242 

J. Woleński

Ritter, Joachim, Karlfried H. Gründer, and Gottdried Gabriel, eds. 1971–2007. Historischer Wörterbuch der Philosohischen Begriffen, 13 vols. Darmstadt: Wissenschafliche Buchgessellschaft. Putnam, Hilary. 1985–1986. “On Comparison of Something with Something Else”. New Literary History 17: 61–79. Runes, Dagobert David. 1942. The Dictionary of Philosophy. New  York: Philosophical Library. Russell, Bertrand. 1903. Principles of Mathematics. Cambridge: Cambridge University Press. Tarski, Alfred. 1933. Pojęcie prawdy w językach nauk dedukcyjnych (The Concept of Truth in the Languages of Deductive Sciences). Warsaw: Towarzystwo Naukowe Warszawskie, Warszawa. ———. 1936a. Über den Begriff der logischen Folgerung. In Actes du Congrès International de Philosophie Scientifique, vol. 7, 1–11. Paris: Herman and Cie. ———. 1936b. Głos w dyskusji nad referatem M. Kokoszyńskiej (Contribution to the Discussion on M. Kokoszyńska’s talk). Przegląd Filozoficzny 39: 425. ———. 1944. The Semantic Conception of Truth and the Foundations of Semantics. Philosophy and Phenomenological Research 4: 341–395. ———. 1992. Drei Briefe an Otto Neurath [25. IV. 1930, 10. VI. 1936, 7. IX. 1936], together with Eng. tr. by J.  Tarski. Grazer Philosophische Studien 42: 1–29. Twardowski, Kazimierz. 1894. Zur Lehre vom Inhalt und Gegenstand der Vorstellungen. Eine psychologische Untersuchung. Vienna: Hölder. Trans. Reinhardt Grossmann as On the Content and Objects of Presentations. A Psychological Study. The Hague: Nijhoff, 1977. ———. 1912. O czynnościach i wytworach. Kilka uwag z pogranicza psychologii, gramatyki i logiki. In Księga Pamiątkowo ku uczczeniu 250-tej rocznicy założenia Uniwersytetu Lwowskiego przez króla Jana Kazimierza, vol. 2. Lvov: Nakładem Uniwersytetu Lwowskiego, 1–33. Trans. Arthur Szylewicz as “Actions and Products. Some Remarks from the Borderline of Psychology, Grammar and Logic”. In On Actions, Products and Other Topics in Philosophy, 103–132. Amsterdam: Rodopi, 1999. Woleński, Jan. 2019. Semantics and Truth. Heidelberg: Springer Nature. Zawirski, Zygmunt. 1934. Stosunek logiki wielowartościowej do rachunku prawdopodobieństwa. Partially translated by Feliks Lachman as “The Relations between Many-Valued Logic and Probability Calculus”. In Zygmunt Zawirski: His Life and Work with Selected Writings on Time, Logic & the Methodology of Science, 112–114. Dordrecht: Kluwer, 1994.

Part III Brentano’s Wider Legacy

12 Modifying Terms and Modification in Husserl and the Brentano School Maria van der Schaar

1

Introduction

In Aristotle’s De Interpretatione (21 a 18 ff., chapter 11), it is noticed that if we say of a particular man that he is a man, this is true, but, if we say of a dead man that he is a man, this is false. The logical inference from This is a dead man to This is a man is invalid. This is caused by the fact, Aristotle explains, that in what is added, namely being dead, some characteristic is contained, namely being no longer alive, that is in opposition to a characteristic contained in being a man. Being alive and being no longer alive cannot correctly be said of the same thing at the same time. One might thus say that calling something a man presupposes that he is a living being, whereas to say that he is dead contradicts this presupposition. Contemporary philosophers have difficulties to account for this phenomenon, because standard forms of conceptual analysis understand a complex concept as a complex of simple, atomic concepts, unrelated to

M. van der Schaar (*) Institute of Philosophy, Leiden University, Leiden, The Netherlands e-mail: [email protected] © The Author(s) 2021 A. Dewalque et al. (eds.), Philosophy of Language in the Brentano School, History of Analytic Philosophy, https://doi.org/10.1007/978-3-030-52211-7_12

245

246 

M. van der Schaar

each other. The conceptual analysis of bachelor gives two unrelated simpler concepts: unmarried and man. The phenomenon of modification is known to us, though, through Timothy Williamson’s proposal to understand knowledge not as belief plus some other concepts. It is rather that belief is to be understood as knowledge minus something: ‘Mere believing is a kind of botched knowing’ (Williamson 2000, 47). In the Brentanian tradition, terms like ‘botched’ in ‘botched knowing’, or ‘dead’ in ‘dead man’, are called modifying terms, just as ‘false’ in ‘false friend’ and ‘fake’ in ‘fake gold’. In the paper, different explanations are given of modifying terms and the idea is generalised to the idea of modification as an operation upon concepts. Both the concepts of dead man and of illusion can be understood as the result of an operation of modification on, respectively, the concept man and the concept of perception. A phenomenological account should not only explain that we are able to construct a new concept, that of dead man, when the term ‘dead’ is applied to ‘man’, but also how one can construct the concept of illusion from that of perception when no modifying term is present. The idea of modification plays an important role in Husserl’s writings. For example, in Ding und Raum, Husserl understands imagination as a modification of perceptual experience (Husserl 1973, § 18, 55). What can this mean if no modifying term is present? It means, to begin with, that we cannot understand what imagination is, unless we have understood what perceptual experience is. In the Cartesianische Meditationen, Husserl writes, the other presents itself phenomenologically as modification of my self (Husserl 1987, § 52, 118). Husserl adds here that the other ego is constituted by means of the ego. In the same way, ‘meine Vergangenheit’, my past, is given to me ‘als vergangene Gegenwart’ (Husserl 1987, 118). My past is given to me as past present, as intentional modification of the present. Again, this means, at least, that I cannot understand the idea of other unless I understand it in terms of the self, and that I cannot understand what the past is unless I understand it in terms of the present. At the same time, it says something about the way we experience the other, the past. The term intentional modification seems to refer to this special way in which the other and the past are given to us, and seems to show us something about the essence of the other, the essence of the past.

12  Modifying Terms and Modification in Husserl… 

247

In Formale und Transzendentale Logik (1929, Beilage II, § 2), Husserl explains that the same object has an a priori possibility to be given to us in different ways (in memory, perception, imagination). Among these, the experiencing way (die erfahrende), being the original one, has a certain priority (einen Vorzug), to which all the others are related as intentional modifications. These others point back to what is non-modified: ‘Intentionale Modifikation aber hat ganz allgemein das Eigene, dass sie in sich selbst auf das nicht-Modifizierte zurückweist’. (Husserl 1987, 276). In Erfahrung und Urteil, Husserl writes: ‘dass blosses Urteilen eine intentionale Modifikation von erkennendem Urteilen ist’ (Husserl 1929, § 5, 15). In other words, mere judging is given to us as intentional modification of judging as knowing. Judging as knowing is prior in the conceptual order to mere judging. And, in (Husserl 1985, § 7, 23): ‘Vermutlichkeit, Wahrscheinlichkeit usw. sind Modifikationen eines ursprünglichen schlichten Glaubensbewusstsein’, i.e. conjecture and probability are to be understood as modifications from an original consciousness of belief or faith. This is not merely to be understood in a genetic sense—it is not meant as an empirical, psychological thesis. It is rather that these modifications of conjecture, doubt and probability, as ways in which we are conscious of something, conceptually presuppose a belief in a world (einen universalen Boden des Weltglaubens, Husserl 1985, 25). Last but not least, the idea of modification plays a crucial role in Husserl’s notion of epoché: the whole natural world is put into parentheses (in Klammern) through a modification of neutralisation (Husserl 1913, § 109). In the epoché, we describe what we perceive as ‘material object’, as ‘tree’: the quotation marks express the meaning-modification of these words (Husserl 1913, § 89). We thus see that the idea of modification plays a crucial role in Husserl’s writings. What does modification mean for Husserl? He uses the idea of modification at crucial places in the Logical Investigations, and he gives there also an explanation of the idea. I will start the paper by tracing the origin of the idea to Brentano and other members of the Brentano School (Sect. 2). Then, Husserl’s account of modification and its different uses in the Logical Investigations will be given (Sect. 3). In the concluding Sect. 4, I present a first proposal for a phenomenological account of modification.

248 

2

M. van der Schaar

Modification in Brentano and the Brentano School

In the Wissenschaftslehre, Bolzano quotes the Scholastic philosopher Savonarola: ‘a dead man has the form and likeness of a man, but is not a man’ (1985–1987, § 19). Whereas in ‘sea fish’ the noun ‘fish’ expresses the head-presentation, in ‘painted fish’, the head-presentation is not expressed by ‘fish’ but by ‘painted’: a painted fish is not a kind of fish, but a kind of painting (1985–1987, § 59). A possible thought, cogitatio possibilis, is not a kind of thought, but rather a kind of possibility (1985–1987, § 23). The class of dead man is not a subclass of the class of men, but a subclass of dead beings, because ‘dead’ expresses the head-presentation. Bolzano thus gives an explanation of modifying terms, in which a deep structure of the language shows the logical relation between the underlying concepts. This explanation may work in some cases, but it does not work in all cases. A fake gun is not a special kind of gun, but neither is it a special kind of fake(-thing); a false friend is not a special kind of friend, but neither is it a special kind of falsity. Brentano presents the phenomenon of modifying terms as a semantic issue, and as a logical issue, as well. In the logic manuscript from the 1880s, Brentano writes: An adjective may add a characteristic, but often it may modify the meaning of the other part, for ex. past king, future battle, dead man […], possible dollar, etc. (Ms. EL 80, 13.211, 188) A past king as such is not more a king than a beggar is […] The addition of the term ‘past’ is not enriching, but modifying, it cancels the concept of king in its original sense and replaces it by something else. (Brentano 1968, 46)

In the first passage, Brentano presents a semantic account of modifying terms. A term like ‘dead’ or ‘past’, if it is used in a certain context, has the function to change the meaning of the noun to which it belongs. Do these terms really modify the meaning of the head-noun? Why think that a change on the level of meaning occurs? The second passage is formulated in a different way: in contexts with modifying terms a new concept

12  Modifying Terms and Modification in Husserl… 

249

arises. If the term ‘past’ is added to a noun, it cancels the concept expressed by the noun and replaces it by a new concept, for example past king. This seems to give a more natural explanation. In a letter to John Stuart Mill from 1873, Brentano also gives a logical account of modifying terms, which seems to be inspired by the account given by Bolzano. Mill gives a counter-example to Brentano’s thesis that all judgements are either affirmations or denials of existence: ( 1) A centaur is a fiction of the poets. (2) There exists a poetic fiction, in which parts of a human body and that of a horse are unified. (1) does not affirm any existence, according to Mill. Brentano answers that (1) hides its true logical form, made visible in (2). The judgement concerns not a centaur, but a poetic fiction: it affirms the existence of a certain kind of fiction (Brentano 1971, II, chapter 7, 61). Like Aristotle, Brentano acknowledges that a learned man is a man, but a dead man is not a man. And, in the logic manuscript, Brentano writes: (1’) A man is dead,

is not to be understood as having categorical form; its logical form is: (2’) A dead man exists.

The latter form shows that ‘dead’ functions in (1’) as modifying term belonging to the term ‘man’ (Ms. EL 80, 13.211, 13.312). In his famous book from 1894, Twardowski uses the distinction between attributive, also called determining or enriching terms, and modifying terms to explain the distinction between content and object of mental acts. He points to an ambiguity in language regarding the term ‘presented’. The same term may occur in some contexts as attributive term; in other contexts as modifying term. A ‘presented landscape’ may thus refer to a landscape that is presented in a painting, or it may refer to a painting, in which a landscape is presented. In the last case, the term is used as a modifying term, for a presented landscape in this sense is not a

250 

M. van der Schaar

landscape. The same sort of ambiguity we have in the complex ‘presented object’. The term may refer to the object, which is presented: a man or a circle. The term may also refer to the content of the act of presenting, which directs the act to this or that object. The content itself is not the object of the act; the term ‘presented’ in ‘presented object’ is thus used as a modifying term. Twardowski mentions the unique logical behaviour of modifying terms, but his account is a purely semantic one. A ‘determination is called attributive or determining if it completes, enlarges […] the meaning of the expression to which it is attached. A determination is modifying if it completely changes the original meaning of the name to which it is attached’ (Twardowski 1977, 11). This cannot be any standard form of ambiguity, for we do not find different meanings in the lexicon for all terms that may be preceded by modifying terms (see Kripke 2011, 346). In his later writings, when Twardowski devotes a small paper to the different kinds of non-attributive terms, he formulates his explanation of modifying terms in a somewhat different way: The function of modifying is composed, first of all, of the function of partially deleting [something] from the content of the [act of ] presentation expressed by the noun, and secondly, of the function of replacing this deleted constituent of the content […] with other positive or negative characteristics. (Twardowski 1927)

Instead of the term ‘meaning’, it is the term ‘content’ that is now used. For Twardowski, the content is not an ideal meaning, but an existing part of the act. A term in a certain context has thus a particular content, corresponding to the act manifested by the linguistic expression. I have argued elsewhere (Schaar 2016) that this account does not presuppose that there occurs a change in meaning on the level of types. It is rather that the occurrence of a complex is given a new content in this particular context. But, as Jan Claas and Benjamin Schnieder have pointed out, we need not only an explanation of the token ‘toy-duck’ in a particular context (Claas and Schnieder 2019, § 6.a.). The problem recurs on the level of types or identical meanings: ‘toy-duck’ expresses an identical meaning in different contexts, and we need an explanation of that phenomenon,

12  Modifying Terms and Modification in Husserl… 

251

as well. They claim that neither of Twardowski’s accounts is satisfying, and I think their argument is convincing. For Anton Marty, I refer the reader to their paper (Claas and Schnieder 2019). Like Twardowski, Marty understands modifying terms as changers of meaning. Furthermore, he introduces a new distinction within the class of non-attributive terms, also taken up by Twardowski in the paper from 1927. We need to make a distinction between modifying and restorative terms. Of the modifying adjectives: ‘past’, ‘merely possible’, ‘merely presented’, ‘merely desired’ (Marty 1908, 514), a purely semantic explanation is given; they change the meaning of the head noun completely. Restorative terms like ‘real’ (wirklich) and ‘truly’ (wahrhaft) can only be understood if a modified meaning is already in use. It is desirable to possess a word that restores (herstellt) and expresses with emphasis the original meaning in contrast to the modified meaning. A real man in contrast to a painted man […] is nothing but a man in the proper sense of the term, that is, something that can rightly be acknowledged as man. (Marty 1908, 515)

As a member of the Brentano School, and as a reader of Bolzano, Husserl is familiar with the distinction between modifying and attributive terms when he writes the Logical Investigations. As we will see in the next section, Husserl has made ample use of the distinction, but also extended it beyond a linguistic and logical phenomenon, as he was willing to speak of modification also in cases where no modifying term occurs.

3

Modification in Husserl’s Logical Investigations

In the fifth Logical Investigation, Husserl speaks of a general phenomenon of modification. It is preceded by an investigation, in which modifying terms seem to be understood as changers of meaning. Husserl speaks of modifications of meaning (Bedeutungsmodifikationen) (Husserl 2009, IV, 329, § 11, the title). Husserl does not begin with the modifying terms, but with cases of suppositio materialis. In normal contexts, the particle ‘if ’

252 

M. van der Schaar

has a certain meaning, but if one uses the term in a context such as ‘“if ” is a particle’, it changes its meaning, for we are now speaking about the term ‘if ’. It has changed from a syncategorematic term into a categorematic term, and the syntactic category has thus also changed. When we say ‘“The earth is round” is a proposition [Aussage]’, we do not make a judgement about a state of affairs that the earth is round, but about a ‘declarative sentence’ (Aussagesatz). The sentence thus functions in an abnormal way as name of itself (Husserl 2009, IV, 331). Quotations marks thus function in the same way as modifying terms, according to Husserl. He is already generalising the function of modifying terms to cover a different phenomenon, in which modifying terms play no role. Husserl tries to capture a general linguistic phenomenon, in which terms may systematically be used in an abnormal way. We thereby reach a general phenomenon of modification, in which modifying terms need not occur. Crucial to Husserl’s understanding of modification is his claim that, from a logical point of view, all systematic ‘changes of meaning’ are to be understood as ‘abnormality’ (Abnormalität), (Husserl 2009, 330). These changes of meaning are not context-relative: they belong to language itself, to the general character of expressions. They are general in that we see them in all languages; they concern ‘allgemeingrammatische Äquivokationen’ (Husserl 2009, 333). These changes are thus not arbitrary. They point to a priori laws, according to which meanings are related to each other. They have their source in the ideal nature of meanings, as Husserl puts it. The relation between these different meanings concerns a characteristic of language in general, rather than a lexical issue. Terms are not ambiguous in the sense that different meanings are given to a term in a lexicon; the ambiguities Husserl speaks of are based on the universal grammar of all languages. He acknowledges, though, that on the level of ideal meanings there are no changes. What changes is rather our way of meaning, the way we use an expression to mean something; the changes concern ‘Änderungen des Bedeutens’ (Husserl 2009, 332). An essential aspect of our understanding of these modifications is that we are able to construct new concepts in many different ways from an original kernel concept (Husserl 2009, 332). In this sense there is an analogy, Husserl writes, with transformations of arithmetical structures (Gebilde). Although Husserl takes the ideal meanings to be prior to the

12  Modifying Terms and Modification in Husserl… 

253

explanatory order to our constructions, one can also give a phenomenological account of modification without presupposing that these ideal meanings are independent of the way we construct them. Husserl himself gives an account, at least, of the phenomenological side of these ideal meanings: these modifications ‘express unique relations between essences that are grounded in the phenomenological content of [our] experiences’ (Husserl 2009, V, 488). How are we to understand the distinction between normal and abnormal? According to Husserl, a term functions normally in its ‘original’ meaning. Terms like ‘originate’ are not to be understood in empirical-­ psychological terms, but rather in terms of ‘founding’ (Husserl 2009, V, 488). If the meaning of the modified complex derives from an origin, this means it is founded on the origin, thus showing a one-sided dependence relation. As Husserl puts it in the fifth Logical Investigation, what is founded thus points back to its foundation (zurückweisen) (Husserl 2009, 486). This does not mean that the foundation is somehow a part of what is founded, although one might say that the original is logically present in what is founded (Husserl 2009, 488). It rather means that one can grasp the derived meaning only if one has grasped the original one. These relations between meanings are grounded in the phenomenological content of our experiences, thus expressing essential relations (Husserl 2009, 488). Or, as one may put it, these relations between original concept and concept expressed by the modified complex are internal, and thus not to be determined by empirical investigations. As Husserl puts it, these foundation relations are a priori relations between different ideal intentional essences (Husserl 2009, § 35, 488). The term ‘normal’ does not seem to involve a normative element. The distinction can be understood as a distinction between standard and non-­ standard use. For non-standard uses the speaker has to give cues to the audience that he is using an expression in a non-standard way. We may use quotation marks or modifying predicates to indicate abnormal use. If one uses an expression without special signs in the context, it is supposed to have its normal use, and to refer to the original concept. If it is used in a context with special signs, it is supposed to have its abnormal use. The listener should be able to construct the specific abnormal uses from the special signs. The distinction between normal and abnormal may thus be

254 

M. van der Schaar

understood as a distinction between different uses of a term. It is crucial, though, for Husserl, that these different uses correspond to a priori relations between ideal meanings, and is therefore essentially a semantic phenomenon. And he is right insofar as these complexes have a different contribution to the truth-value of the sentence if they are used in an abnormal way. What we also see is that the operation of modification covers all kinds of modification, depending on the special signs used in the context, and is thus not to be understood in any well-defined sense, in the way we understand the negation to be an operator. In order to extend the linguistic phenomenon of modifying terms to an account of modification in general, Husserl starts with a familiar problem of how to understand the relation between perception and illusion. Contemporary philosophers of perception are beginning to understand that one cannot explain perception as a mere appearance plus an external relation to a state of affairs; and then explain illusion as a mere appearance without the plus. For Husserl, such an understanding of the relation does not fit our first-person experience. How can we understand the relation between perception and illusion from a first-person, phenomenological point of view? ‘Can we understand perception as an act-complex from which we can separate a mere presentation as a separate act?’ (Husserl 2009, V, § 27). No, Husserl says, and he gives an example. In the wax museum, we see a lovely lady on the stairs asking us to come in. At first, a) we perceive … a lady. Having recognised the delusion, b) we experience an illusion of seeing a lady. At the same time, we perceive a doll, who presents a lady. So, something completely different comes in the place of the perception in (a). How are we to understand the relation between a perception (what I experience as a perception), and an illusion (what I experience as an illusion)? According to Husserl, the perception does not consist in an act of mere presentation together with a belief moment. The perception is not

12  Modifying Terms and Modification in Husserl… 

255

a complex phenomenon in that sense. Case (a) is the simple and normal case, while (b), realising that the ‘perception’ is a mere illusion, is the complex, the abnormal case. ‘The mere presentation, the illusion, points to a deprivation [Mangel] […] There is a priority [Vorzug] on the side of the perception’ (Husserl 2009, V, § 28). One first has to understand what perception is; one then can understand illusion in terms of perception and a certain lack. If one is familiar with the Aristotelian tradition, one may think of the way in which blindness is explained. Blindness is understood as a privation with respect to man; not being able to see is not a privation with respect to stone. In order to understand what blindness is, one first has to understand what being able to see means. Being able to see comes first in the conceptual order. In the same way, perception is prior in the conceptual order to illusion. There is thus a conceptual relation between perception and illusion that is similar to the relation between man and dead man, although no modifying term is present. One may reconstruct it as a form of linguistic modification, when one says that illusion is to be understood as an illusive perception, in which case ‘illusive’ is used as a modifying term. Why is this problem relevant for Husserl in the Logical Investigations? Although perception is for Husserl not judgemental, the fact that perception does not include an act as a common element it shares with illusion, shows something important about the notion of judgement. In Logical Investigation V, Husserl aims to understand the relation between an act of judgement and an act of mere presentation, mere thinking. Does the act of judgement consists in an act of mere presentation together with a belief moment, as Brentano thought? Or, are we rather to understand the act of presentation in terms of the act of judgement? Husserl endorses the latter thesis: ‘To every judgement there belongs its modification: an act which merely presents what the judgement takes to be true’ (Husserl 2009, V, § 38). The two acts of judging and mere thinking are related as positing and non-positing: judging is thetic; merely thinking is not. They may have the ‘same’ content, while differing in quality. Strictly speaking the two acts have each a matter that are instantiations of the same ideal meaning. For Husserl, the act of mere thinking is to be understood in terms of the act of judgement. Act of judgement is prior in the conceptual order to mere thinking. Mere

256 

M. van der Schaar

thinking is to be understood as a modification of the act of judgement. This is called qualitative modification: a modification of the act. We have seen above that relations of modification are a priori relations between ideal meanings. If one is not willing to endorse Husserl’s logical realism, one can understand his point in terms of capacities. One is not able to realise the capacity, mere thinking, unless one is able to realise the other, judging (Husserl 2009, 489). Thus, there can be no beings who can think, but who cannot judge. For Husserl, the act of judgement does not consist in an act of presentation together with a belief moment. Judging does not involve two acts: an act of grasping the content and the act of acknowledging the content to be true, as Frege has claimed. Nor is it correct that each act of judging presupposes an act of presentation, which is to account for the intentionality of the act of judging, as Brentano would say. For Husserl, the judging itself, together with its matter, accounts for the intentionality of the act; no separate act of grasping or presentation is needed for that.

4

 Phenomenological Account A of Modification

In the past, I have given a logical account of modification.1 I still endorse this approach, although I now think that this logical account should be supplemented by an analysis of what happens on the level of presuppositions. This is not the place to elaborate this point. But I can give the logical properties of modifying terms. First, we have to make a distinction between the general class of non-attributive terms, which includes not only the modifying terms, but also relative, restorative and modal terms. ‘Small’ and ‘good’ are relative terms; ‘former’ and ‘past’ are modal terms. I come back to the restorative terms below. Non-attributive terms have a common logical behaviour. For all non-attributive terms A, belonging to a noun N, inference-schema (I) is invalid: (I) a is an A N. Therefore, a is an N & a is A.

12  Modifying Terms and Modification in Husserl… 

257

Thus, from the judgement that he is the former president, one is not entitled to draw the conclusion that he is the president. And, from the fact that she is a big mouse, it does not follow that she is big. Modifying or privative terms can then be understood as a subclass of non-attributive terms. They can be distinguished from other non-­ attributive terms by their distinct logical behaviour. A is a modifying or privative term, iff inference (II) is valid: (II) a is an A N. Therefore, a is not an N & a has the appearance of an N.

From the judgement that this is a toy-duck, one is entitled to draw the conclusion that this is not a duck, but has, or aims to have, the appearance of a duck. This account of modifying terms is incomplete, for we still do not understand what modification is. The modifying term signifies to the reader the presence of the phenomenon of modification, but how is he to understand a complex that might be completely new to him? Can we also give an account that explains the function of restorative terms, like ‘true’ in ‘true friend’? These terms do not have any straightforward distinctive logical behaviour, for a true friend is after all a friend. Still, they do qualify as non-attributive terms, for when we have made the judgement that he is a true friend, we are not entitled to draw the conclusion that he is true. Restorative terms are unique, for it only makes sense to say that he is a true friend, if there has been a suggestion that there is someone who is not a true friend. The negative use wears the trousers, as Austin put it. The idea of a non-true or false friend comes first in the conceptual order in relation to the idea of true friend. Below I develop a phenomenological account of modification that can be extended beyond the purely linguistic, giving both an account of modifying terms and of the idea of modification in general. Let me first explain how Scholastic philosophers understood the phenomenon, in order to see what is missing there.2 Their distinction between signification and supposition allows for a position where one can explain that no change of meaning takes place. Instead, in certain contexts, the noun is given a special supposition.

258 

M. van der Schaar

(1) Man consists of three letters. (Quotation marks did not exist at the time.) According to this position, the occurrence of the term ‘man’ in (1) has material supposition, standing for the word ‘man’ itself. The meaning as signification of a term does not change in different contexts; if a term has more than one signification, it is ambiguous, and this is precisely not the case here. A term has a fixed meaning or signification; it is rather that the occurrence of the term gets a special supposition due to the context of the sentence. The term ‘man’ does not get a different meaning in ‘this is a dead man’ or ‘this man is dead’, it is rather that the occurrence of ‘man’ does not have its standard supposition, due to the context of the sentence. One may thus argue that in the context of the sentence: (2) This is a fake gun, the supposition of the occurrence of ‘gun’ includes fake guns, while in the next context: (3) This is a gun. the occurrence of ‘gun’ does not include fake guns in its supposition. The fact that we are not entitled to infer (3) from (2) is accounted for: (2) may be true, while (3) is false. This proposal is, at least, incomplete: it sounds ad hoc; and, no explanation of modification is given. Why does the occurrence of ‘gun’ apply to fake guns in (2)? Furthermore, no account is given of the conceptual order, so essential to understand modification. No account is given of the a priori relations that obtain between the concepts or meanings false friend and friend. Nor is an account given of a related phenomenon, the special function of restorative terms, like ‘true’ in ‘true friend’. As Husserl has pointed out, these concepts are related on the level of ideal meanings; these relations do not differ for different subjects, and are thus not arbitrary, psychological relations between the ways we conceive these meanings. Does this mean that we have to be logical realists regarding ideal meanings, as Husserl himself was at the time he was writing the

12  Modifying Terms and Modification in Husserl… 

259

Logical Investigations? These meanings are to be understood as concepts, which can be constructed in the same way by each of thinking agent; they are thereby not dependent upon our acts insofar as they belong to this or that particular subject. The relations between these concepts are not arbitrary and subjective, but have their foundation in a general grammar and semantics that may belong to any language that thinking beings have developed. Or, to put it differently, they are founded in our logical capacity to think and reason. On this account, concepts can be understood as being constructed in an act of thinking, understood as an actualisation of such a capacity. These concepts and their relations need thus not to be understood as completely independent of our acts of thinking. An analysis in terms of construction of concepts can account for the a priori relations between friend and false friend, between knowing and botched knowing (belief ), and between perception and illusion. If we start with the concept of friend, taking a defining characteristic, such as to be trusted, away from it, and putting appearing to be a friend in its place, we obtain the concept false friend. We can then construct the concept true friend by taking away the characteristic appearing to be a friend, and substitute it by the original characteristic: to be trusted. The construction of friend is thus different from that of true friend. For this reason the two concepts are not identical; their construction is not an accidental fact belonging to these concepts, for it lays bare the a priori, internal relations between them. A similar account can be given of the relation between concepts like judging and mere thinking, which we understand without the use of any modifying term. The concept mere thinking is obtained by taking away a defining characteristic of judging, its positing, thetic character, and putting in its place appearing to be a judging. Mere thinking appears to be judging insofar as it has the same propositional form, the form of a declarative. A phenomenological account of modification would thus be able to: a) give an account of our ability to construct and understand new cases; b) give an account of the conceptual order of the relevant concepts; c) explain that restorative terms are highly complex: one cannot understand what a true friend is, unless one has understood what a false

260 

M. van der Schaar

friend is, and one cannot understand what a false friend is, unless one has understood what a friend is; d) account for the fact that the relations between these concepts are a priori; the analysis is not a psychological one, but allows for an involvement of the first person in the construction of concepts. e) Furthermore, it has no need to say that a change of meaning takes place; f ) and, can be used to supplement the logical account that was introduced above. Concepts may thus be analysed and related to each other in different ways than on a standard account of conceptual analysis.3

Notes 1. See (Schaar 2014, 2016) where I discuss the views on modifying terms proposed by linguists, such as Barbara Partee. 2. I refer the reader to Ebbesen (1979), for a more nuanced account of the different views. 3. This paper is based on a talk given at the conference Philosophy of Language in the Brentanian Tradition, Université de Liège, September 2018, at the German Philosophy Workshop, University of Chicago, March 2019, and in Leiden, September 2019, at a symposium dedicated to Göran Sundholm’s retirement. I thank the audiences for their inspiring questions and comments.

References Bolzano, Bernard. 1985–1987. Wissenschaftslehre [1837], ed. Jan Berg. In Bernard Bolzano-Gesamtausgabe, Vols. I, 11, 1–2, ed. Eduard Winter and Jan Berg. Stuttgart-Bad Cannstatt: Frommann. Brentano, Franz. 1968. Über Wahrnehmung modo recto, modo obliquo und die Zeitwahrnehmung [1914]. In Psychologie vom empirischen Standpunkt, vol. III. part 1, chapter 5, ed. Oskar Kraus, 37–52. Hamburg: Felix Meiner. ———. 1971. Psychologie vom empirischen Standpunt [1874]. Vol. 2. Hamburg: Felix Meiner.

12  Modifying Terms and Modification in Husserl… 

261

———. n.d. Ms. EL 80. Logik. Claas, Jan, and Benjamin Schnieder. 2019. Determining and Modifying Attributes. In Anton Marty and Contemporary Philosophy, ed. Giuliano Bacigalupo and Hélène Leblanc, 59–96. London: Palgrave Macmillan. Ebbesen, Sten. 1979. The Dead Man is Alive. Synthese 40: 43–70. Husserl, Edmund. 1913. Ideen zu einer reinen Phänomenologie und phänomenologischen Philosophie, Book 1, ed. Karl Schuhmann. The Hague: Nijhoff, 1977. ———. 1929. Formale und Transzendentale Logik. Halle: Max Niemeyer. ———. 1973.  Ding und Raum; Vorlesungen 1907, ed. Ulrich Claesges. The Hague: Martinus Nijhoff. ———. 1985. Erfahrung und Urteil [1939], ed. Ludwig Landgrebe. Hamburg: Felix Meiner. ———. 1987. Cartesianische Meditationen [1929],  ed. Elisabeth Ströker. Hamburg: Felix Meiner. ———. 2009. Logische Untersuchungen [1900-01],  ed. Elisabeth Ströker. Hamburg: Felix Meiner. Kripke, Saul. 2011. Unrestricted Exportation and Some Morals for the Philosophy of Language. In Philosophical Troubles. Collected Papers, vol. 1, 322–350. Oxford and New York: Oxford University Press. Marty, Anton. 1908. Untersuchungen zur Grundlegung der allgemeinen Grammatik und Sprachphilosophie. Halle: Niemeyer. van der Schaar, Maria. 2014. Wooden Horses and False Friends: On the Logic of Adjectives. In The History and Philosophy of Polish Logic: Essays in Honour of Jan Woleński, ed. Kevin Mulligan, Katarzyna Kijania-Placek, and Tomasz Placek, 103–116. Basingstoke: Palgrave Macmillan. ———. 2016. Kazimierz Twardowski: A Grammar for Philosophy. Leiden, Boston: Brill and Rodopi. Twardowski, Kazimierz. 1927. On the Logic of Adjectives. In On Actions, Products and Other Topics in Philosophy, ed. Johannes L.  Brandl and Jan Woleński, 141–143. Amsterdam: Rodopi, 1929. ———. 1977. On the Content and Object of Presentations. A Psychological Investigation [1894].  Trans. Reinhardt Grossmann. The Hague: Martinus Nijhoff. Williamson, Timothy. 2000. Knowledge and Its Limits. Oxford, New  York: Oxford University Press.

13 The Early Husserl on Typicality Hamid Taieb

1

Introduction: Some Typicalities

As philosophers, we are—or at least we have been for a very long time— the victims of a ‘definitionalist prejudice’, in both our semantics and our theory of concepts. We have tended to think that our (common) names refer to things by expressing meanings, or concepts, which are structured like definitions. On this view, ‘man’, for example, refers to men by the intermediary of a meaning, or concept, which represents a series of ‘necessary and sufficient’ features that all and only men possess. In this respect, all items meant by the name, or gathered together by the concept, are strictly identical. This vision of the meanings of words and of the content of concepts as having ‘definitional structure’ is said to be inherited from Plato; at any rate, it has circulated for a very long time among philosophers without being contested (on all this, I follow Margolis and Laurence 1999, 2019; for the reference to Plato, see 1999: 10).

H. Taieb (*) Fakultät für Geisteswissenschaften, Philosophisches Seminar, Universität Hamburg, Hamburg, Germany © The Author(s) 2021 A. Dewalque et al. (eds.), Philosophy of Language in the Brentano School, History of Analytic Philosophy, https://doi.org/10.1007/978-3-030-52211-7_13

263

264 

H. Taieb

In the second half of the twentieth century, however, this image came under intense criticism. Some cognitive scientists defended the thesis that our concepts, and by their intermediary our names, gather together not identical things, but more or less similar ones, which are organised around a ‘prototypical case’—that is, a sort of standard which things within the extension resemble more or less. They also held that the extension of such concepts—called ‘prototype concepts’—is vague, since things can be less and less similar to the prototype without there being any sharp boundary that can be drawn as to where something is no longer included in the class. This theory is meant to develop Wittgenstein’s notion of ‘family resemblances’, which states that our concepts have—and our words refer to—an extension made of more or less similar items and which lacks sharp boundaries (Rosch and Mervis 1975 and Hampton 1995, following Wittgenstein 1953; on this whole narrative about concept structure, see again Margolis and Laurence 1999, 2019). An interesting development about prototype concepts is the thesis that they are context-sensitive. Depending on the context in which we are using a name and mobilising a given concept, the standard that we use to fix the extension will vary. For example, if someone tells us that ‘Stacey volunteered to milk the animal whenever she visited the farm’, we will take a cow as the model of an animal, whereas when someone says ‘Frank asked her father to let her ride the animal’, the model that we will take is a horse. Thus, in the first case, the concept will be about ‘cow-like living beings’, but in the second case it will be about ‘horse-like living beings’. (The view and the examples are from Roth and Shoben 1983, quoted and discussed in Del Pinal 2016, 2911–2912; see also Prinz 2012.) This not only shows that the definitionalist is wrong to think that we have a strictly delimited series of features in mind when applying names and concepts, since we have at best typical cases and blurred boundaries around them. It also reveals that even our models for the fixing of extension vary from one situation to another. According to this view, a concept is a ‘rich array of information’ stored in ‘long-term memory’ and ‘the subsets of features activated in each token use can vary’ (Del Pinal 2016, 2912). In other words, prototype concepts are mental representations: they are understood as cognitive ‘entries’ storing ‘information’. So in the debate on the ontology of concepts, when it is asked whether concepts are Fregean-like

13  The Early Husserl on Typicality 

265

abstract entities or rather mental representations, defenders of the prototype view clearly opt for the latter (For an overwiew of this debate, see Margolis and Laurence 2019). As indicated, the usual source which is mentioned when people treat of the origins of the prototype theory is Wittgenstein’s notion of family resemblances (Rosch and Mervis 1975). However, as emphasised by Mulligan (1990, 2012), Wittgenstein’s theory of family resemblances had been anticipated by Austro-German authors, in particular Brentano and his heirs. More specifically, the very idea that the reference of names and the extension of concepts is fixed on the basis of typical cases had been explicitly defended by authors in the Austro-German tradition long before Wittgenstein: one finds in Marty (1908, 527–531), at the turn of the twentieth century, the idea that (common) names in ordinary language have as their meaning a ‘concept built following a type’. Similarly, the late Husserl holds that our non-scientific concepts are ‘type concepts’ expressed by ordinary language (1939, 2012).1 Alexius Meinong (1969, 480–492) also hints at a similar theory of concepts. In fact, the idea that our concepts might not have a definitional, but rather a typical structure was quite widespread in the German-speaking world. It was seemingly introduced by the German philosopher Benno Erdmann (1894), who himself relied on the work of William Whewell (1858). Besides Marty, the late Husserl, and Meinong, it seems that prototype concepts were accepted by Wilhelm Jerusalem (1902, 98–103) and by Johannes von Kries (1916, 10–13).2 Interestingly, the early Husserl, in his first Logical Investigation, also defends the view that our ordinary language is sensitive to typicality. More precisely, he holds that our (common) names refer to more or less similar things organised around a typical case, which is itself contextdependent, and that the limit of their application is vague. Apart from Benoist (2010), his views have never been compared with contemporary prototype theories. It is notable, however, that the early Husserl explains the same phenomena that the prototype theory is meant to explain, but without admitting type concepts: rather, he thinks that typicality is a linguistic feature found at the level of ordinary language, and that it is connected with fluctuations in our acts of meaning or so-called ‘meaning-intentions’. Concepts, he maintains, are abstract, ‘ideal’ entities

266 

H. Taieb

which have definitional structure. This is an interesting move, as it seems to suggest that what people have taken to be a feature of our concepts might in fact be a merely linguistic phenomenon which can be explained while leaving the structure of concepts untouched. Moreover, in order to explain it, it is not necessary to treat concepts as mental representations, for typicality can still be accounted for even if, from an ontological point of view, concepts are described as abstract entities. In this paper, I will present and evaluate the early Husserl’s account of typicality. In the first section, I will briefly introduce the later Husserl’s view on types, in order to contrast this with his earlier account. In the second, longest section, I will present in detail the early Husserl’s views on typicality. In the third section, I will evaluate his position, by asking whether his earlier account can do without prototype concepts. I will answer that it cannot, and that this might explain why Husserl later adopted a theory of type concepts.

2

 Brief Excursus: The Later Husserl A on Type Concepts

The late Husserl thinks that the concepts that we build in our ordinary life are ‘type concepts’ (Typenbegriffe) (Husserl 1939, 2012; on his theory, see Lohmar 1998 and Bégout 2002). These concepts pick out ‘commonalities’ (Gemeinsamkeiten) in ‘open groups of similarities’ (Husserl 2012, 233.25–32). Their extension seems to be a network of more or less similar things: Husserl states that types have a ‘horizon of sameness, of similarity’ (Husserl 2012, 388.8–10). But although these concepts group together more or less similar items, it is unclear where to put their exact boundary, that is, the level of similarity at which something will no longer be included in the class. Husserl describes these types as ‘vague generalities’, holding that their extension lacks determinate boundaries (Husserl 1939, 114; see also Bégout 2002, 60). The later Husserl treats type concepts as psychic items, since they are mental representations built on the basis of our ordinary experience (Husserl 1939). Type concepts are opposed to abstract, ideal concepts,

13  The Early Husserl on Typicality 

267

which are grounded on the identity (rather than similarity) of their members and are exact (rather than vague), that is, they have sharply delimited extensions (Husserl 2012). While it is clear that typicality is a feature of concepts, the late Husserl also claims that it is mirrored at the linguistic level. In fact, he holds that type concepts find their expression in ‘natural language’ (natürliche Sprache) (2012, 276.15 and 315.21–23).3 Although I have not found a text where this is clearly stated, his idea might be that our ordinary language (common) names refer to more or less similar items and have vague boundaries of application. At any rate, this would be in line with his earlier views on typicality, developed in the Logical Investigations, where he states that the meaning of our ordinary (common) names is sensitive to typicality—that is, it is based on more or less similar typical examples—and has vague boundaries. However, the interesting point is again that the early Husserl explains typicality without admitting type concepts. The only concepts that exist are ideal concepts, with definitional structure. This is a more parsimonious view, which I will now present, before considering whether the view is in fact too parsimonious, and whether type concepts are still required in order to account for typicality at the linguistic level.

3

The Early Husserl’s Account of Typicality

The context in which Husserl discusses typicality are the chapters in the Logical Investigations devoted to ‘ambiguous expressions’. Ambiguous expressions include ‘essentially occasional expressions’, that is, indexicals, whose meanings are constantly renewed in different situations of utterance, and ‘equivocal terms’, that is, words with different meanings, but which are limited in number and fixed in advance and not on the basis of the situation—for example, ‘dog’, ‘which sometimes means a type of animal, and at other times a foot or a grate’ (see §§ 24–29 of the first Logical Investigation in Husserl 1984, transl. Findlay). In addition to ‘essentially occasional expressions’ and ‘equivocal terms’, Husserl (1984, 93.5–94.6) counts among ambiguous expressions ‘vague’ ones, namely (common) names used in daily life, such as ‘tree’, ‘shrub’,

268 

H. Taieb

‘animal’, and ‘plant’. The meaning of such terms is said to ‘fluctuate’ (schwanken). Their meaning depends on ‘typical examples’: Vague expressions have no single meaning-content, the same in all cases of their application: their meaning is oriented towards typical examples, only partially conceived with clearness and definiteness, typical examples which tend to vary from case to case, perhaps even in a single train of thought. The examples, stemming from what are, or from what seems to be, genuinely unified fields, yield a number of concepts, more or less cognate or related, which emerge in turn according to the circumstances of our talk and its varied thought-promptings. (Husserl 1984, 93.11–21; transl. Findlay, slightly modified)

Note that Husserl makes a distinction between the meaning of what he calls a ‘universal name’ (universeller Name) and its extension: the meaning is a concept, whereas the extension is made up of the things that fall under the concept (1984, 51.25–55.33). As regards vague names, their meaning seems to vary following ‘typical examples’ (typische Beispiele). But what are exactly these typical examples? Unfortunately, Husserl does not say much more about this. However, when he discusses these issues, he gives as reference on types a paper by Erdmann (1894; see Husserl 1984, 94, note 1). Now, in the paper in question, Erdmann does not talk of Beispiele, but of Muster, translating with this term the word ‘example’ found in Whewell (1858). What are ‘examples’ for Whewell? According to him, every class is organised around a type, which is ‘an example […] eminently possessing the characters of the class’. This might be a ‘species’ for a ‘genus’, or a ‘genus’ for a ‘family’, that is, it is a general entity directly subordinated to the class of which it is the type.4 So, a genus, for example, is made up of a series of species, with one of them being the type, and the others resembling it more or less, with no clear criterion as to where to stop the resemblance series (1858, II, 121–122). Is the same point to be found in Husserl? Husserl holds that the meanings of words such as ‘animal’ vary and are ‘oriented towards typical examples’. He adds that the ‘examples […] yield a number of concepts, more or less cognate or related’ (diese Beispiele […] bestimmen

13  The Early Husserl on Typicality 

269

verschiedene, aber in der Regel verwandte oder beziehungsvolle Begriffe). So, apparently the meaning of the term ‘animal’ varies and orients itself following examples to which there correspond concepts. It would certainly be a reasonable interpretation to hold that the examples in question are, e.g. cow or horse, as in Whewell—that is, general entities directly subordinated to the class of which they are typical examples. Husserl says that the concepts which correspond to the typical examples are ‘more or less cognate or related’. Does he mean that they are more or less similar in terms of content? In that case, the word ‘animal’, for example, would have as its meaning a series of more or less similar contents: ‘cow’, ‘horse’, etc. Seemingly, a word such as ‘animal’ would then refer to more or less similar things because it means more or less similar contents, which vary depending on the context. At any rate, Husserl’s idea seems to be that ‘animal’ means ‘cow’, or ‘horse’, or some other kindred concept, and that the fixing of the exact meaning depends on the context of utterance.5 Husserl holds that the application of words of ordinary language comes with borderline cases. He seems to do so when he talks of ‘expressions standing for relatively simple genera and species of phenomenal properties’, such as names of colours: Within certain ranges and limits their application is unhesitant, i.e. in fields where the type appears clearly, where it can be evidently identified and evidently distinguished from remotely unlike characters, e.g. ‘signal-­ red’ and ‘coal-black’, andante and presto. But these fields have vague borders, and flow over into correlative spheres comprehended in the same genus, and so give rise to transitional regions where application varies and is wholly uncertain. (Husserl 1984, 93.34–94.6, transl. Findlay)

Clearly, then, ‘vague names’ are vague because the domain of their application has blurred boundaries: that is, there are cases in which it is unclear whether or not the word applies to, for example, some specific reddish colour. Now, Husserl has a problem with these fluctuations of meaning. Indeed, he has a theory of ‘ideal meanings’. Ideal meanings are, for Husserl, atemporal, abstract objects, divided into concepts and

270 

H. Taieb

propositions, which constitute, respectively, the meanings of (common) names and of sentences. These meanings are instantiated as parts of our mental acts when we use names and sentences. As ideal entities, however, they exist independently of being instantiated, and as such are Platoniclike universals (see, among others, 1984, 104.21–106.32 and 617.34–618.1, as well as Simons 1995). Now, since these meanings are independent of their being thought, it is unclear in what sense there could be any sort of ‘fluctuation of meaning’. Aren’t meanings stable, or ‘fixed’ (fest), atemporal entities? But then, what is to be said about ambiguous expressions, including terms of ordinary language, whose meanings change depending on the context? Hence there arises a problem, one which is explicitly addressed by Husserl: We now have to consider whether these important facts of fluctuation of meaning are enough to shake our conception of meanings as ideal (i.e. rigorous) unities, or to restrict its generality significantly. Those ambiguous expressions we called essentially subjective, in particular, as also our distinction between vague and exact expressions, might make us doubtful on this point. Do meanings themselves divide into objective and subjective, into meanings fixed (feste) and meanings changeable on occasion? Must we, in other words, so interpret this difference, with seeming obviousness, that it becomes one between meanings that are ideal unities, on the one hand, fixed species untouched by the flux of our subjective picturing and thinking, and such, on the other hand, as live submerged in the flux of subjective mental experiences, and are transitory events, at one time there, and at the next moment not? (Husserl 1984, 94.18–95.3, transl. Findlay)

Husserl’s answer is straightforward: ‘variations in meanings’ (das Schwanken der Bedeutungen) are in fact ‘variations in the act of meaning’ (Schwanken des Bedeutens), also called ‘meaning-intentions’ (Bedeutungsintentionen). In other words, what fluctuates are our meaning-­ intentions, not the meanings themselves. How should we understand this? First of all, the meaning meant in this or that case is always an ideal meaning. In fact, Husserl thinks that even ‘essentially occasional expressions’—that is, indexicals—could be replaced by objective expressions. All circumstance-sensitive words (‘subjective

13  The Early Husserl on Typicality 

271

expressions’) have equivalent circumstance-unsensitive words (‘objective expressions’) in terms of which they could be expressed, which would establish a firm relation between the word and the ideal meaning, connected by one and the same meaning-intention: […] the content meant by the subjective expression, with sense oriented to the occasion, is an ideal unit of meaning in precisely the same sense as the content of a fixed expression. This is shown by the fact that, ideally speaking, each subjective expression is replaceable by an objective expression which will preserve the identity of each momentary meaning-intention. (Husserl 1984, 95.5–11, transl. Findlay)

Husserl’s claim is that everything that is pointed at by means of language is not just describable in relation to the occasion of utterance, but absolutely determinable. ‘This’, ‘that’, ‘here’, ‘there’ are all translatable into objective expressions: What is, has its intrinsically definite properties and relations, and if it has natural, thinglike reality, then it has also its quite definite extension and position in space and time, its quite definite ways of persisting and changing. But what is objectively quite definite, must permit objective determination, and what permits objective determination, must, ideally speaking, permit expression through wholly determinate word-meanings. (Husserl 1984, 95.20–29, transl. Findlay)

In short, the linguistic phenomenon of ambiguity in general is avoidable, at least ‘ideally speaking’, since Husserl (1984, 95.12–16, transl. Findlay) recognises that this is ‘impracticable’, or even impossible—by this he probably means impossible for us human beings, perhaps due to the fact that such determinations are infinite. At any rate, this ‘objectivability’ applies all the more to vague names. Although the meaning of ‘animal’ changes depending on the context, it is in each case an ideal, stable meaning that is expressed. What seems to happen is that the meaning switches from one typical case to another depending on the context, so that ‘animal’ means, in turn, ‘cow’, ‘horse’, etc. On such a view, there is no distinction between meanings understood

272 

H. Taieb

as ‘fixed species untouched by the flux of our subjective picturing and thinking’ and meanings understood as ‘transitory events, at one time there, and at the next moment not’ and ‘submerged in the flux of subjective mental experiences’. Husserl can thus say: ‘Plainly therefore, considered as such, meanings do not differ essentially among themselves’ (Husserl 1984, 96.14–16, transl. Findlay). But then how can we make sense of the claim that vague names ‘change their meaning’ from one context to the other? As indicated above, what is in fact changing is not the meaning itself, but the meaning-intention, or the ascription of meanings to words: Actual word-meanings are variable, often changing in a single spell of thought, by their nature mainly adjusted to the occasion. Rightly seen, however, such change in meanings is really change in the act of meaning. In other words, the subjective acts which confer meaning on expressions are variable, and that not merely as individuals, but, more particularly, in respect of the specific characters in which their meaning consists. (Husserl 1984, 96.16–23, transl. Findlay)

The idea seems to be that a speaker can ascribe different meanings to ‘animal’ (for example) on different occasions; these variations, however, do not affect the nature of the meanings themselves, which all remain fixed. It seems clear that for Husserl, the concepts which are meant by the various uses of a vague name have definitional structure and are sharp. The meanings which are mobilised when a vague name is used are based on different typical examples. As Husserl says, these examples ‘yield a number of concepts, more or less cognate or related’. Now, these concepts clearly seem to be both definitional and sharp: they are ‘firmly determinate word-meanings’ and they represent ‘firmly determinate properties and relations’ of things; the words which mean them are ‘expressions having exact meanings’ (Husserl 1984, 95.17–96.3, transl. Findlay, slightly modified). In short, all objects are made up of a precise number of determinate features and the concepts which represent these objects mirror these features exactly. Meanings are thus definitional and sharp concepts. For the early Husserl, there is nothing like a type concept.

13  The Early Husserl on Typicality 

4

273

Evaluation

The early Husserl makes room for certain phenomena that the prototype theory of concepts is also meant to explain. He accepts that in ordinary language, the meaning of our (common) names depends on typical cases which vary according to the context. He seems to think that these various cases are more or less similar to one another, or at least that they are ‘more or less cognate or related’. He also holds that the limits of the application of these words are vague, since they are cases in which it is not clear whether the word still applies or not. Strikingly, whereas contemporary philosophers modify the structure of our concepts to account for this, Husserl maintains that all these linguistic phenomena can be explained by means of fluctuations in meaning-intentions. This is an interesting solution, inasmuch as it tries to explain in purely linguistic terms phenomena that people usually try to account for by modifying the structure of our concepts. Indeed, for Husserl, one and the same word has various meanings (e.g. ‘animal’ means ‘cow’, ‘horse’, etc.), depending on the context. These variations are attributed to the shifts in the meaning-­intentions, or the ascription of meanings to words, while the meanings themselves which are expressed are all definitional and sharp concepts. But is Husserl’s avoidance of type concepts plausible when explaining typicality? All things considered, I would say that it is not. First of all, one point which is not clear in Husserl’s text is whether the meaning of, for example, ‘animal’ is equivalent to that of the names of typical examples, such as ‘cow’, ‘horse’, etc., depending on the various contexts of utterance, or whether it is broader than these specific names, which merely fix the standard of resemblance, in which case the word will mean in turn ‘cow-like living beings’, ‘horse-like living beings’, etc. (as in Roth and Shoben 1983 and Del Pinal 2016, quoted in the Introduction above). I would tend to go for the second option, for otherwise ‘animal’ on the one hand, and ‘cow’, ‘horse’, etc. on the other, would be mere synonymous terms in various contexts, what would be an unfortunate result, since the first is obviously the name of a genus, the others of species. It seems therefore that the meaning of vague names should include some sort of open range of similarities. However, the early Husserl owes us an

274 

H. Taieb

explanation of how exactly this will work, all the more so in the context of a theory which admits ideal meanings, which have definitional structure and are sharp. Moreover, and independently of this issue, it seems to me that the way vague names are supposed to work in Husserl in fact requires the existence of underlying type concepts. Husserl’s account seems to imply that to every vague name there corresponds some information stored in the memory of the utterer. Indeed, he would agree that people apply ‘animal’ to cows, horses, etc., but not to, say, Belmag lamps. However, this store of information, which gives boundaries to the meaning-intention, seems to be a blind spot in Husserl’s account, as he does not explain either what it is or how it is structured. Now, it seems legitimate to ask whether this store is in fact a sort of concept. Note that, as Husserl himself admits, the ambiguity of vague names is not a mere equivocation. Rather, the different meanings of vague names are intimately related to one another as members of one and the same genus. Husserl himself seems to point this out when he says that vague names apply to ‘correlative spheres comprehended in the same genus’, or when he holds that the typical examples ‘stem from what are, or from what seems to be, genuinely unified fields’. But then, isn’t the store of information which allows us to ‘keep in mind’ that these species have an intimate connection a generic concept?6 Recall that in contemporary philosophy, a type concept is a ‘rich arrays of information’ stored in ‘long-term memory’, and that ‘the subsets of features activated in each token use can vary’ (Del Pinal 2016, 2912). In Husserl, one vague name means a series of ‘cognate or related’ typical examples which vary according to the context. So could he deny that the information about the meaning of a word is stored in memory and forms a sort of concept? In addition, Husserl has a problem with vagueness, which seems impossible to explain in his framework. That is, though he wants ordinary language to be vague, it is not clear how he can account for its vagueness. He tends to say that a vague name has various meanings, which change according to the context; however, the combination of various similar sharp concepts does not create vagueness (Sainsbury 2012). Think of a series of cookie cutters that you press on a pastry: you will end up having one or many bizarre forms, but nothing like vague

13  The Early Husserl on Typicality 

275

boundaries. It seems then that if one wants to account for vagueness, one has to renounce sharp concepts, and introduce non-sharp ones which will allow for an open structure of relations of similarity. The early Husserl has no such concepts, and yet he speaks of ‘vague borders’ and ‘transitional regions where application varies and is wholly uncertain’, which seems to appeal to non-sharp conceptual contents. One way to explain this in the prototype theory is to hold that concepts have an undetermined ‘similarity comparison process’ which is built into their very structure and which leads to vagueness: one compares cases to the prototype, and includes the more similar cases within the category, but without knowing where exactly inclusion should stop. This might explain Husserl’s later introduction of type concepts: their specific structure is able to account for the fact that every type implies a ‘horizon of similarity’ and is a ‘vague generality’.7 To conclude, in order to account for the kind of fluctuation of meaning-­intentions that Husserl is interested in, the best move would be to accept concepts of a sort that very much resembles what contemporary philosophers call ‘prototype concepts’. Such a consideration might be among the reasons that led the later Husserl to hold that there are type concepts in addition to ideal concepts.8

Notes 1. On these issues in Marty, as well as their anticipation in Brentano (see Taieb 2020). On similar topics in the later Husserl, see “The Structure and Domain of (Proto)type Concepts: A Husserlian Approach” (Taieb Ms.). In the present article, I focus on the early Husserl, whose theory, interestingly, differs not only from those of Brentano and Marty, but also from his own later account. 2. I thank Mark Textor for the reference to Jerusalem. On Von Kries, see the various discussions in (Mulligan 2012). 3. As indicated, I explore Husserl’s theory of type concepts in my paper “The Structure and Domain of (Proto)type Concepts: A Husserlian Approach”. 4. This resembles the way prototypes are understood in contemporary philosophy, namely, as subordinated concepts (see Margolis and Laurence

276 

H. Taieb

1999, 28, note 35). Although Whewell does not say what the item in question would be for a most specific species (e.g. English Cocker Spaniel), one option would be to hold that it is still a general entity, although not a species in the sense of a natural kind (e.g. an English Cocker Spaniel of a certain colour and shape). 5. Note that Husserl has a limited account of contextualism: the variation of meaning seems to be restricted to entities directly subordinated to the same class. There is a stronger sense of contextualism in contemporary philosophy of language, in which there is no limitation as to the variation that the meaning of a word can undergo in different contexts (on contextualism, see Recanati 2012). 6. See also (Benoist 2010, 135) who emphasises that ‘vague names’ in the early Husserl are not mere equivocal terms, and that this hints at the acceptance of a non-standard type of concept from which their meaning is constructed. According to Benoist, the early Husserl in fact anticipates the prototype theory. In my opinion, the early Husserl would be unhappy with the admission of prototype concepts, although I think that they are an unavoidable consequence of his theory. 7. On how this ‘similarity comparison process’ would be able to explain vagueness, see (Barsalou 1987, 116; Margolis and Laurence 2019; as well as Hampton 2007, 379). For more on this in Husserl’s later account of types, see again my paper “The Structure and Domain of (Proto)type Concepts: A Husserlian Approach”. 8. This paper was written in the context of an Alexander von Humboldt postdoctoral fellowship on “Prototype Concepts in Austro-German Philosophy”. A first draft of the paper was presented at the conference “Philosophy of Language in the Brentanian Tradition”, held in 2018 at the University of Liège. I thank the participants for their comments, especially Kevin Mulligan, Sébastien Richard, and Basil Vassilicos. I am also grateful to Mark Textor for the discussion that we had on an earlier draft of this paper. Finally, I thank Charlotte Gauvry for her written comments on an earlier draft.

13  The Early Husserl on Typicality 

277

References Barsalou, Lawrence W. 1987. The Instability of Graded Structure: Implications for the Nature of Concepts. In Concepts and Conceptual Development: Ecological and Intellectual Factors in Categorization, ed. Ulric Neisser, 101–140. Cambridge: Cambridge University Press. Bégout, Bruce. 2002. Un air de famille : la théorie husserlienne des types. Recherches husserliennes 17: 51–85. Benoist, Jocelyn. 2010. Concepts. Introduction à l’analyse. Paris: Cerf. Del Pinal, Guillermo. 2016. Prototypes as Compositional Components of Concepts. Synthese 193: 2899–2927. Erdmann, Benno 1894. “Theorie der Typen-Einteilungen”. Philosophische Monatshefte 30/3–4: 15–49 and 129–158. Hampton, James A. 1995. Concepts as Prototypes. The Psychology of Learning and Motivation 46: 79–113. ———. 2007. “Typicality, Graded Membership, and Vagueness”. Cognitive Science 31: 355–384. Husserl, Edmund. 1939. Erfahrung und Urteil. Untersuchungen zur Genealogie der Logik, ed. Ludwig Landgrebe. Prague: Akademia. ———. 1984. Logische Untersuchungen. Zweiter Teil (Hua XIX/1–2), ed. Ursula Panzer. The Hague: Martinus Nijhoff. Trans. John N.  Findlay as Logical Investigations. London: Routledge, 1970. ———. 2012. Zur Lehre vom Wesen und zur Methode der eidetischen Variation (Hua XLI), ed. Dirk Fonfara. New York: Springer. Jerusalem, Wilhelm. 1902. Lehrbuch der Psychologie. 3rd ed. Vienna: Braumüller. Lohmar, Dieter. 1998. Erfahrung und Kategoriales Denken. Hume, Kant und Husserl über vorprädikative Erfahrung und prädikative Erkenntnis. Dordrecht: Kluwer. Margolis, Eric, and Stephen Laurence. 1999. Concepts and Cognitive Science. In Concepts: Core Readings, ed. Eric Margolis and Stephen Laurence, 3–81. Cambridge, MA: MIT Press. ———. 2019. Concepts. In Stanford Encyclopedia of Philosophy (Summer 2019 Edition), ed. Edward. N.  Zalta. https://plato.stanford.edu/archives/ sum2019/entries/concepts. Marty, Anton. 1908. Untersuchungen zur Grundlegung der allgemeinen Grammatik und Sprachphilosophie. Erster Band. Halle: Niemeyer.

278 

H. Taieb

Meinong, Alexius. 1969. Abstrahieren und Vergleichen. In Gesamtausgabe, ed. Rudolf Haller and Rudolf Kindinger, vol. 1, 443–494. Graz: Akademische Druck- u. Verlagsanstalt. Mulligan, Kevin. 1990. Marty’s Philosophical Grammar. In Mind, Meaning and Metaphysics. The Philosophy and Theory of Language of Anton Marty, ed. Kevin Mulligan, 11–27. Dordrecht: Kluwer. ———. 2012. Wittgenstein et la philosophie austro-allemande. Paris: Vrin. Prinz, Jesse. 2012. Regaining Composure: A Defence of Prototype Compositionality. In The Oxford Handbook of Compositionality, ed. Wolfram Hinzen, Edouard Machery, and Markus Werning, 437–453. Oxford: Oxford University Press. Recanati, François 2012. “Contextualism: Some Varieties”. In The Cambridge Handbook of Pragmatics, eds. Keith Allan and Kasia M. Jaszczolt, 135–150. Cambridge: Cambridge University Press. Rosch, Eleanor, and Carolyn Mervis. 1975. Family Resemblances: Studies in the Internal Structure of Categories. Cognitive Psychology 7: 573–605. Roth, Emilie M., and Edward J. Shoben. 1983. The Effect of Context on the Structure of Categories. Cognitive Psychology 15: 346–378. Sainsbury, Richard M. 2012. Concepts Without Boundaries. In Vagueness: A Reader, ed. Rosanna Keefe and Peter Smith, 251–264. Cambridge, MA: MIT Press. Simons, Peter M. 1995. “Meaning and Language”. In The Cambridge Companion to Husserl, eds. Barry Smith and David W. Smith, 106–137. Cambridge: Cambridge University Press. Taieb, Hamid. 2020 Ordinary Language Semantics: The Contribution of Brentano and Marty. British Journal for the History of Philosophy 28: 777–796. ———. n.d. Ms. The Structure and Domain of (Proto)type Concepts: A Husserlian Approach. Von Kries, Johannes. 1916. Logik. Grundzüge einer kritischen und formalen Urteilslehre. Tübingen: Mohr. Whewell, William. 1858. The History of Scientific Ideas. Vol. 2. London: Parker. Wittgenstein, Ludwig. 1953. Philosophical Investigations, ed. Elizabeth Anscombe and Ruth Rhees. Trans. Elizabeth Anscombe. Oxford: Blackwell.

14 Wundt and Bühler on Gestural Expression: From Psycho-Physical Mirroring to the Diacrisis Basil Vassilicos

1

Introduction

Contemporary research on gestures—as an empirical, interdisciplinary body of work involving anthropology, linguistics, philosophy, and psychology—can be characterised as having a couple of overarching traits. In the first place, it seems disposed toward a kind of empirical but also egalitarian prudence when it comes to the relation of gesture to those communication phenomena typically designated as having to do with language or speech. With increased attention being given to gestured (e.g. signed) languages as robust and wholly developed communicative systems in their own right (i.e. not merely derivative of other linguistic systems) has also come a reluctance to reduce gestural expression to verbal expression, or vice versa.1 A second feature has to do with how gestures are delineated as ‘integral’ phenomena for scientific investigation in their own right, namely as fully fledged ‘partners’ if not self-sufficient

B. Vassilicos (*) Mary Immaculate College, University of Limerick, Limerick, Ireland e-mail: [email protected] © The Author(s) 2021 A. Dewalque et al. (eds.), Philosophy of Language in the Brentano School, History of Analytic Philosophy, https://doi.org/10.1007/978-3-030-52211-7_14

279

280 

B. Vassilicos

components within any human communicative interaction, with the potential if not requirement to contribute just as much as that other much more well known and oft-studied partner, i.e. verbal language. That is, there is a strong focus today on the multi-modality of human communicative action, meaning, depending on one’s point of view, co-­ speech gestures or co-gestured speech. A simple example can illustrate this multimodal approach to gestures and language. Two people are bent over a bit of exposed dirt in a field, in discussion with each other about something they are excavating.2 In so communicating, they might use words, but also their hands and the tools they are holding. For instance, the one person might complete a verbal phrase with a gesture with her right hand that traces an outline in the dirt. In response, the other might show an understanding of this tracing as the joint object of scrutiny by repeating the same pointing gesture and then take it as the point of departure for both her own gesture with her trowel and her further verbal communication. Two things are noteworthy here. First, if by ‘gesture’ what is meant is body information provided in this example of communicative interaction, then only certain types of body information seem relevant to be being counted as gestures: namely, those that are express or intentional movements or configurations of the body, and not such things as one’s rate of pulse or breathing, one’s pallor, or one’s twitches. This is nonetheless a fluid and pragmatic distinction. The lower threshold of gestural expressions may be quite hard to establish definitively, and this means certain ‘body information’, like one’s stance or mien, may pass from one to the other side of this distinction, based on circumstances or other factors. Importantly, this fluidity does not seem to impinge upon how, as stressed by Kendon (2004, 11–13), humans seem adept at judging just which body information is relevant to the interaction as ‘deliberate, expressive action’, that is, as gestures.3 Second, in this example the persons involved are deploying different sorts of communicative ‘resources’, each with their own distinct features, yet which are nonetheless ‘interdependent phenomena’ (Streeck et  al. 2011, 3). The first person’s initial verbal phrase is a kind of directive whose object and content can only be understood via her gesture, and the interlocutor can immediately grasp this. Likewise, the second person’s

14  Wundt and Bühler on Gestural Expression… 

281

own gesture in response will provide the referent for what she might go on to express verbally; indeed, one can easily imagine a case where whatever else she will say verbally will have already been expressed in her trowel gesture. The nature of the semantic and performative relationships between these distinct phenomena seems to run in both directions, from both gesture to speech and speech to gesture. From the moment there is recognition of distinctive communicative resources enmeshed in human communicative interaction, there arise questions about the conditions, rules, or norms—broadly speaking, the ‘grammar’—under which this enmeshing occurs, in the co- or alternating performance of gesture and speech. For instance, the more someone would be committed to a kind of ‘semantic holism’, the more they would be required to come to a position on the underlying nature of the fundamental (inter)relation of gesture and speech which multimodality would denote. This is a fascinating but challenging question for contemporary philosophy and linguistics.4 In the latter, for one, one finds numerous careful attempts to identify and catalogue different sorts of multimodality in human communication, in different contexts and cultural situations, but a basic concern remains as to how best to arrive at hypotheses about why such phenomena of multimodality should take certain forms rather than others. Should such rules or conditions be understood primarily in terms of the regulating limitations and competencies of the human organism in its psychic and corporeal constitution? Are they strictly dictated by the material conditions of the human interaction,5 or by the historical customs and cultural conventions of a person’s upbringing and surroundings? Notably, proposals of this last sort would fall under a relativistic or even psychologistic position on the notion of the ‘grammar’ of multimodal communication. By contrast, are such rules or conditions dictated by some sort of intrinsic, adequative link between whatever communicative motivation is going on ‘inside’ an individual and the external expressive (gestural and/ or speech) utterance, as has been shown to be the case for Wittgenstein and Scheler in the context of emotional expressions (Mulligan 2012, 63)? Might the rules for combining gestural and verbal expression be thought of  in terms of dependent, co-dependent, or independent parts and wholes, or in terms of how a certain Meinen or communicative intention

282 

B. Vassilicos

with a determinate content gets parsed across these different semiotic resources, while still meeting sufficiency conditions for expression in each mode and within the whole? May there be certain structural conditions imposed upon such multimodal communication (across gesture and verbal expression) whose necessity is independent of any specifically human communicative function or need? What might they be and how might one identify them? In the ensuing, we aim to broach these questions by exploring how Wundt’s and Bühler’s respective theories of gestural expression would treat them. This will require some careful positioning of each figure with respect to this line of questioning. It does not go without saying that what Bühler investigates as gestures entirely correlates with how gestures are studied today (Friedrich 2012, 210, note 14), and the same might be said for Wundt as well. It will thus have to be shown how there are sufficient conceptual grounds for connecting these past thinkers with today’s approaches, and to do so we shall proceed in the following manner. With Wundt, the goal is to see how he advocates a problematic position on the relationship of gesture and language, which stems from his underlying conception of expression. Such a demonstration will allow us to recognise and develop a critique of a neo-Wundtian position that has surfaced in contemporary work on gestures. With Bühler, on the other hand, the goal will be to outline a positive contribution to these questions, which is, on the one hand, structurally focused and phenomenologically grounded but which on the other remains empirically informed and thus commensurable with recent accounts of gesture.

2

Wundt: Expression and Grammar

The verdict on Wundt’s theories of language by Danziger, already in 1983, seems categorical and unstinting; ‘[they are] of limited interest today’ (1983, 308). For Danziger, with his emphasis on Wundt’s conception of Völkerpsychologie, perhaps the only source of curiosity is that of the historian in what might have been. But can Wundt say nothing to us today?

14  Wundt and Bühler on Gestural Expression… 

283

There is indeed still much to appreciate in Wundt, at least in light of his analysis of gestural expression: the considerable granularity of his catalogue of different kinds of gestures; his proto-pragmatic recognition that facial expressions provide the basic tone (Grundton) for interpreting gestures (1977, 86); the innovative manner in which he broaches the issue of the grammar or syntax of gestures. These reasons aside, one further motivation still to take an interest in Wundt is that his intellectual legacy seems very much alive and well today, albeit under another, not very explicitly Wundtian form. What is meant here is the widely influential ‘Growth Point’ (GP) theory of the psychologist David McNeill and his colleagues, which has become paradigmatic for a number of studies on gesture today (De Ruiter and De Beer 2013, 1015; Kendon 2004, 77–78).6 For McNeill, the notion of ‘growth point’ is meant to capture how a person’s thoughts can come to different means of expression, through either speech or gesture or both; the growth point is ‘[t]he initial organising impulse of an utterance, and the starting point for developing its meaning; [it is] a minimal, irreducible, psychological unit that […] is a microcosm of the whole utterance’ (Duncan et al. 2007, 5). Such a notion already indicates a kind of affinity with the Wundtian view of communication (Mitteilung),7 but even more decisive for establishing the link to Wundt is how gestures and co-speech gestures are understood by McNeill. Of the gestures he deems most interesting and relevant for investigation, he writes that they are ‘idiosyncratic spontaneous movements of the hands and arms accompanying speech [which] can be taken to reveal the utterance’s primitive stage’ (McNeill 1992, 40); with these gestures, ‘people unwittingly display their inner thoughts and ways of understanding events of the world […]. Gestures are like thoughts themselves […] [and] belong, not to the outside world, but to the inside one of memory, thought, and mental images’ (McNeill 1992, 12).8 Statements like these seem to have a Wundtian conception of gestural expression at their core. To understand why that is and the questions they invite, we need to understand the role of the principle of psycho-physical parallelism (PPP) in Wundt’s own theory of language and expression. Wundt’s PPP may be understood as a neo-Leibnizian thesis about the ‘contents of experience’ (Erfahrungsinhalt) which makes two related claims. First, it holds that these contents are available to two sorts of

284 

B. Vassilicos

scientific investigations, one ‘indirect’, via brain- or neuroscience, and the other ‘direct’, via psychology. Second, it postulates that every process on the one side of that distinction has to have a corresponding process on the other side (Wundt 1896–1897, 389). As cited by Rieber, Brentano apparently claimed that Wundt ‘forfeits the unity of mental life’ by saddling himself with the PPP; in so doing, ‘he gives himself three problems instead of one’ (2001, 152). These are, ostensibly, the relation of the neuro-physical processes to experience, the relation of psychological processes to experience, and the relation of the unity of the neuro-physical and the psychological to experience. In Wundt’s defence, one could say he sought to define the PPP only as a sort of ‘general heuristic principle’ for phenomena that seemed quite disparate (Fahrenberg 2012, 230). These are, as Ungeheuer explains, phenomena in the bio-physical system of the human being that do not seem to obey the law of the conservation of energy (in terms of inputs and outputs) (1984, 16). As such, Wundt could be seen to be advocating a kind of methodological dualism with which to explore ‘a unitary, monistic conception of life’, for instance, as evinced in the two very different approaches to gestural expression taken in the first two chapters of the Sprache volume of his Völkerpsychologie. The interesting problems arise from the way that Wundt comes to rely on the PPP when he turns his attention to human social phenomena and especially what he calls the ‘drive to communicate’. As commentators like Bühler (1933) and Nerlich and Clarke (1998) have pointed out, this parallelism—initially serving as a methodological safeguard against confusing two sets of facts or two frames of references—morphs into something more, namely a descriptive framework by which Wundt analyses human communication in general and gestural expression in particular. As Bühler stresses in his critique of Wundt in the former’s Ausdruckstheorie, the PPP goes from being a more cautious claim about the intimate relationship between the neuro-physical and the experiential domains to a much more ambitious or ‘generous’ (Bühler’s term) thesis. Namely, the PPP for Wundt also comes to imply an expressive thesis concerning a ‘mirroring’ between the inner psychic state and the ‘outer’ gestural expression (Bühler 1933, 133) and a pro-social thesis about the mirroring between the psychic states of the persons involved in communicative interaction.

14  Wundt and Bühler on Gestural Expression… 

285

To understand the thrust of Bühler’s critique of such an ambitious thesis and its relevance today, it will be useful to clarify how Wundt understands gestural expression according to the PPP. On the whole, for Wundt, the expressible contents of the psychic life of human beings are constituted by affects, as counterparts to the physical excitations of the body.9 These affects, which are themselves complexes of sensations that alone would not be expressed (Wundt 1904, 45, 47), give rise to a binary drive that underpins the basic forms of human expressive activities. On the one hand, the drive pushes for comprehension (Verständigung) of the affect, namely by associating it with a representation (Vorstellung) that can then be linked to other representations and their respective affects. On the other, the drive pushes for communication (Mitteilung) of the affect via bodily movements. In this respect, Wundt can then claim that ‘the gesture is the direct expression of the concept which at the moment governs the affect’ (Wundt 1977, 147).10 There can be no doubt that Wundt sees such an analysis of gestural expression as wholly consistent with the PPP (Wundt 1904, 90). However, processes running in parallel with each other does not necessarily mean that they mirror each other, yet this is just the direction in which Wundt takes his interpretation of the PPP.  This can be seen in how the PPP underwrites Wundt's view of the relation between speaking (a physiological activity) and understanding (a mental activity). He regards speaking and understanding as mirror image processes, a view that allowed him to dispense with any particular psychological theory of how understanding, especially understanding of speech in a social context, is possible (Knobloch 1988, 415). Moreover, further evidence of the link between the PPP and a notion of mirroring can be found in the manner in which Wundt delineates only certain gestures to be relevant to his study of the psychology of human communication. In a manner that shows a striking relationship to McNeill’s perspective mentioned earlier, Wundt claims that those gestures that come from ‘arbitrary agreement’, i.e. via conventions, are of little importance for the psychologist of language (Wundt 1977, 70). Much more relevant are those gestures that constitute an expressive ‘means derived from the direct drive to communicate [which are] on the whole unpremeditated’ and thus that demonstrate a ‘greater primitiveness’ (Wundt 1977, 75). This nomination by Wundt of only

286 

B. Vassilicos

certain gestures to be investigated leaves questions about the use of arbitrary or conventional gestures something of a mystery to him; they may stem from an affect or they may not, but Wundt’s psychologist of language will never be able to get to the heart of the matter one way or another. The primitive, psychologically interesting gestural expressions, on the other hand, manifest a direct inner connection to the affect that generates the drive to communicate, and indeed do so in a regular or constant fashion—what he calls in a passage that Bühler emphasises, ‘inseparable characteristic expressive movements’ (Wundt 1904, 100). With such an assertion, Wundt is perhaps not too far off from Ekman and Friesen’s thesis that human emotions have certain universally shared facial expressions across all peoples of the world (1975). This idea would demonstrate one meaning of the notion of mirroring to which Bühler’s critique alludes; a faithful and constant mirroring between affects in the body and the movements to which they give rise, where affect and movement reflect each other to such an extent that the occurrence of the one even seems to motivate the occurrence of the other.11 However, there is still another sense of mirroring that comes into play in Wundt’s account, which can be seen as a further implication of the previous one; there is also mirroring between individuals, where expressive movements in a communicator can give rise to (co-)expressive movements in the addressee—what he calls Mitbewegungen—which in turn link back in the addressee’s mind to affects and respective concepts. As he writes, ‘[…] the [gestural] expression provides a firm basis for the mirroring [Widerspiegelung] of the affective movement [Gemütsbewegung] in the addressee’ and ‘excites further concepts associated with the gesture, develops the gesture, or perhaps even elicits its own opposite’ (1977, 147). One might find something chilling about the social world hereby sketched by Wundt—one to some extent also present in McNeill (2012, 65, 154–156)—where all human beings are linked in some great chain of mirror neurons. More seriously, such a deployment and transposition of the PPP into a principle of pro-social mirroring between individuals seems to expose a serious flaw in Wundt’s analysis, for which Bühler will take him to task; Wundt seems to miss entirely the interactive, person-to-­ person interdependent character of gestural expression.

14  Wundt and Bühler on Gestural Expression… 

287

This can be seen in how Wundt approaches our earlier question about the rules or conditions according to which gestures are performed (with or without speech). In analysing the performance of a series of gestures that form a certain sentence (i.e. comparing how gestures flow and are linked together with respect to how words in speech flow and are linked together), Wundt considers the following question: why might the order of the subject and predicate be reversed, as seems more often to be the case in gestural communication?12 His answer is both striking and entirely consonant with his approach. He argues that there are two reasons for such a reversal. On the one hand, gestures have a certain ‘slowness’ of performance with respect to speech, which is due to the time it takes for the hands to make sequential movements. On the other, certain affects have a higher degree of excitation and thus a higher pressure to be expressed (Wundt 1977, 123). To accommodate these two factors, the communicating individual finds it necessary to reverse subject and predicate in cases of ‘animated, excited speech’ (Wundt 1977, 124). To use a metaphor, it is as if the gesturing communicator were like an overworked postal worker having to deliver parcels in function of their urgency or encumbrance for the sender, instead of according to address, order of intake, etc. The implication in such a view of communication is that it is the human capacities for processing mental content plus the physical limitations of the human body that chiefly condition gesture performance. There can be no question for Wundt of the gestural expression being structurally determined, for instance, in function of the conditions under which another person would understand such expression or with respect to an aim to evaluate the response by that other person to one’s own gesture. This means that, on Wundt’s view, if we aim to understand under what conditions gestural expression can be performed, then we ought foremost to focus our scientific attention on the conditions under which an affect can be manifested—for instance, whether the affect is strong (‘excited’) or weak, confusing or clear, new or old, etc. For the theorist of language of a Martyian persuasion, for whom manifestation (Kundgabe) and intimation (Kundnahme) form two interrelated, indispensable, yet distinct functions of human communication (Cesalli 2009, 47, 51), such an emphasis on the conditions of the manifestation

288 

B. Vassilicos

of the affect will seem like a rather problematic view for understanding the conditions under which gestural and verbal expression are related. This view nonetheless seems influential in some current work on gesture. In particular, it surfaces McNeill’s ontogenetic view, as a further development of his GP theory, that the origin of language lies in the happy circumstance that our brains are organised in a certain way (2012, 154). It seems moreover to underlie the motivation to rely upon aphasia studies in order to formulate and evaluate models of gestural expression. In a study by De Ruiter and De Beer (2013), extensive consideration is given to how current competing theories may generate and accommodate the ‘modules’ or psychic capacities that seem to be affected in non-fluent aphasia; their corresponding aim is to formulate a general theory of the psychic capacities that condition or constrain gestural and linguistic expression for aphasic and non-aphasic speakers alike. While such work has its place, the question remains whether these lines of inquiry are the only or most important ones when it comes to the conditions upon multimodal expression. Put in terms of Bühler’s critique, such perspectives risk approaching gestural and verbal expression, and their multimodal relationship, from the theoretical perspective of a ‘Diogenes im Fass’, which is to say, from the perspective of an organism closed in on itself (Bühler 1977, 37, 47); that is, from the perspective of an organism for whom the constraints upon the verbal or gestural expression of an affect are tantamount to the constraints upon the manifestation of that affect within the organism.

3

Bühler: Diacrisis, Steering, Structure

To explore the potential for Bühler’s contribution to understanding the conditions under which gestural and verbal expression are integrated and related, let us consider once more the above example of two persons communicating at a dig site. We can think of the gestures and the verbal speech of a communicating person as two expressly shared streams of information, addressed to another person, and as being about the same state of affairs, with a common communicative purpose driving both streams. When conceived in this way, the two streams show different

14  Wundt and Bühler on Gestural Expression… 

289

kinds of relationships with each other; some seem merely possible ways of being related, while others seem necessary. On the side of the possible, the one stream may but need not hold a supplementary or repetitive relationship with respect to the other. The supplementary relationship means each stream can offer information not contained in the other, for instance, by way of completion or amplification, as in the above example where a gesture might provide a referent for the verbal utterance of a relative pronoun. A repetitive relationship means each stream can reiterate information given in the other, so as to enable the one to accent the other, or to enable a referring back in the one stream to information given in the other (anaphora); in the example above, the one archaeologist may trace a line with their gesture and a moment later also verbally describe the path of a line in the dirt. On the other hand, considered from the perspective of their mutual integration in acts of communication, there are other relationships between the two streams of gestures and verbal expression that seem required. One such would be their mutual non-interference with each other; each stream has to be able to offer information without hindering the information given in the other stream. This might be thought of as a no-noise principle of multimodal communication; uptake of the one, e.g. gestural, stream ought not prevent uptake of the other, e.g. verbal expression, stream, nor vice versa. Another such candidate, related to the latter, would be the relationship of coordination or fit between the two streams. Such coordination is necessary if interaction and integration between the two streams is to be achieved; even when the one stream might undermine or contradict the other (e.g. in irony, sarcasm, or deception), still the two streams must be ‘in sync’ with each other. Such coordination, though required, can nonetheless take different forms. It might be syntactic, as in the one person’s gesture constituting the ‘noun’ of the verbally expressed prepositional phrase, or it may be temporal, as in the emphasis of the so-called ‘beated’ or anticipatory gesture which must fit the temporal frame of the verbal expression to which it would be matched. These are just a few ways that gestures and speech may interact and be integrated, according to either their possible or necessary relationships. Yet how might one best elaborate further, refine, or correct such a characterisation of the integration of these two streams in communication? On

290 

B. Vassilicos

our understanding, there are at least two Bühlerian concerns to be explored regarding this framing of the relationships between gesture and speech. A first concern would point to a crucial facet thus far missing from the depiction of the two streams. Namely, to which kind of meaningful form or structure of communication are these two streams to be understood as contributing? That is, Bühler would insist that there can never be two such streams in a communicative vacuum; he would thus call for consideration of the communicative structures (Sprachgebilde) and communicative products (Sprechwerke) in which these two streams and their interaction would always have to fit (Bühler 2011, 57–58). A second concern has to do with the communicative functions fulfilled by the streams above (Bühler 2011, 35). It is easy to grasp that the one or other stream of information can be involved in representing something the communicator is thinking about (Darstellung), or in manifesting something the communicator is feeling (Ausdruck), or in drawing the listener’s attention to something in the environment (Appell), or indeed in some combination of these. However, when these streams of information would be directed at a receiver in an act of communication, by what means would the receiver be guided to the relevance of the one stream for the other, as well as their mutual relevance for the ‘whole’ meaning of the communicator's streams? This is a question not only when the streams would diverge in their functions, but just as much when those functions would align; it seems necessary that the receiver grasp their importance for each other in order to make sense of the streams, just as it seems necessary for the communicator to intend for their relationship to be taken in a certain way. For instance, by what means should a receiver distinguish what is merely extrinsic or ‘supplementary’ about the one stream for the other stream, from what might be essential to their relationship? What is needed here, Bühler would suggest, is a principle of steering by which the relationship between the streams and their functions could be understood; in other words, what is needed in order to understand the relationship of gestures and verbal speech is a concept of diacrisis. For the moment, let us focus on the second of these two concerns. Diacrisis (or sometimes diacritics) is a concept that Bühler draws from his investigations of phonology and the processes of perception thereby involved (2011, 42).13 On the one hand, it refers to a process of

14  Wundt and Bühler on Gestural Expression… 

291

abstraction, whereby a person, an ear, or a finger can learn to distinguish the relevant from the irrelevant (Bühler 2011, 53), thus turning some of the information provided into a kind of background, context, or noise, and the remainder into the content of a perceptual apprehension. On the other hand, diacrisis for Bühler would equally indicate the result of such a process of abstraction regarding some quantity or quantities of information—aural, tactile, or symbolic; namely, this result is an (re-)orientation or a guiding of an individual in any given situation where a ‘decision between several possibilities has to be made’ (Bühler 2011, 176). Per Bühler’s examples, this diacrisis means that via perceptual content one can be guided to a recognition that, e.g. there is a one-euro and not a two-euro coin in one's pocket. In respect of a Bühlerian conception of multimodal expression via gesture and speech, two further aspects of the notion of diacrisis are of interest. First, although such diacrisis may be arrived at by an individual under their own power, it may also be instigated or initiated in communicative interaction. When a person returns home, for instance, and says ‘I’m home’ or ‘It’s me’, the person upstairs hearing that utterance ‘is supposed to perform a personal diacrisis, more or less as if a personal name had been spoken’ (Bühler 2011, 109); the speaker wants the listener to attend in the utterance only to those ‘distinguishing marks’ of the speaker’s sound of voice that allow the listener to identify just who is the speaker. This kind of utterance is a kind of vocal pointing or steering of the listener to an aural identification; it should enable the listener to decide just who is speaking.14 At the same time, Bühler holds that diacrisis can occur not just in speech, but also in gestures, or in both at once. Either gesture or utterance can count as ‘islands of language emerg[ing] from within the sea of silent but unequivocal communication’ (2011, 176) as we guide and are guided by others and as we coordinate our activities with others.15 Second, if diacrisis concerns a kind of steering towards a decision between certain possibilities of meaning, an important caveat here is that these are not mere empty possibilities but are rather ones constrained by two related but still distinct factors. First, the possibilities navigated by the diacrisis are already ‘steered by the matter at issue’ (Sachsteuerung) (Bühler 2011, 75–76), where such a matter may be understood as a content or state of affairs in respect of which certain understandings,

292 

B. Vassilicos

responses, or actions—that is, certain acts and courses of actions—are most likely. This aspect of the diacrisis reflects Bühler's structural leanings, according to which he would investigate the subject-independent and law-like conditions under which humans communicate and act. Second, Bühler stresses that either the material world or other factors, like the dispositions of other persons, may restrict possibilities in light of which the diacrisis is effected. He thus refers to the ‘material clues’ and ‘material guidance’ (Stoffliche Steuerung) of the likely choices that one might make in a café, in getting on a bus, in hearing a voice in one’s home, and the like (Bühler 2011, 77, 194 ff.). Yet here one ought also to recall that for Bühler diacrisis takes place at many different levels of the communication process (Bühler 2011, 52); there is already diacrisis in the sound and the phoneme, between one gesture or word and another, between a representational function and an appellative function of a gesture or word, and so on. Combining these thoughts, we arrive at the idea that every diacrisis, as a presentation and a choice of a meaningful possibility, occurs in a respect of a diacrisis at another level of complexity, higher or lower, in the communicative process. What this Bühlerian view thereby implies is that gestures and speech are answerable not only to their environments and the trajectories (Vorbahnungen) of the matter at issue jointly attended to by the speaker and the receiver (Knobloch 1988, 419). In addition, they may be crucially answerable to each other, in their relationship to the matter at hand and in providing a context for the performance of each. Thus, gestures may be instrumental in operating alongside speech in both aiding the distinction of one sound from another, one word from another, one concept from another, one kind of utterance from another, one kind of speaker’s intent from another. And the same would obtain for speech in relation to gestures as well. How might this notion of the diacrisis expand or revise the above characterisation of multimodal expression via speech and gesture according to their possible and necessary relationships? One contribution concerns the question whether the relationships of non-interference and coordination between gesture and speech constitute a complete picture of the necessary conditions of their integration and interaction. One hesitation about that picture could run as follows; even if it seems correct to insist that  the

14  Wundt and Bühler on Gestural Expression… 

293

one stream (e.g. of gestures) ought not disrupt the comprehension of the other, this condition could also be met by someone merely ignoring the one or other stream. Our description would then miss the very phenomenon it means to target, namely the interaction of gesture and speech. Hence, in insisting upon a necessary relationship of non-interference, the implication cannot be that one stream can simply replace the other—at least, not if one takes these co-speech gestures not to be sheerly ‘self-­ directed’ or cognitive-facultative bodily movements with no meaning for the other person and no intention to be a matter of uptake. As an improvement upon that picture, one may suggest that to the non-interference relationship has to be appended a non-negligible or diacritical relationship. According to this latter requirement, each stream has to occupy some position of relevance or importance with respect to the other stream with which it would be integrated (greater, lesser, equal). It would then be in view of this mutual relationship of relevance that the receiver of the integrated streams, gesture and speech, would then attend now primarily to the one or the other, or indeed both at once, in order to grasp the function of each stream, the nature of their roles with the communicative act, the course of action to which they may lead, and so on. To put this differently, the diacritical relationship of the two streams would ensure that their receiver would be steered to the relevance of both streams for the matter at hand and the communicative situation. One might indeed wonder how that happens, yet on our view such a relationship would not necessarily entail that each stream must carry some ‘meta-­ information’ about its relevance to the other stream. In some cases, as in the above example of an elliptical statement completed by a trowel gesture, this diacritical requirement seems fulfilled quite straightforwardly, for instance via the coordination (in this case alternation) of gesture and speech. In others, especially where the two streams are co-occurring and not consecutive, it may be more challenging to understand how this requirement is met, and here we can refer to an interesting distinction already made by Wundt between Hilfegebärden and Hauptgebärden; the former being gestures aimed at helping the comprehension of gestures, or in our example, aimed at directing attention to the mutual importance of the one or other stream, in a manner similar to deictic references in speech. Moreover, if we expand our multimodal scope for a moment, one

294 

B. Vassilicos

might consider whether such steering occurs strictly via gestures or speech or whether it may also even occur via a yet third stream of information, such as prosody, the exchange of eye gazes, or bodily stances. In the foregoing, we have attempted to demonstrate both why it can still be important to understand Wundt and the criticisms of him, and why Bühler can offer an account of gestural and multimodal expression which is at once amenable to empirical input and structural in its aims. However, this has been but an indication of the potential of Bühler's insights. For further development thereof, one area would be the question of how the necessary and possible relationships of gesture and speech fall within Bühler’s fourfold scheme of speech as “Act, Work, Action, and Structure” (2011, 57). We might thereby explore what other relationships between gesture and speech we have missed, and thus take Bühler’s scheme as a guide for systematically exploring and interpreting the wide taxonomy of gesture actions and gesture texts recorded by anthropologists, linguists, and psychologists in today’s and yesterday’s human world. Where are they populous within Bühler's scheme? Where might each need to be filled out further, and why?

Notes 1. See (Kendon 2007, 24; Kendon 2004, 73, 75; Streeck 1993, 276). Bühler, for one, already appealed for a scientific openness to the variety of functions of gesture in warning against such a reductivism: “No, gesture is gesture and language is language, and the use of mimicking expressions and gestures in human interaction would be in a bad way if everything had to be underpinned and adequately translatable (interpretable) into spoken language” (2011, 178 [trans. changed]). There are some questions then about any tendency to interpret Bühler’s earlier work on gestural expression (1933) as claiming that “there is no principal difference between vocal and bodily (e.g. sign) language” (Müller 2013, 203–4). In fact, this overlooks how in that work Bühler’s point was more nuanced. He disagrees with Engels and claims that gestures, like speech, also have a capacity for sophistication and differentiation in expression (1933, 40). This does not entail that gestural expression would attain such fea-

14  Wundt and Bühler on Gestural Expression… 

295

tures in just the same way as speech does or that there would no differences between the two forms of expression; earlier in the same passage (39), Bühler points to the particular capacity of speech to combine linguistic functions in a way freer and smoother than how they may be combined in gestures. 2. This example is drawn from (Streeck et al. 2011). 3. See (Müller 2014) for a further discussion of this aptitude. 4. See (Müller et al. 2013) and (Fricke 2013) for two recent approaches to such questions from the perspective of linguistics. (Müller et al. 2013) embraces McNeill’s notion of gestural expression, which leaves open opportunity for a more Bühlerian contribution; Fricke takes ‘grammar’ in a narrower sense than considered here. 5. See Kendon’s ‘context of use’ thesis (2004). 6. Streeck (1993) already hints at this link. See (McNeill 2012, 130–31) for an explicit parallel between the GP and Wundt’s notion of the ‘sentence’. 7. As in Wundt’s assertion that ‘[l]anguage is the mirror of the human being in the totality of its psychic achievements’ (1977, 148–149). 8. See also (McNeill and Duncan 2000, 143), which describes ‘co-expressive speech-synchronised gestures’ as ‘a “window’ onto thinking”’. 9. (Wundt 1904, 50). 10. See (Araujo 2016, 52), for a more complex picture of the notion of ‘affect’ in the earlier Wundt. 11. See (Wundt 1904, 134–135). 12. See how Wundt analyses a series of gestures which he renders as meaning ‘the child was hit, the man was angry’. 13. See (Mulligan 1988, 209). 14. In this respect, we might even go so far as to say such diacrisis goes to the heart of what Bühler calls in his Krisis the ‘mutual steering’ (Gegenseitige Steuerung) and the ‘interpersonal process of appeal’ (Auslösung). 15. See also (Bühler 2011, 160). In this regard, diacrisis has an important pragmatic function, in the linguistic sense of the term.

References Araujo, Saulo de Freitas. 2016. Wundt and the Philosophical Foundations of Psychology. Cham: Springer. Bühler, Karl. 1933. Ausdruckstheorie. Jena: Fischer.

296 

B. Vassilicos

———. 1977. The Psychophysics of Expression of Wilhelm Wundt. In The Language of Gestures, ed. Georg Herbert Mead et  al., 30–54. Mouton: The Hague. ———. 2011. Theory of Language. Trans. Donald Fraser Goodwin. Amsterdam: John Benjamins. Cesalli, Laurent. 2009. Zweck vs. Leistung. Les Deux Fonctionnalismes de Marty et Bühler. Verbum 21 (1–2): 45–64. Danziger, Kurt. 1983. Origins and Basic Principles of Wundt’s Völkerpsychologie. British Journal of Social Psychology 22: 303–313. De Ruiter, Jan P., and Caroline De Beer. 2013. A Critical Evaluation of Models of Gesture and Speech Production for Understanding Gesture in Aphasia. Aphasiology 27 (9): 1015–1030. Duncan, Susan D., Justine Cassell, and Elena T.  Levy. 2007. The Dynamic Aspect of Language. In Gesture and the Dynamic Aspect of Language, ed. Susan D.  Duncan, Justine Cassell, and Elena T.  Levy, 3–13. Amsterdam: John Benjamins. Ekman, Paul, and Wallace V.  Friesen. 1975. Unmasking the Face: A Guide to Recognizing Emotions from Facial Expressions. Englewood Cliffs, NJ: Prentice-Hall. Fahrenberg, Jochen. 2012. Wilhelm Wundts Wissenschaftstheorie. Ein Rekonstruktionsversuch. Psychologische Rundschau 63 (4): 228–238. Fricke, Ellen. 2013. Towards a Unified Grammar of Gesture and Speech: A Multimodal Approach. In Body—Language—Communication: An International Handbook on Multimodality in Human Communication, ed. Cornelia Müller et al., vol. 1, 733–755. Berlin: De Gruyter. Friedrich, Janette. 2012. Le concept d’expression chez Karl Bühler. Intellectica 57: 199–217. Kendon, Adam. 2004. Gesture: Visible Action as Utterance. Cambridge: Cambridge University Press. ———. 2007. On the Origins of Modern Gesture Studies. In Gesture and the Dynamic Aspect of Language, ed. Susan D. Duncan, Justine Cassell, and Elena T. Levy, 13–28. Amsterdam: John Benjamins. Knobloch, Clemens. 1988. Die Bedeutung von Bühlers ‘Axiomatik’ für die Psycholinguistik. In Karl Buhler’s Theory of Language, ed. Achim Eschbach, 415–433. Amsterdam: John Benjamins. McNeill, David. 1992. Hand and Mind. Chicago: University of Chicago Press. ———. 2012. How Language Began: Gesture and Speech in Human Evolution. Cambridge: Cambridge University Press.

14  Wundt and Bühler on Gestural Expression… 

297

McNeill, David, and Susan F. Duncan. 2000. Growth Points in Thinking-for-­ Speaking. In Language and Gesture, ed. David McNeill, 141–161. Cambridge: Cambridge University Press. Müller, Cornelia. 2013. Gestures as a Medium of Expression: The Linguistic Potential of Gestures. In Body—Language—Communication: An International Handbook on Multimodality in Human Communication, ed. Cornelia Müller et al., vol. 1, 202–217. Berlin: De Gruyter. ———. 2014. Gesture as ‘Deliberate Expressive Movement’. In From Gesture in Conversation to Visible Action as Utterance, ed. Mandana Seyfeddinipur and Marianne Gullberg, 127–152. Amsterdam: John Benjamins. Müller, Cornelia, Jana Bressem, and Silva H.  Ladewig. 2013. Towards a Grammar of Gestures: A Form-Based View. In Body—Language— Communication: An International Handbook on Multimodality in Human Communication, ed. Cornelia Müller et  al., vol. 1, 707–733. Berlin: De Gruyter. Mulligan, Kevin. 1988. On Structure. Bühler’s Linguistic and Psychological Examples. In Karl Buhler’s Theory of Language, ed. Achim Eschbach, 203–226. Amsterdam: John Benjamins. ———. 2012. Wittgenstein et la philosophie austro-allemande. Paris: Vrin. Nerlich, Brigitte, and David D. Clarke. 1998. The Linguistic Repudiation of Wundt. History of Psychology 1 (3): 179–204. Rieber, Robert W. 2001. Wundt and the Americans. In Wilhelm Wundt in History: The Making of a Scientific Psychology, ed. Robert W. Rieber and David Robinson, 145–160. New York: Springer. Streeck, Jürgen. 1993. Gesture as Communication I: Its Coordination with Gaze and Speech. Communication Monographs 60: 275–299. Streeck, Jürgen, Charles Goodwin, and Curtis LeBaron. 2011. Embodied Action in the Material World. In Embodied Interaction. Language and Body in the Material World, ed. Jürgen Streeck, Charles Goodwin, and Curtis LeBaron, 1–29. Cambridge: Cambridge University Press. Ungeheuer, Gerold. 1984. Bühler und Wundt. Bühler-Studien 2: 9–67. Wundt, Wilhelm. 1904. Völkerpsychologie, vol. 1 (Die Sprache). 2nd ed. Leipzig: Engelmann. ———. 1896–1897. Grundriss der Psychologie. Leipzig: Engelmann. Translated by Charles Hubbard Judd as Outlines of Psychology. Leipzig: Engelmann. 1977. ———. 1977. The Language of Gestures. In The Language of Gestures, ed. Georg Herbert Mead et al., 55–152. The Hague: Mouton.

15 On Being Guided, Signals and Rules: From Bühler to Wittgenstein Kevin Mulligan

1

A Guide to Guidance

Appearances, demonstratives, facts, knowledge, name tags, people, perceptions, pictures (misleading or not), quotation marks, rules, street signs, traffic signals, utterances, written signs, and many, many other things guide, lead, steer, control and influence us. In the writings of Brentano’s heirs and Wittgenstein, we find many verbs for such relations: trigger/auslösen follow/befolgen steer, control/steuern influence/beinflussen guide, lead/führen allow oneself to be led by/sich leiten lassen indicate/anzeigen

K. Mulligan (*) Faculty of Communication, Culture and Society, Università della Svizzera italiana, Lugano, Switzerland e-mail: [email protected] © The Author(s) 2021 A. Dewalque et al. (eds.), Philosophy of Language in the Brentano School, History of Analytic Philosophy, https://doi.org/10.1007/978-3-030-52211-7_15

299

300 

K. Mulligan

The philosophers of guidance are Martinak, Marty and Ahlman, Husserl and Scheler, Bühler and Wittgenstein. Wittgenstein and his great predecessor, Bühler, sometimes use the term ‘signal’ (Signal) for the first term of the relation of guidance.1 ‘Understanding’, says Wittgenstein, ‘is like knowing how to go on, and so it is an ability: but ‘I understand’, like ‘I can go on’, is an utterance, a signal (Signal) (Wittgenstein 1980, I, § 875). Bühler systematically opposes signals to indicators (Anzeichen), sometimes called ‘utterances’ (Äusserung) and ‘expressions’ (Ausdruck), and also to symbols (Symbol). Signals guide or steer (leiten, steuern), symbols represent (darstellen) objects and states of affairs, and indicators indicate (anzeigen) or express mental or psychological states, functions and acts. One of the more striking illustrations of guidance or steering in Wittgenstein’s writings is to be found in the series of simple interactions described at the beginning of Philosophical Investigations, for example that of the builder who calls out a word and the assistant who then brings a stone. This is, among other things, an example of a use of language to ‘influence other people’ (Wittgenstein 1980, § 491). But, as Bühler had pointed out in 1927, the real variety of guidance only emerges if we consider both non-verbal and verbal interactions and their relations: The apprentice who hands something to his master in the course of carrying out some common task, the surgical assistant and the surgeon—here the fact that the activities go hand in hand is brought about by a steering directed towards a goal in a shared perceptual field. […] But if the focus of a desired steering transcends in any way the shared perceptual domain then […] a higher-order contact is needed to make steering possible. If we take the case of contact of the highest possible kind, the speaker appeals to the ideas [Vorstellungen] and concepts of her listener. But before this there lies a range of intermediate steps and a labyrinth of ways of transcending [the shared perceptual domain] and of possibilities of overcoming this range. (Bühler 1929 [1927], 40–41, see 90)

What exactly is it to be guided by something? What are the main forms the phenomenon takes? What are the phenomena which presuppose, which are founded on guidance?

15  On Being Guided, Signals and Rules: From Bühler… 

301

The first question raises many other questions. Does guidance involve any sort of experience or is the notion a non-experiential one? Is it a passive or active phenomenon? Is it sometimes, if we allow ourselves some Husserlian jargon, a matter of passive activity and sometimes a matter of active passivity, and sometimes even a matter of passive passivity? Is the relation a causal relation? An example of association? Of motivation? Are the items which guide us reasons? What is the relation between guidance in the animal world and in the human world? Are expressions such as ‘guide’ ‘lead’ and ‘steer’ family resemblance terms? Is the concept of guidance definable? One answer to the second question is that what we say and do influences the doxastic, epistemic, affective and conative states of others as well as their behaviour. But we are guided, too, by inanimate items— rules, norms, knowledge and by forms, the so called subjective apriori. Is this great variety exhibited by guidance a reason for thinking that it has no essence or nature? What sort of things presuppose, are founded on the phenomenon of guidance? Many if not all forms of social interaction. Consider for example the social category of a community (Gemeinschaft). Bühler suggests that there can be no community without reciprocal guidance: In human life there are many different accidental gatherings of hermits; we exclude them from the concept of genuine community life. […] It is enough for our purposes if we emphasise one feature, the fact that the meaningful behaviour of the members of a community is subject to reciprocal steering. (Bühler 1929 [1927], 39)2

Indeed psychological contact, knowledge or understanding of the minds or souls of others involves reciprocal steering (Bühler 1929 [1927], 83–5, 99–101). Reciprocal steering of one individual by another is by no means the only type of social cement. Rule-following, in particular our blindly following rules, syntactic, semantic, pragmatic, moral, aesthetic, prudential, culinary, as well as the rules of etiquette and for driving on the left, is also important. We are guided by or, as Husserl puts it, allow ourselves to be guided by rules. Is following the same rules prior to belonging to the

302 

K. Mulligan

same community? Does the priority run in the other direction? Or is there perhaps no such relation of priority? In 1926, Ahlman endorses the first answer: We determine whether two individuals belong to the same linguistic community by finding out whether they observe the same norms and not vice versa—the norm system has priority over the linguistic community. (Ahlman 1926, 15)

A third type of social cement resembles the first in linking only individuals but differs from it in not involving reciprocal guidance. Exemplars or models, such as teachers, heroines and indeed popular musicians, guide or influence us. So, too, do counter-models. (‘I don’t want to turn into that sort of person!’) Models are not leaders, as Scheler points out. Models and counter-models, unlike leaders, may be real or fictive. Leaders lead by telling us what to do, unlike models. Just as following a rule resembles but is not the same thing as obeying an order, so too, following an exemplar or model is like but is not the same thing as obeying the order of a leader. The philosophy of guidance, then, is inseparable from social philosophy, the philosophy of perception and the philosophy of language. In what follows, I first look at the discussions of reading as an example of being guided (Sect. 2) by Bühler and Wittgenstein, the view that ‘guide’ is a family resemblance term and the view that the concept of guidance is what Bühler called a ‘synchytic [synchytisch] concept’, a concept which groups objects ‘according to a manifold similarity, that is, one that is not determined on the basis of only one aspect’ (Bühler 1982 [1934], 222; 2011, 247). I then set out the different views of Bühler and Wittgenstein about an analogy between a person’s being guided by something and a machine’s being controlled by something and between the corresponding possibilities (Sect. 3). In the final section, I look at the relations between the views of early Bühler and early Wittgenstein on rule-awareness (Sect. 4). The central aim of the paper is to show that some aspects of the thought of the later Wittgenstein amount to a very radical version of part of the philosophies of some of Brentano’s heirs. Brentano’s heirs, as is now

15  On Being Guided, Signals and Rules: From Bühler… 

303

well-known (perhaps even in the field of the history of analytic philosophy), moved along two very different paths. Husserl and Meinong, Reinach and Scheler multiplied entities, modalities and essentialisms and set out thoroughly anti-naturalist and anti-pragmatist views. Marty, on the other hand, argued that much of the multiplication was unnecessary. He developed a very thorough account of meaning-nominalism, in which the functions of expressing mental states and modifying the mental states of interlocutors are prominent, an account embedded in an understanding of language and other social phenomena as the product of complicated, unplanned processes of selection. Husserl and many others, on this account, were tempted by linguistic pictures, succumbed to the temptations and from then on were misled by just these pictures. This, Marty argues, is particularly true of Husserl’s 1900–1901 dream of founding the philosophies of language and logic and so of many parts of philosophy on the theory of sense, nonsense and ‘logical grammar’. Marty’s philosophy is in some respects a Helvetic compromise between the view of the multipliers and the views of (late) Brentano himself, an uncompromising form of nominalism combined with the conviction that all philosophers of a different mind must be misled by linguistic fictions. Marty’s philosophy of language was developed above all by Bühler, Ahlman and Landgrebe and anticipated by Martinak. In spite of their divergences, Brentano and his most famous heirs all took seriously the idea of philosophy as a theoretical enterprise. From the point of view of the development of Austro-German philosophy, Wittgenstein’s most original innovations were two in number. First, the claim that the project of philosophy as a theoretical enterprise is taken seriously only by those who are in the grip of misleading pictures. Second, the suggestion that what look like essential, non-contingent philosophical claims, in particular claims made by Brentano’s heirs, are at best reminders about how words ought to be used. The discussions of guidance by Bühler and Wittgenstein, as we shall now see, illustrate some aspects of this large picture.

304 

2

K. Mulligan

Reading and Guidance

Two claims are central in the descriptions of guidance and of reading given by Bühler and then Wittgenstein. First, the extraordinary variety of these phenomena. Second, their independence from particular experiences and sensations. Husserl often describes and analyses the phenomenon of reading. In his 1900–1901 Logical Investigations he uses the phenomenon in order to consider the relation between visual perception and ‘intellectual’ acts such as judging, supposing and meaning that p and to argue that there is nothing sensory about the latter. His conclusion influenced the Würzburg School of psychology, to which Bühler belonged. In their descriptions of reading as being guided, Bühler and Wittgenstein begin by setting aside what Husserl called intellectual acts: [H]ow do things stand in the case of optical linguistic symbols […]? In fluent reading, our eyes wander jerkily over the lines taking in one word-­ image after another. Of course, the thoughts, emotions and interests of the reader do not remain with the black forms which stand on the white background of the paper but with the objects being dealt with. […] As far as the process [Geschehen] [of reading] is concerned, one can say that the sensory impressions intervene [eingreifen], triggering and steering. If I read a book out loud, the choice of tempo, whether I read loudly or softly, the right intonation, whether I read phlegmatically or in a lively way, this is up to me. […] And if we want to characterise briefly the intervention [Eingreifen] of optical perceptions in these activities, there are probably no better terms than triggering and steering. (Bühler 1929 [1927], 75)

And when Wittgenstein invites us to ‘study the use of the expression “to be guided” by studying the use of the word “reading”’ he tells us that by the latter he means […] the activity of translating script into sounds, also of writing according to dictation or of copying in writing a page of print, and suchlike; reading in this sense does not involve any such thing as understanding what you read. (Wittgenstein 1972, Brown Book, § 66, 119)

15  On Being Guided, Signals and Rules: From Bühler… 

305

In the Blue Book, we are told that ‘guided’ is used in many different senses (Wittgenstein 1972, 41). There is, Wittgenstein also says, a ‘family of cases of deriving’ sound from text and […] in the same way we also use the word ‘to read’ for a family of cases. And in different circumstances we apply different criteria3 for a person's reading. (Wittgenstein 1968, § 164.)

In the following passage, Wittgenstein arrives at the conclusion that no one feature is essential to all cases of reading after mentioning only a handful of the numerous examples of types of reading he gives elsewhere: […] our eye glides [gleitet] over printed lines differently from the way it glides over arbitrary pothooks and flourishes. (I am not speaking here of what can be established by observing the movement of the eyes of a reader.) The eye glides, one would like to say, with particular ease, without being held up; and yet it doesn’t skid [rutscht]. And at the same time involuntary speech goes on in the imagination. That is how it is when I read German and other languages, printed or written, and in various styles.—But what in all this is essential to reading as such? Not any one feature that occurs in all cases of reading. (Compare the process [Vorgang] in the case of reading ordinary print with reading words which are printed entirely in capital letters, as solutions of puzzles sometimes are. What a different process it is!— Or reading our script from right to left.) (Wittgenstein 1968, § 168, modified transl.)

Steering, Bühler asserts in 1929, ‘can be described without referring to any “mental processes”’ (Bühler 1929 [1927], 42).4 But when he asserts that ‘the entire domain of perception can be dealt theoretically with the help of […] three functions, signals, indicators and symbols’ he also asserts that triggering and steering apply not only to ‘the aspect of behaviour’ but also to ‘the aspect of experience’. Steering ‘describes the process [Geschehen] [of perception] from the lowest animals to the most complicated perceptions of symbols’ (Bühler 1929 [1927], 75). Although Bühler insists on the variety of guidance, he does not describe anything like the variety of cases distinguished and discussed by Wittgenstein. He is particularly interested in the variety of perceptual guidance:

306 

K. Mulligan

Someone who reads in the ordinary way and remains with the content of his thoughts does not behave with respect to sense-impressions in the same ways as someone who makes an ‘observation’. The visual impressions of the reader are only there in order to excite certain ideas, thoughts and feelings etc. They are not noticed and considered in order to find out what it is that is seen and how it is. This is always the case when signs and indicators carry out their specific function. In the case of observations, it is quite different. […] Not all perceptions are observations; this is clearly shown by the example of ordinary reading. It is also shown by other examples: to consider an object within the aesthetic set [Einstellung, attitude] is to be oriented towards enjoyment. […] There are as many kinds of perception as there are specific aims of perception; arriving at a perceptual judgment is only one such aim. (Bühler 1930 [1918], 137, my emphasis)

He also distinguishes many different kinds of signals. Indexical expressions (Zeigwörter) are both signals and symbols (Bühler 1982 [1934], 90, 190). There are reception signals, action signals, position signals and individual signals. ‘That’ and ‘I’ are reception signals when they trigger eye-movements and so a reception and thus differ from action signals, such as imperatives (Bühler 1982 [1934], 107). ‘Here’ is a position signal, ‘I’ an individual signal (Bühler 1982 [1934], 95). Anaphoric uses of language are a sui generis type of guidance (Bühler 1982 [1934], 122–123). A speaker’s knowledge steers his grasp of utterances (Bühler 1982 [1934], 65, 171, 172). Appearances and sense impressions steer behaviour, bodily movements and goals and do this either immediately or immediately (Bühler 1929 [1927], 73–77, 80). What psychologists call a set (Einstellung) is to be understood in terms of possible steerings (Bühler 1929 [1927], 121). This ‘wealth’ of ways of guiding and of goals, Bühler says, is often overlooked. The ‘structural monism’ of the Berlin Gestalt psychologists, for example, cannot do justice to the variety of guidance (Bühler 1926, 1929 [1927], 121–122), unlike the Vienna Gestalt psychologists. The wealth and variety of signals and guidance corresponds to a variety of different types of understanding. ‘By nature, understanding comes in at least three kinds, in linguistic inquiry’ (Bühler 1982 [1934], 13, my emphasis). To the difference between symbols, signals and expression

15  On Being Guided, Signals and Rules: From Bühler… 

307

there correspond three types of understanding, the understanding of symbols, of signals and of expression, three ‘keys to understanding’. He calls them: ‘expression-keys’, ‘signal-keys’ and ‘symbol-keys’ (Ausdrucksschlüssel, Signalschlüssel, Symbolschlüssel) (Bühler 1982 [1934], 13). There are, he repeatedly asserts, different levels, strata and depths of understanding (Bühler 1926, 495, 507, see 474; 1929 [1927], 85, 135; 1930 [1918], 225).5 Already in 1909 Bühler notes that ‘understand’ and ‘sense of speech [einer Rede]’, which he calls ‘corresponding expressions’, are ‘not univocal’ and refer to the ‘ambiguity of the concept of understanding’ (Bühler 1909, 104, 109). One of the reasons why it is so difficult to understand understanding is that ‘in understanding, we complete a great deal, specify much that is indeterminate’ (Bühler 1909, 119). Yet another type of understanding is pictorial understanding. Bühler agrees with Wittgenstein that ‘there is understanding and failure to understand’ pictures and drawings, ‘understanding’ a picture may mean ‘various kinds of thing’; one example Wittgenstein gives is: A picture is perhaps a still-life; but I don’t understand one part of it: I cannot see solid objects there, but only patches of colour [Farbflecke] on the canvas. (Wittgenstein 1968, § 526)

Reading off colour contrasts between colour patches in a physical field or context, says Bühler, is one thing. Taking in the pictorial values in the context of a painting a quite different thing: Colour contrast […] is a phenomenon that can be read off from the symphysical field of the coloured patches [Farbflecken]. The case of the ‘context’ of the pictorial values in a painting as a whole, however, is substantially different. If a painter mixes the same grey on his palette three times and three times places physically the same patch of grey in a developing picture, this patch can three times (or more often) receive a different pictorial value in the context of the painting; for example, it can give the impression of being a shadow or a light reflection or the colour of an object [Gegenstandsfarbe] (for example a spot of dirt on a white table cloth)— quite regularly and automatically for the viewer who is in a normal state of readiness to receive. The structural law of the pictorial values of a painting

308 

K. Mulligan

is completely different from colour contrast; these pictorial values are in a synsemantic field and receive certain field values in it. The patches of colour (in general: the sense data) must be assigned a sign. (Bühler 1982 [1934], 165–166).

3

 isleading Pictures and Pianolas: M Machines and Modality, Rules and Rigidity

For someone to be actually guided by something is one thing. For her to be such that she could be guided by it is another thing. Similarly, for a machine to be actually guided by something is one thing. For it to be such that it could be guided in this way is another thing. This analogy was of great interest to Bühler and Wittgenstein. Unlike Bühler, Wittgenstein thinks that when philosophers talk about modality in this way they often give in to the temptation to see possibilities, abilities and powers as states of the bearers of these possibilities. As we shall see, to see modalities in this way is to be in the grip of a misleading picture, indeed it is to be guided by and misled by a picture. This particular misleading picture of modality had been described in 1908 by Marty, the Swiss philosopher to whom Bühler owes several of the guiding principles of his theory and philosophy of language. From the point of view of Marty and Wittgenstein, what Bühler says about the possibilities open to machines looks like a good candidate for exemplifying the mistake of assimilating modalities to states. In 1926 Bühler announces that he has borrowed the term Steuerung from the vocabulary of physics. There is control and steering not only of ‘dead systems’ but also in ‘animal and human communities’. It is a concept which applies to both behaviour and experience. Properly understood, it promises to help us understand intentionality. Brentano’s pathbreaking discovery of intentionality was of ‘so to speak static’, point-­ like intentionality. But intentionality of this sort is a mere abstraction from dynamic intentionality, in which an individual experience is a control-­lever (Bühler 1926, 487, 479).

15  On Being Guided, Signals and Rules: From Bühler… 

309

Whoever constructs a machine equips it with ‘degrees of freedom’ (Freiheitsgraden, Toleranzen). Indeed ‘there is no machine without such degrees of freedom’. And ‘what we find in the domain of technology, we may also assume to be realised and used in the domain of organic steering’ (Bühler 1926, 496–497). There is ‘no material machine with only fixed connections, that is without grades of freedom’ (Bühler 1926 504–505). Bühler is so pleased with his distinction between the rigid or fixed connections and degrees of freedom of machines and organisms that he applies it to natural language and to a number of different linguistic phenomena. ‘Intersubjective communication with the signs of natural language is exceptionally imperfect in conforming to the requirement’ of […] meaning constancy: the same word—the same meaning, wherever it is used. But the author of this book […] finds that the rigidity of a stiff riding boot may well have advantages, for example for the horse rider; for the speakers of the clear language of science are proud riders who insist on rigid, well-defined word-meanings. However, there are other advantages for intersubjective communication that are provided by a certain plasticity of the meaning spheres of our naming words. Modern technology knows that one can and must work with degrees of freedom in mechanical engineering; organisms have known this much longer. As in very complicated modern machines and the organs of organisms, there are safety devices to correct the degrees of freedom of the meaning spheres of our naming words. Supersummativity and subsummativity of attributive complexes increase the productivity of language to a remarkable extent and make possible laconic naming. This presupposes, of course, that a correction of the indeterminateness and equivocity of the complexes formed in this way is available for use. (Bühler 1982 [1934], 350)

It is in his working out of the ideas in this passage (see Mulligan 2012, chap VI; 1997) that Bühler anticipates contemporary, strongly contextualist accounts of meaning, accounts which Wittgenstein, too, is held to have anticipated. Are the possibilities open to machines and the abilities of human states thereof? Wittgenstein notes in Brown Book cases where someone is actually guided by something only if she also has certain abilities:

310 

K. Mulligan

B is guided by the particular combination of words in one of our three sentences if he could also have carried out orders consisting in other combinations of dots and dashes. And if we say this, it seems to us that the ‘ability’ to carry out other orders is a particular state of the person carrying out the orders. […] And at the same time we can’t in this case find anything which we should call such a state. (Wittgenstein 1972, 117)

What is true of the way we speak about abilities is, Wittgenstein thinks, also true of the way we talk about de re possibility in general: There are […] various reasons which incline us to look at the fact of something being possible, […] as the fact that he or it is in a particular state. Roughly speaking, this comes to saying that ‘A is in the state of being able to do something’ is the form of representation we are most strongly tempted to adopt. (Wittgenstein 1972, 117)6

This is just the error Marty denounces in 1908, […] the common perpetration of the fiction which consists in bringing what is not real […], in thoughts about it, far too close [heranrücken] to the real. […] Forces, […] and all possibilities […] are treated as things, which have effects and are effected. […] And one proceeds in the same way in a thousand other cases. (Marty 1976 [1908], 354)

As Wittgenstein puts the point, We say: ‘It isn't moving yet, but it already has the possibility of moving’— ‘so possibility is something very near reality’. (Wittgenstein 1968, § 194)7

He is here referring to a machine: We use a machine, or the picture of a machine, as a symbol for the particular action of a machine. […] ‘The machine’s action seems to be in it from the start’ means: we are inclined to compare the future movements of the machine in their definiteness to objects which are already lying in a drawer and which we then take out. (Wittgenstein 1968, § 193)

15  On Being Guided, Signals and Rules: From Bühler… 

311

When we consider whether someone is guided by certain signs, there is an inclination […] to say some such thing as that we could only decide this question with certainty if we could look into the actual mechanism connecting seeing the signs with acting according to them. For we have a definite picture of what in a mechanism we should call certain parts being guided by others. In fact, the mechanism which immediately suggests itself […] is a mechanism of the type of a pianola. Here, in the working of the pianola we have a clear case of certain actions, those of the hammers of the piano, being guided by the pattern of the holes in the pianola roll. (Wittgenstein 1972, 118)

Wittgenstein also distinguishes between what Bühler had called rigid or fixed connections and degrees of freedom: The machine as symbolizing its action: […] We talk as if these parts could only move in this way, as if they could not do anything else. How is this— do we forget the possibility of their bending, breaking off, melting, and so on? Yes; in many cases we don't think of that at all. We use a machine, or the drawing of a machine, to symbolize a particular action of the machine. For instance, we give someone such a drawing and assume that he will derive the movement of the parts from it. (Wittgenstein 1968, § 193)

But when we are concerned with predicting the actual behaviour of a […] machine we do not in general forget the possibility of a distortion of the parts and so on.—We do talk like that, however, when we are wondering at the way we can use a machine to symbolize a given way of moving— since it can also move in quite different ways. (Wittgenstein 1968, § 193)

Hintikka once claimed that Wittgenstein was interested in a problem which is common to both our following rules and to a machine’s being guided: […] Wittgenstein’s problem about rules […] has nothing to do with intentionality. Wittgenstein could have raised the same problem about machines instead of humans—and, in fact, did so. How is a machine guided by its

312 

K. Mulligan

blueprint? This question is parallel with the question: How does a rule guide my actions when I follow it? […] Wittgenstein discusses how the motions of a machine are ‘guided’ by its blueprint in Philosophical Investigations I, secs. 193–194 among other places. (Hintikka [n.d.])

Whatever we should think about Hintikka’s account of Wittgenstein’s problem about rules, it seems clear that Bühler thought that the understanding of machines and language had much in common: My impression is that what is commonly called the ‘understanding’ of a machine is a many-sided and in certain circumstances very complicated matter. […] The understanding of language is no different. (Bühler 1929 [1927], 135)

4

 arly Bühler and Early Wittgenstein E on Rule-Awareness

The views about mind, language and colours criticised by ‘the later’ Wittgenstein are often views which are not to be found in the writings of Frege, Russell and Moore or in the philosophical tradition, but rather in the writings of Brentano’s heirs (see Mulligan 2012). Occasionally, Wittgenstein tells us that a view he is criticising is a view he once held. Thus he says: [E]arlier I thought at one time that grammatical rules are an explanation of what I experience on one occasion when I use the word. They are as it were consequences or expressions of the properties which I momentarily experience when I understand the word. (Wittgenstein MS, 116, quoted by Hintikka [n.d.])

Is this view to be found in the Brentanian tradition? In 1908, Bühler put forward the account of thinking he had arrived at as a member of the already mentioned Würzburg School, an account in which certain types of experience are said to be ‘direct expressions of rules’:

15  On Being Guided, Signals and Rules: From Bühler… 

313

When we want to express a difficult thought we first choose a sentence form for it, we first become internally aware of the operational plan, and it is then this plan that marshals the words. When we understand a complex compound sentence we gain a knowledge of its grammatical structure, we know of the relations between the individual parts of the entire form. That also happens when we speak; when, for example, we begin a parenthesis with ‘when’ and suddenly break off at the end of the subordinate clause, we become conscious of the fact that we had expected something; what we expect is not only a completion with regard to the topic, but a grammatical completion: we expect a main clause. What we become specially aware of in all of these cases is what always or almost always mediates between thoughts and words, incidentally and without being particularly noticed: a knowledge of the sentence form and of the relationship of the parts of the sentence to each other, something that must be regarded as a direct expression of the grammatical rules that are active in us. (Buhler 1908, 84 ff., my emphasis)8

Bühler continued to endorse his 1908 account of a type of awareness which is a direct expression of rules active in us. He quotes approvingly the passage from his 1908 article quoted above in (Bühler 1982 [1934], 254). As already mentioned, some of the claims of the Würzburg psychologists derive from reflection on Husserl’s Logical Investigations. This is true of Bühler’s account of rule-awareness, which he presents as a modification of Husserl’s account of so-called categorial intuition: What is an awareness of rules [Regelbewusstsein]? A thought in which something that from a logical point of view we call a rule dawns on us. But this is not an unambiguous description. I can also simply mean [meinen] a rule as I can any other object. Awareness of rules is not this thinking of a rule [an eine Regel denken] but rather thinking a rule or thinking in a rule [denken einer Regel oder in einer Regel]. The object of rule-awareness is not the rule but the state of affairs or objective it refers to, is applicable to, from which it is perhaps derived. (Bühler 1907, 335, see 339)

Bühler’s later views about the place of rules and conventions in his masterly ‘theory’ of language are merely hinted at in various places.9 A much more thoroughly developed philosophy of linguistic rules is that

314 

K. Mulligan

given by Ahlman who, like Bühler, develops ideas of Marty. Like Marty, Bühler and Wittgenstein, Ahlman thinks of words as essentially tools or implements, a view roundly rejected by such heirs of Brentano as Husserl and Scheler. Like the early Bühler and Wittgenstein, Ahlman thinks there is an intuitive grasp of rules: Individuals normally have an intuitive awareness of the rules. […] More or less distinctly, more or less exactly they have in mind [vorschweben] the (ideal) content of these norms, which they try to realise in speaking and understanding; individuals judge each other’s linguistic activity and understanding of what is said from the viewpoint of these norms. (Ahlman 1926, 9)

But is our most basic grasp of the rules we follow and infringe really to be understood in the ways suggested by Bühler, early Wittgenstein and Ahlman? Two philosophers who came to the conclusion that this grasp must be non-cognitive are Scheler and then Wittgenstein.10,11

Notes 1. Scheler and Bühler also use the expression ‘Signalement’. 2. In 1931 Charlotte Bühler and Paul Lazarsfeld add: ‘The fact of reciprocal steering must be the constitutive element of social psychology’ (Bühler and Lazarsfeld 1931, 335). 3. In 1909 Bühler attributes to Paulhan what he takes to be the correct idea of using as ‘an objective criterion of understanding a correct reaction to perception of language’ (Bühler 1909, 115). He continues to use the term ‘criteria’, sometimes, like early Wittgenstein, as a synonym of ‘symptom’, sometimes sharply distinguishing the two. Bühler’s fullest account of the nature of criteria (Bühler 1930, 367–377) identifies criteria with reasons or justifications for that for which the criteria are criteria, for example the certainty of memory. He sharply distinguishes the relation of justification and causal relations. But, like Husserl, he also talks of an experience of the relation of justification, of noticing the relation between certainty and the criteria it is based on, an idea Wittgenstein will criticise. Elsewhere Bühler uses ‘criteria’ simply to mean conditions

15  On Being Guided, Signals and Rules: From Bühler… 

315

for being F. On criteria and symptoms and related notions in Bühler, Wittgenstein and Husserl, see (Mulligan 1990). 4. The concept of ‘reciprocal steering’, according to Charlotte Bühler, is a concept which ‘in principle already transcends introspection’ (Bühler and Lazarsfeld 1931, 335). 5. On understanding, see (Bühler 1929 [1927], 124, 133, 135; 1930 [1918], 225, 406). 6. The ‘form of representation’ mentioned here is an example of the phenomenon of the so called subjective a priori, as Scheler calls it, mentioned at the beginning of § 1 above. 7. On the misleading pictures and associated temptations denounced by Marty and Wittgenstein and their views of such pictures and temptations, see (Mulligan 2019). 8. On the use and understanding of Darstellung (representation), Symbol and Zuordnung (correlation) in early Bühler and early Wittgenstein, see (Mulligan 2012, chapter V). 9. See for example (Bühler 1926, 494). 10. For the details, see (Mulligan 2006). 11. Versions of the material in this paper were presented most recently in Liège (2018), under the title “On Being Guided: Bühler in the Brown Book”, and in Vienna (2013). I am grateful to the audiences for their comments. Thanks, too, to Peter Simons for help with engineering terminology.

References Ahlman, Erik. 1926. Das Normative Moment im Bedeutungsbegriff, Helsingfors. (Suomalaisen Tiedeakatemian Toimituksia, Annales Academiae Scientiarum Fennicae, B, XVIII, Helsinki 1923–1926). Bühler, Charlotte, and Paul Lazarsfeld. 1931. Der Behaviorismus. In Einführung in die neuere Psychologie, ed. Emil Saupe, 332–341. Osterwieck-Harz: Zickfeldt. Bühler, Karl. 1907. Über Gedanken. Archiv für die gesamte Psychologie 9: 297–365. ———. 1908. Über Gedankenerinnerungen. Archiv für die gesamte Psychologie 12: 24–92.

316 

K. Mulligan

———. 1909. Über das Sprachverständnis vom Standpunkt der Normalpsychologie aus. In Bericht über den dritten Kongress für experimentelle Psychologie, 94–130. Leipzig: J.A. Barth. ———. 1926. Die Krise der Psychologie. Kant-Studien 31: 455–526. ———. 1929 (1927). Die Krise der Psychologie. Jena: Fischer. ———. 1930 (1918). Die geistige Entwicklung des Kindes. Jena: Fischer. ———. 1982 (1934). Sprachtheorie. Die Darstellungsfunktion der Sprache. Stuttgart: Fischer. ———. 2011. Theory of Language. The representational function of language. Trans. Donald Fraser Goodwin. Amsterdam: John Benjamins. Hintikka, Jaakko. n.d. Did Wittgenstein Follow the Rules? (or was he guided by them?). Available on: http://people.bu.edu/hintikka/Papers_files/DID%20 WITTGENSTEIN%20FOLLOW%20THE%20RULES.pdf Marty, Anton. 1976 (1908). Untersuchungen zur Grundlegung der allgemeinen Grammatik und Sprachphilosophie. Hildesheim and New York: Olms. Mulligan, Kevin. 1990. Criteria and Indication. In Wittgenstein—Towards a Re-evaluation, ed. Rudolf Haller and Johannes Brandl, 94–105. Vienna: Hölder–Pichler-Tempsky. ———. 1997. The Essence of Language: Wittgenstein's Builders and Bühler's Bricks. Revue de Métaphysique et de Morale 2: 193–216. ———. 2006. Soil, Sediment and Certainty. In The Austrian Contribution to Analytic Philosophy, ed. Mark Textor, 89–128. London: Routledge. ———. 2012. Wittgenstein et la philosophie austro-allemande. Paris: Vrin. ———. 2019. Misleading Pictures, Temptations and Meta-Philosophies: From Marty to Wittgenstein. In Anton Marty and Contemporary Philosophy, ed. Giuliano Bacigalupo and Hélène Leblanc, 197–232. Basingstoke: Palgrave Macmillan. Wittgenstein, Ludwig. 1968. Philosophical Investigations. Oxford: Blackwell. ———. 1972. The Blue and Brown Books. Oxford: Blackwell. ———. 1980. Remarks on the Philosophy of Psychology. Vol. I. Oxford: Blackwell.

Index1

A

Ahlman, Erik, 300, 302, 303, 314 Ajdukiewicz, Kazimierz, 3, 24, 227–239 Albertazzi, Liliana, 2, 4, 77, 86, 88, 90n1, 91n6 Ameseder, Rudolf, 2 Aquinas, Thomas, 120–123, 125 Araujo, Saulo de Freitas, 295n10 Aristotle, 72n2, 133, 245, 249 Augustine, 121, 124, 125, 133 Austin, John Langshaw, 18, 73n9, 257 B

Bacigalupo, Giulano, 3, 136n2 Bacon, Roger, 133

Bain, Alexander, 59 Barsalou, Lawrence, 276n7 Beaney, Michael, 26 Bégout, Bruce, 266 Benoist, Jocelyn, 3, 57, 163n3, 265, 276n6 Bentham, Jeremy, 40–44, 52n9 Benussi, Vittorio, 2 Berlage, Andreas, 85 Bernays, Paul, 233 Boghossian, Paul, 20 Bolzano, Bernard, 36, 42–44, 46, 47, 51, 52n6, 53n12, 53n13, 53n14, 57, 58, 123, 127–130, 135, 153, 199, 207, 209–211, 248, 249, 251 Bonald, Louis de, 73n5, 120–122, 127

 Note: Page numbers followed by ‘n’ refer to notes.

1

© The Author(s) 2021 A. Dewalque et al. (eds.), Philosophy of Language in the Brentano School, History of Analytic Philosophy, https://doi.org/10.1007/978-3-030-52211-7

317

318 Index

Boole, George, 61, 230 Bourdeau, Michel, 96 Brandl, Johannes, 96 Bredeck, Elizabeth, 91n3 Brentano, Franz, 1–10, 15–20, 43, 57–72, 77–90, 95–115, 119–135, 143–163, 190, 210, 231, 247–251, 265, 284, 299 Bühler, Charlotte, 314n2, 315n4 Bühler, Karl, 3, 15, 25, 36, 50, 51, 71, 123, 279–294, 295n14, 295n15, 299–314 Burge, Tyler, 20

D

Dąmbska, Izydora, 3 Danziger, Kurt, 282 Davidson, Donald, 236 De Beer, Caroline, 283, 288 De Ruiter, Jan P., 283, 288 Del Pinal, Guillermo, 264, 273, 274 Dewalque, Arnaud, 2, 22, 26n5, 112 Dretske, Fred, 20 Dummett, Michael, 21, 35, 36, 38–41, 45, 47, 48, 50, 51, 52n4, 57 Duncan, Susan D., 283 E

C

Carnap, Rudolf, 12, 24, 209, 215, 222n21, 230, 239 Castañeda, Héctor-Neri, 211, 219, 220 Cavallin, Jens, 191, 201, 202 Centrone, Stefania, 128 Cesalli, Laurent, 3, 4, 7, 17, 19, 54n21, 78, 82, 122, 123, 126, 131–133, 135, 136n2, 136n5, 287 Champeaux, Guillaume de, 219 Champeaux, William of, 86 Chase, James, 96 Chisholm, Roderick M., 6, 212, 219, 222n17, 222n20, 222n24 Chrudzimski, Arkadiusz, 161, 176 Claas, Jan, 250, 251 Clarke, David D., 284 Coffa, J. Alberto, 57 Conimbricenses, 125, 126

Ebbesen, Sten, 260n2 Eisler, Rudolf, 229 Ekman, Paul, 286 Erdmann, Benno, 265, 268 F

Fahrenberg, Jochen, 284 Falk, Eugene Hannes, 174 Fennell, John, 5 Findlay, John N., 129, 212, 213, 216, 219, 220n4, 221n7, 221n11, 267, 271 Fisette, Denis, 3, 13, 23, 152, 155, 191, 203, 232 Fonseca, Pedro da, 127 Fréchette, Guillaume, 3, 13, 14, 21, 63, 119, 130, 136n2, 136n5, 146, 152 Frege, Gottlob, 5, 12, 15, 18, 21, 24, 35–41, 44, 47–51, 51n1, 52n2, 52n3, 52n9, 57, 58, 61,

 Index 

63, 69, 70, 123, 152, 154, 208, 209, 211, 221n6, 221n10, 221n11, 221n12, 222n15, 229, 230, 232, 256, 312 Fricke, Ellen, 295n4 Friedrich, Janette, 3, 282 Friesen, Wallace V., 286 G

Gakis, Dimitris, 91n5 Gallerand, Antoine, 152, 163n3 Gauvry, Charlotte, 21 Goclenius, Rudolf, 229 Grabmann, Martin, 135 Grätschenberg, Richard, 129 Grice, Paul, 19, 123

319

196, 199, 200, 202, 203, 209, 220–221n4, 222n22, 245–260, 263–275, 300, 301, 303, 304, 313, 314, 314–315n3 I

Ingarden, Roman, 2, 4, 14, 22, 23, 96, 143–163, 163n1, 163n2, 164n6, 164n7, 164n9, 169–185, 189, 190, 203 J

Jacquette, Dale, 211, 212 Janoušek, Hynek, 115n1 Jerusalem, Wilhelm, 265, 275n2 Jevons, William Stanley, 61

H

Hacker, Peter M.S., 21, 35, 36, 40 Hale, Bob, 52n4 Haller, Rudolf, 16, 192 Hampton, James A., 264, 276n7 Heidegger, Martin, 96 Heyse, Paul Johann von, 59 Hintikka, Jaakko, 211, 311, 312 Hobbes, Thomas, 106, 107 Höfler, Aloïs, 2, 130, 220n1, 221n7 Huemer, Wolfgang, 2, 82 Humboldt, Wilhelm von, 45, 59, 276n8 Husserl, Edmund, 2, 13, 14, 22, 24, 25, 50, 51, 86, 91n5, 95–97, 123, 129, 130, 133, 144, 145, 152–155, 157–163, 163n2, 163n3, 164n6, 169, 190–194,

K

Kasabova, Anita, 128 Kastil, Alfred, 3, 14, 82 Katkov, Georg, 3 Kendon, Adam, 280, 283, 295n5 Kenny, Anthony, 36, 37 Knobloch, Clemens, 285, 292 Kokoszyńska, Maria, 236 Kotarbiński, Tadeusz, 3, 237 Kraus, Karl, 3, 14, 16, 17, 77, 78 Kraus, Oskar, 3, 16, 18, 47, 48, 63, 69, 77, 78, 136n2 Kriegel, Uriah, 2, 3, 10 Kries, Johannes von, 265, 275n2 Kripke, Saul, 250 Krug, Wilhelm, 229 Künne, Wolfgang, 128

320 Index L

Lambert, Karel, 211 Landgrebe, Ludwig, 71, 303 Lange, Friedrich-Albert, 85 Laurence, Stephen, 263, 264, 275n4, 276n7 Lazarsfeld, Paul, 314n2, 315n4 Le Rider, Jacques, 85, 88, 89, 91n3 Leblanc, Hélène, 3, 22, 73n5, 125, 136n2 Leclercq, Bruno, 23, 73n8, 149, 151 Leśniewski, Stanisław, 3, 24, 27n6, 232–235, 237, 239 Lewis, Clarence Irving, 24, 234 Libardi, Massimo, 78 Libera, Alain de, 135 Liedtke, Frank, 19, 123 Lindenfeld, David F., 147 Locke, John, 106, 222n20 Lohmar, Dieter, 266 Longworth, Guy, 15, 19, 123 Lotze, Rudolf Hermann, 5 Łukasiewicz, Jan, 3, 13, 24, 155, 232–235, 239 M

Mach, Ernst, 16, 22, 77, 85, 89 Majolino, Claudio, 54n21, 122, 126, 129, 133, 135, 136n5 Malherbe, Olivier, 22, 144, 163n1, 186n13 Mally, Ernst, 2, 211, 216 Marek, Johann Christian, 163n2, 220n1 Margolis, Eric, 263, 264, 275n4, 276n7 Marmo, Costantino, 125

Marthelot, Perrine, 73n10 Martinak, Eduard, 2, 4, 130, 300, 303 Marty, Anton, 2–4, 7, 8, 14–20, 22, 26n3, 26n4, 45–51, 53n10, 53n12, 54n18, 57, 59, 60, 68, 73n3, 73n4, 78, 87, 88, 119–135, 136n2, 136n4, 145, 146, 163n2, 190, 251, 265, 275n1, 300, 303, 308, 310, 314, 315n7 Mauthner, Fritz, 4, 10, 16, 17, 21, 77–90 Mayer-Hillebrand, Franziska, 18, 54n17, 59, 63, 69, 82 McDowell, John, 20 McNeill, David, 283, 285, 286, 288, 295n4, 295n6, 295n8 Meier-Oeser, Stephan, 125, 134 Meinong, Alexius, 2, 4, 14, 22, 23, 26n5, 57, 95, 97, 107, 108, 110, 147–151, 153, 162, 164n9, 207–220, 265, 303 Mervis, Carolyn, 264, 265 Mill, John Stuart, 37, 43, 44, 51n2, 59, 60, 66, 67, 97, 106, 107, 111, 249 Miskiewicz, Wioletta, 144 Mitscherling, Jeff, 176 Moltmann, Friederike, 20 Moore, George Edward, 15, 115n2, 312 Morran, Dermot, 96 Müller, Cornelia, 295n3, 295n4 Mulligan, Kevin, 3, 4, 7, 16–19, 25, 27n7, 47, 48, 54n21, 69, 72n1, 78, 82, 131, 265, 275n2, 276n8, 281, 309, 312, 315n3, 315n7, 315n8

 Index  N

Nájera, Elena, 91n5 Nerlich, Brigitte, 284 Nietzsche, Friedrich, 85 O

Ogden, Charles Kay, 41 P

Parsons, Terence, 211, 214, 215 Partee, Barbara, 260n1 Paśniczek, Jacek, 211 Pfänder, Alexander, 2 Plato, 5, 86, 219, 222n20, 263 Poli, Roberto, 26n2 Post, Emil, 233 Prinz, Jesse, 264 Putnam, Hilary, 20, 236 Q

Quine, Willard van Orman, 6, 21, 35, 39–41, 44, 51, 211, 221n11 R

Rapaport, William, 211, 216 Raynaud, Savina, 136n2 Recanati, François, 19, 20, 123, 276n5 Reinach, Adolf, 2, 14, 27n7, 303 Reynolds, Jack, 96 Richard, Sébastien, 22, 147, 171, 276n8 Rollinger, Robin D., 4, 6, 115n1, 136n1

321

Rosch, Eleanor, 264, 265 Roth, Emilie, 264, 273 Routley, Richard, 211, 212, 214, 222n16, 222n18 Russell, Bertrand, 12, 15–17, 24, 40, 41, 47, 58, 67, 86, 89, 108, 209, 211, 221n11, 229, 230, 312 Ryle, Gilbert, 6, 10, 15, 16, 22, 67, 68, 95–115, 221n11 S

Sainsbury, Richard Mark, 274 Sauer, Werner, 115n1 Savonarola, Girolamo, 248 Schaar, Maria van der, 15, 24, 158, 189, 201, 203, 250 Scheler, Max, 2, 91n5, 281, 300, 302, 303, 314, 314n1, 315n6 Schlick, Moritz, 6 Schnieder, Benjamin, 250, 251 Schuhmann, Karl, 3, 96 Searle, John, 6, 20 Sellars, Wilfrid, 6 Seron, Denis, 21, 85, 89, 92n8, 97, 176 Shoben, Edward J., 264, 273 Sierszulska, Anna, 147, 149 Simons, Peter, 12, 57, 163n3, 172, 315n11 Sluga, Hans, 57 Smith, Barry, 14, 103, 152, 163n3 Soames, Scott, 20 Srzednicki, Jan, 26n3, 48 Stainton, Robert, 52n4 Stein, Edith, 2 Steinthal, Heymann, 45, 59, 122

322 Index

Stout, George, 15 Streeck, Jürgen, 280, 295n6 Stróżewski, Władysłąw, 178 Stumpf, Carl, 2, 52n9, 163n4, 191, 194–196, 199, 200, 202, 203

Voltaire, 84 Von Hildebrand, Dietrich, 2 Vrahimis, Andreas, 96 W

T

Taieb, Hamid, 3, 19, 24, 25, 73n7, 119, 126, 136n2, 136n5, 275n1 Tarski, Alfred, 3, 24, 231, 234–237, 239 Textor, Mark, 3, 20, 275n2, 276n8 Thomasson, Amie L., 96, 116n10 Twardowski, Kazimierz, 2, 4, 9, 11–15, 22–24, 26n4, 57, 64, 73n7, 73n8, 130, 144–146, 152, 153, 155–158, 160, 163, 163n2, 163n5, 164n7, 189–203, 207–220, 220n4, 221n5, 221n7, 227–239, 249–251

Weiler, Gershon, 16, 77, 78, 86–88, 91n5 Whewell, William, 265, 268, 269, 276n4 Williamson, Timothy, 246 Windelband, Wilhelm, 190 Wisdom, John, 41 Wittgenstein, Ludwig, 10, 15–19, 21, 22, 24, 25, 36, 40, 47, 58, 67, 69, 70, 77, 78, 91n5, 233, 264, 265, 281, 299–314 Woleński, Jan, 23, 201, 235 Wright, Crispin, 52n4 Wundt, Wilhelm, 25, 45, 59, 279–294, 295n6, 295n7, 295n10, 295n12

U

Ungeheuer, Gerold, 284 Z V

Vaihinger, Hans, 16, 22, 77, 88, 91n4 Vitasek, Stephan, 2

Zalta, Edward N., 211, 216 Zawirski, Zygmunt, 227, 228, 232, 234 Zimmerman, Robert, 129