Philippine and Chamorro Linguistics Before the Advent of Structuralism 9783050056197, 9783050052144

Citation preview

Philippine and Chamorro Linguistics Before the Advent of Structuralism

Koloniale und Postkoloniale Linguistik Colonial and Postcolonial Linguistics Band 2 Herausgegeben von Stefan Engelberg, Peter Mühlhäusler, Doris Stolberg, Thomas Stolz und Ingo H. Warnke

Lawrence A. Reid, Emilio Ridruejo, Thomas Stolz (Eds.)

Philippine and Chamorro Linguistics Before the Advent of Structuralism

Akademie Verlag

Bibliografische Information der Deutschen Nationalbibliothek Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliografie; detaillierte bibliografische Daten sind im Internet über http://dnb.d-nb.de abrufbar.

© 2011 Akademie Verlag GmbH, Berlin Ein Wissenschaftsverlag der Oldenbourg Gruppe. www.akademie-verlag.de Das Werk einschließlich aller Abbildungen ist urheberrechtlich geschützt. Jede Verwertung außerhalb der Grenzen des Urheberrechtsgesetzes ist ohne Zustimmung des Verlages unzulässig und strafbar. Das gilt insbesondere für Vervielfältigungen, Übersetzungen, Mikroverfilmungen und die Einspeicherung und Bearbeitung in elektronischen Systemen. Redaktion: Cornelia Stroh Einbandgestaltung: hauser lacour, nach einer Idee von Susanne Hackmack Druck: MB Medienhaus ,Berlin Bindung: Norbert Klotz, Jettingen-Scheppach Dieses Papier ist alterungsbeständig nach DIN/ISO 9706 ISBN 978-3-05-005214-4

Table of contents

LAWRENCE A. REID, EMILIO RIDRUEJO &THOMAS STOLZ Preface ..........................................................................................................................

7

EMILIO RIDRUEJO The first grammar of a Philippine language? The anonymous Arte de la Lengua Sambala y Española [Grammar of the Sambal and Spanish languages] (1601) ........... 11 JOHN U. WOLFF The Vocabulario de Lengua Tagala of Fr. Pedro de San Buenaventura (1613) ............. 33 REBECA FERNÁNDEZ Hispanic Philippine historiography: a brief description of the structure of the Calepino Ylocano (18th C.) ................................................................................................................ 49

OTTO ZWARTJES Oyanguren de Santa Inés‘s grammar of Tagalog (Tagalysmo Elucidado 1742): towards a reconstruction of 18th century reflections on comparative typology........... 63 PIERRE WINKLER Subject, Topic, Passive and Perspective in Functional (Discourse) Grammar and in Philippine Missionary Grammar................................................................................... 87 IWAR WERLEN Wilhelm von Humboldt and the role of the verb in Tagalog........................................ 117

6

Table of contents

LAWRENCE A. REID Seidenadel‘s grammar of Bontoc Igorot: one hundred years on......................................... 141 KLAUS ZIMMERMANN The Diccionario español-chamorro (1865) by Padre Fray Aniceto Ibáñez del Cármen: a historiographical characterization of a pedagogic-lexicographic discourse type in late colonial Austronesia................................................................... 163 THOMAS STOLZ The Gramática chamorra............................................................................................. 183 THOMAS STOLZ German and Dutch contributions to Chamorro studies (1800–1920)............................. 201 THOMAS STOLZ, CHRISTINA SCHNEEMANN, BARBARA DEWEIN & SANDRA CHUNG The mysterious H. Who was the author of Die Chamoro Sprache?............................ 227 List of contributors ..................................................................................................... 243

LAWRENCE A. REID (HONOLULU/OSAKA), EMILIO RIDRUEJO (VALLADOLID) & THOMAS STOLZ (BREMEN)

Preface

Only a small segment of the world‘s languages boast of a long documented history. To the overwhelming majority however, this privilege does not apply. Thus, in many cases, it is illusory to write a philologically informed diachronic grammar. Historical linguistics in terms of reconstruction is often feasible if genetically related languages exist. However, reconstructing proto-stages does not exhaust the agenda of the historical linguist. One wants to know more about relatively recent developments. For many languages spoken in former colonies of western powers, this includes the dynamics of contact-induced change wherever the language of the colonisers has exerted influence on the lexicon and grammar of the languages of the colonised. Moreover, early documents are helpful when it comes to language planning, revitalisation or standardisation. Speech-communities need access to earlier stages of their language to recover lost territory in the realms of lexicon, idioms, constructions, etc. in order to prepare their languages for Ausbau without relying too much on the languages of their colonisers. Among the languages which are underprivileged in terms of historical documentation, there are the countless autochthonous languages of the Pacific and their distant relatives in Southeast Asia. On closer inspection, one realises that this is too sweeping a statement as some languages of this region stand out from the vast majority of their neighbours. They stand out because there are bits of text in and/or about these languages which predate the turn of the 19th century. Apart from the occasional case of autonomous pre-colonial indigenous literacy, we find examples of a noteworthy tradition of Missionary descriptive linguistics during Spanish colonial dominance in the Philippines and in the Marianas ranging from the mid-16th century to the late 19th century. Owing to the withdrawal of Spain in the aftermath of the Spanish-American War of 1898, other foreigners entered the scene and thus US-American, German, Dutch and other scholars began to write extensively on and sometimes also in the languages of the newly acquired territories. When the 2nd World War settled the question of political dominance in this region in favour of the United States, those traditions based on languages other than English soon became obsolete and have fallen into oblivion. Since the knowledge of the languages of former (and sometimes only ephemeral) colonisers (including clerical Latin) among the native speakers of Philippine and Micronesian lan-

8

Lawrence A. Reid, Emilio Ridruejo & Thomas Stolz

guages is negligible nowadays, the older descriptive sources dealing with their native languages are not easy to consult. To the language problem, we must add the problem that some of the texts have never appeared in print while those that have are hidden in publications which are hard to come by in university libraries and elsewhere. This lack of accessibility of the older sources may create the image of languages without a written history of their own. This image in turn might have a negative effect on the speakers whose loyalty towards their native language largely depends on the cultural value associated with the language. If it does not have a history of written records (rather than the history constructed in modern accounts by outsiders), it may not be able to compete with the more prestigious English (or other first world language). With a view to remedying this situation, specialists of the history of the languages of the Philippines and Micronesia agreed to get together to exchange ideas and make plans for a future joint project. The meeting took the shape of a two-day workshop on Philippine and Micronesian Linguistics before the advent of structuralism (28–29 August, 2008). The workshop was part of the programme of the 11th International Conference on the History of the Language Sciences hosted by the University of Potsdam/Germany. Scholars from Germany, the Netherlands, Spain, Switzerland, and the US gave talks on the early (non-English) sources of the indigenous languages of the Philippines and the Marianas. The presentations and the ensuing discussion proved to be so stimulating that it was decided not only to publish the papers based on the earlier talks but also to establish a network among the participants and other interested scholars in order to pursue the common interests further. Doing research in this sector of the history of linguistics is fascinating. Furthermore, from this endeavour, Austronesian linguistics and the modern speechcommunities of the languages under scrutiny may gain considerably as long lost sources of linguistic and cultural value have come back into focus such that in the not too distant future, they will be accessible to everybody again (or for the first time ever). Consistent with the programme of the above workshop, the papers in this edited volume are organised geographically and chronologically. Older sources of Philippine languages are discussed first. In the second part, the focus is on Chamorro, the indigenous Austronesian language of the Marianas. Emilio Ridruejo (Valladolid/Spain) sets the scene with his presentation of the anonymous first grammar of a Philippine language published in Spanish perhaps as early as 1601. Further early Spanish contributions to the descriptive linguistics of the languages of the Philippines are discussed in the papers by John U. Wolff (New York/USA) on the first Tagalog dictionary, Rebeca Fernández (Valladolid/Spain) on Pedro Vivar‘s treatment of Ilocano, Otto Zwartjes (Amsterdam/Netherlands) on the early comparative approach adopted by Oyanguren de Santa Inés, Pierre Winkler (Haarlem/Netherlands) on the affinities of the Philippine Missionary grammarians with modern Functional Grammar, Iwar Werlen (Berne/Switzerland) provides an appraisal of Wilhelm von Humboldt‘s treatment of the verb in Tagalog and Lawrence A. Reid (Honolulu/USA) concludes the Philippine section with a discussion of the achievements of Seidenadel in the realm of Bontoc Igorot grammar.

Preface

9

In the Chamorro section the work of the Spanish cleric Fray Aniceto Ibáñez del Carmen is the topic of two papers: Klaus Zimmermann (Bremen/Germany) scrutinises his Spanish-Chamorro dictionary from the point of view of a constructivist theory of language (contact) whereas Thomas Stolz (Bremen/Germany) highlights some linguistically striking aspects of the Spanish grammar written in Chamorro by Ibáñez del Carmen in the mid-19th century. In his second paper, Thomas Stolz surveys German and Dutch contributions to the study and description of Chamorro from the early 19th century until the end of the 1st World War. In the final paper, Thomas Stolz, Christina Schneemann, Barbara Dewein (Bremen/Germany) and Sandra Chung (Santa Cruz/USA) jointly solve the problem of the identity of H. Costenoble, the author of the most sizable grammar of Chamorro. The editors would like to express their gratitude to the Fritz Thyssen Foundation (Cologne/Germany) for their generous grant with which the workshop in Potsdam was supported financially. We are also indebted to Cornelia Stroh (Bremen/Germany) whose expertise in editorial matters brought this volume to fruition within only a year after the workshop. Honolulu/Osaka Valladolid Bremen The editors, August 2009

EMILIO RIDRUEJO (VALLADOLID/SPAIN)

The first grammar of a Philippine language? The anonymous Arte de la Lengua Sambala y Española [Grammar of the Sambal and Spanish languages] (1601)

Abstract This article describes the content of a manuscript kept in the Philippine Augustinian Archives in Valladolid, entitled Arte de la Lengua Sambala y Española [Grammar of the Sambal and Spanish languages]. The manuscript, which dates back to1601, is a compendium that was apparently compiled from several grammatical studies and a vocabulary list, all of them about the Sambal language spoken in Bolinao, presently a municipality in the north of the Sambal Peninsula in central Luzon. As a whole, the work can be categorised as part of the Philippine grammatical tradition of the 17th Century because of the common grammatical structure and terminology that it shares. Therefore, it is probably not the first grammar of a Philippine language, but it certainly has very early features and the writing date of some paragraphs is old and may be close to 1601.

1. The manuscript In the Archive (not the Library) of the Philippine Augustinian convent in Valladolid there is a manuscript which contains a Zambal or Sambal grammar in the shelf marked 1182/2.1 The manuscript is written on European style paper measuring approximately 190 x 139 millimetres. It consists of 157 pages, although the page numeration only goes up to 167, due to a gap between pages 79 and 90. The manuscript is bound in the Spanish style, probably dating from the 18th or early 19th century. It is in a relatively satisfactory state of repair. However there are a number of patches dues to dampness, holes made by insects and several torn pages. It has been restored, probably during the binding process, using strips of paper stuck to the edges of some pages. These strips are not intended to hide the writing, although at times they do so, and prevent some lines from being read. The text has been written by different hands, some of which date from the 17th century, and there are also technical corrections and superimposed comments. There is no index. On the final page the following note is made in a reddish ink: ―Este Arte hallé tirado en el Pueblo de Bolinao, quando estuve allí el año 1780. Fr. Agustín María‖ [I found this Grammar on the ground in the village of Bolinao, while I was there in the 1

Sueiro Justel (1997, 2003) does not quote it. However, it appears in Santiago Vela‘s bibliography (1913–1935) and in Hernández (1990).

12

Emilio Ridruejo

year 1780. Fr. Agustín María.]. This is followed by a rubric. The front cover, without a doubt written by the same person that wrote the colophon, shows the date 1601. Bolinao, the village where the manuscript was found, is today a municipality of the province of Pangasinan, on the north-west coast of Luzon. Evangelisation by the Spaniards began early in this region, in 1583.2 According to the census of 2000, Bolinao has a population of over 60,000 inhabitants. Today it is not part of the Zambales province, which is to the south, although it did belong to that province during colonisation and until 1903, when North American authorities introduced changes in regional boundaries.

2. Authorship Although the work is anonymous, it is possible to identify the person who wrote the colophon, prologue and the first few pages. It was almost certainly Fray Agustín María de Castro, as suggested by Santiago Vela (1913–1935, vol. 5 of 1920: 165), as his initials coincide with those found in the text, as do the dates with the time he spent in Zambales. Father Castro (1740–1801) was an Augustinian born in La Bañeza (León), who was ordained in Valladolid in 1756, and after travelling to Mexico, arrived in the Philippines in 1759. He fought against the British in 1762 and, once peace was restored, he travelled around all the provinces of the islands (Santiago Vela 1913–1935, vol. 1 of 1913: 660–665). During his travels he collected material from convent archives which he used to write a biographical work of the Augustinian missionaries in the Orient (Castro 1954 [1780]). Fr. Agustín María de Castro, on the front cover and in the prologue of the manuscript, does not attribute the work to a particular author, as he does in his book Osario. In the biography of Fr. Esteban Marín (approx. 1556–1601 or 1602), the first Augustinian monk posted to the evangelisation of the Sambal people, considers him to be the author of the Arte de la lengua sambala he saw in Bolinao.3 However, Santiago Vela indicates that this monk did not minister in that region for more than two years and was moved from there in 1587.4 Other contemporaries suggested by Santiago Vela as possi2

3

4

Father Martín de Rada visited Bolinao coast and harbour in 1575 during an expedition to China, but he was not welcomed. (Rodríguez Rodríguez 1978: 272; 331) ―He wrote Arte y Diccionario [Grammar and Dictionary] of the Tina language or old Sambalic, a volume in quarto, 420 pages, that I saw and read in the Archive of the Santiago convent in Bolinao. In Iba and Masinloc I saw and read many of his sermons written in this language. He also wrote Arte y Catecismo [Grammar and catechism] in the Igorot language, a volume in quarto, manuscript, 32 pages, that I have in my hands‖ (Castro 1954 [1780]: 85–86). However, the work mentioned here by father Castro does not match in length the manuscript kept in Valladolid. Father Esteban Martín had professed in 1576 and was assigned to Bolinao in 1585. In 1587 he was appointed Prior in Batac, then in Tagurin (1589), Laoag (1590) and again in Batac (1595). Father Martín died in 1601 or 1602 murdered during a military expedition against the Ilongot or Italon people in the mountainous region of Pampanga (Rodríguez Rodríguez & Álvarez Fernández 1992 I: 435).

The first grammar of a Philippine language?

13

ble authors are Fr. Francisco Martínez Ramos (approx. 1543–1608 or 1609), made Prior of Bolinao on the 31 of October 1600,5 and Fr. Juan de las Casas (ý1607), who arrived in the Philippines in 1592 and who, after working for an unknown amount of time in Zambales, died – shot dead by the indigenous people with their arrows (Santiago Vela 1913–1935, vol. 5 of 1920: 165). The work has also been attributed to Fray Rodrigo de Aganduru (1584–1626), who in the early 17th century made numerous trips to the Philippines, Malacca, Ceylon, Goa and, later, to the Middle East. However, there seems to be no apparent reason for this other than the possibility that this friar was an erudite traveller (Santiago Vela 1913–1935, vol. 1 of 1913: 32). The author of the main body of this work is a cultured person, who makes extensive use of technical abbreviations and grammatical terminology. In some parts of the manuscript we are led to believe there is the work of a less educated writer or a copier – it is not known which. There are spelling errors which reveal a seseo6 (çera, for será [‗it will be‘], asse, for hace [‗he/she/it does‘], Anon. 1601: 36) which is not unusual, as this pronunciation is part of Philippine Spanish, yet there is also some confusion between /e/ and /i/ (denero, for dinero [‗money‘] Anon. 1601: 116; pidi for pedí [‗I asked‘] Anon. 1601: 39), an error which is only likely to occur among indigenous people. It is even possible that this is a dictated copy, as certainly is the case of other Philippine grammatical manuscripts (Ridruejo 2006), as there are identical errors, for example, the incorrect interpretation of the word Dios [‗God‘] in the term gerundios [‗gerunds‘] (Anon. 1601: 114).

3. Language The manuscript contains a grammar of the Zambal or Sambal language (both spellings are possible, given the general seseo found in the Philippines). While on the front cover it is called lengua sambala, in the text there is a distinction between zambal, when referring to the language in Spanish, and sambal, when referring to it in the language itself (Anon. 1601: 118).7 5

6

7

According to Rodríguez Rodríguez & Álvarez Fernández (1992 II: 73–74) there was no person named Francisco Martínez Ramos but historians have mixed up two friars both named Francisco Martínez and attributed biographical data from both to a single fictitious individual. One of the friars professed in Segovia in 1563 and was Prior in Taguesi in 1593, the last date we have heard of him. The other friar, who is supposed to be the one in Bolinao, is believed to have died in 1608 or 1609 in Manila. Seseo is the term used to refer to the pronunciation, today occurring in Latin America and some regions of Spain, of [θ] as [s]. It seems unlikely that this difference occurs because the name of the language itself and the people who spoke it were linked among the Spaniards with the term zambo, which refers to a racial group, a mix of Blacks and Indians, as documentation on zambo in this sense can only be found in the CORDE (Corpus Diacrónico del Español – Diachronic Corpus of Written Spanish) after the 18th century, i.e., much later than its use by the people of Luzon.

14

Emilio Ridruejo

This language, considered as one of the lesser Philippine tongues (Constantino 1971) was spoken – and continues to be spoken – in the centre of Luzon, in the Zambales and Pangasinan provinces. Today three varieties of the language are known: Bolinao, Botolan in the Zambales province, and Tina, also in Zambales, in the north of the province, on the border with Pangasinan. The Sambalic languages are closely related to the Ayta languages (Stone 2006), and have been shown to be genetically related to Kapampangan in the Central Luzon microgroup of Philippine languages (Blust 1991). The grammar in question clearly reflects the Bolinao dialect of zambal. The manuscript contains numerous allusions to the language of this town (Anon. 1601: 8v, 19v, 28, 36v, 77) and the forms it covers are those described today in the variant of this area. Examples of this are the forms of the 1st person exclusive plural pronoun (sa)kami and the 2nd person plural pronoun (sa)kamo.

4. Content of the manuscript The manuscript is intended to be a grammar of the Sambal language. This is not only stated in the 1780 title, but also in a heading found immediately before the Vocabulary section: ―Finis Artis Bolinao. Deo gratias‖ (Anon. 1601: 139v). However, given that it is a compiled volume, it also contains other parts, the most important of which is the vocabulary, although the page numeration is consistent throughout the book, with the exception of the gap previously mentioned. The work starts with a prologue, ―Al religioso lector [To the religious reader]‖, written in the same hand as the colophon, where there is no interesting detail about the author or the grammar itself, but merely a topical encouragement for apprentices in their studies. After this, the grammar itself begins, the first pages of which are also written in the same hand as the prologue. However this fragment ends on page 2v, without reaching the bottom of the page, with the words: ―no obstante lo dicho, algunos nombres ay determinados a un sexo, cuya copia está en el Arte grande l[ibro]1, nº 1 [notwithstanding the above, some nouns are of a definite gender, a copy of which is in the large Grammar, book 1, nº1]‖. This reference to the Arte grande suggests that in 1780 there was a more detailed work on the Sambal language in existence at the Santiago Monastery in Bolinao. It might have been the 420-page work referred to by Fr. Castro in his Osario. Page 3 repeats the content of the previous page, regarding the lack of noun gender in the Sambal language, and it would seem that the author of the colophon and the front cover, Fr. Castro, may have found these first pages missing, and copied these first two pages from another work. Furthermore, on page 3, immediately after dealing with gender, there is an epigraph, ―Cap 2 de los Appelativos [Chap. 2 on Appellatives]‖, written in the same hand as the majority of the manuscript, which is crossed out. There are also other gaps and omissions. For example, on page 6 we find the beginning of a descrip-

The first grammar of a Philippine language?

15

tion of pronouns, yet the first person singular is missing, which must have been on one of the previous lost pages. After the grammatical sections, on page 140, we find the start of the ―Bocabulario de los demás adverbios por las letras A, b, c [Vocabulary of the remaining adverbs under letters A, b, c]‖. This claims to be a continuation of the Book devoted to adverbs, yet it includes many other lexical items, nouns and verbs. It is therefore a vocabulary of the Sambal language in alphabetical order. This vocabulary extends until page 160, at the beginning of the letter V, where it is cut short. The remainder of the manuscript includes various elements which seem to have been added upon binding, perhaps in an arbitrary fashion. Page 161 (in very poor condition) deals with pronunciation, and may be the remains of a longer section on this topic. The following page, which is of considerable theoretical interest, deals with translation difficulties, although it does not talk specifically about the Sambal language. Finally, the manuscript ends with a ―Capítulo único de la Poesía desta lengua [a chapter on the Poetry of this language]‖ which indeed does tend to appear in this end position in other grammars. Here there is a poetic measure and a number of examples of Sambal poetry.

5. Organisation of the grammar The strictly grammatical part of the manuscript consists of seven sections, each one divided into notes and chapters. The author, or the compiler, has tried to present the material in the same way as in Latin grammars, and has dealt with types of words, or parts of speech. According to this pattern, the first section deals with nouns, pronouns and relatives, partitives, ordinals and distributives. There is no second section. The third section, which is very short, deals with participles; the fourth, with prepositions or particles; the fifth, with verbs; the sixth, with adverbs; the seventh, with interjections, and the eighth with conjunctions. Apparently, therefore, the structure is the same as that found in the Analogy or Morphology section of a Latin grammar and in many other grammatical works on European, American or Philippine languages. However, in Philippine grammars it was difficult to follow this pattern. In Philippine languages, words consist of a root or theme, plus a wide range of affixes (prefixes, infixes, suffixes), with many different functions, many more than the inflectional morphemes in Indo-European languages. This made it impossible to fit grammatical material into a model with nominal and above all, verbal, paradigms. Grammarians resolved this difficulty by establishing different categories in the structure of the work. Firstly, particles whose function can be compared with the inflectional morphemes under the paradigms of traditional Greek and Latin grammars, and secondly, those which add meanings not considered to be part of the Classical tradition or which correspond to derivational affixes. In this way, Blancas (1560–1614) and, in a more eclectic manner, López (1617), divide their grammars (Blancas 1997 [1610]; López 1617) into two parts, one which reflects tradi-

16

Emilio Ridruejo

tional grammars covering the variation of different kinds of words, and another which deals with affixes and the results of combining them with different roots. In the Arte pampanga (1621) by Coronel (d. 1630), there is also an initial section which deals with parts of speech, followed by chapters devoted to the ―explicación de las preposiciones y verbos compuestos [explanation of prepositions and compound verbs]‖, which describes particles in alphabetical order. Given this division of paradigms and particles, in the Bolinao grammar there is an attempt to include particles into the pattern of parts of speech. There is only a very short fourth section, dealing with prepositions or particles, only one paragraph long, where the author limits himself to saying that: Las preposiciones o particulas son de quatro maneras. Vnas son de nombres y pronombres, de que ya hemos tratado. Otras son las de los verbales, de las quales trataremos jonto con las composisiones o particulas verbos actiuos. Otras son las de los verbos pasibos de que trataremos despues de los verbos actiuos [There are four types of prepositions or particles. Some are nouns and pronouns, which we have already seen. Others are verbal, which we will see together with the compositions of particles of active verbs. Others are those of passive verbs, which we will examine after active verbs] (Anon. 1601: 35v). In this way, particles are described when dealing with adjectives in chapter 13 of the first section: ―de ordinario los adjetiuos se forman con ma, vg. maabiaco, yo soy bueno, maabigca, tu eres bueno... [adjectives are normally formed with ma, e.g. maabiaco, I am good, maabigca, you are good…]‖ (Anon. 1601: 27), and the function of certain particles, ca, nica, na dian, manca, etc. is described in the formation of comparatives, diminutives, superlatives, etc. It is, however, in the fifth section, on verbs, where different chapters, from 3 to 12, are devoted to the particles ca, mo, mag, magsi, ma, mang, mi, qui, maqui, maca, mama, mameca, pa y mipa. After these, the passive forms are described, in between two other chapters devoted to pa and ypa. Another particularity of the Bolinao grammar is found in the position of the section describing the Sambal alphabet, letters and sounds. In Philippine grammars, this content was placed in one of two positions. Sometimes it appears at the beginning of the respective works, providing the reader with an instrument for reading the texts, as some special letters are used, such as the hatted . In other cases – and most frequently – the description of letters and sounds comes at the end, following the order already seen in Nebrija‘s (c.1444–1522) Institutiones latinae. This is the case of Blancas‘ (1560– 1614) work, among others. In the Sambal Grammar, however, the description of the indigenous alphabet, letters and sounds is placed in a note before the fifth section, dealing with verbs. This position is so unusual that one of the later correctors or commentators felt the need to add an observation on page 1v, indicating where this content could be found. There is only one

The first grammar of a Philippine language?

17

possible explanation for this fact, and that is that the fifth section on verbs at some point was independent and separate from the rest. Finally, there is another part of Philippine missionary grammar, which is not as common as in other missionary traditions and consists of a considerably important section devoted to the lexicon and expressions referring to trade activities. The descriptive vocabulary of purchasing, sales, payment, contracts, measurements, etc. is extensively represented in the grammars of Philippines languages from the first works written until those published in the 19th century, either for their eminently practical purpose or for technical reasons.8 The Bolinao Grammar fits this tradition and provides an extensive description of this terminology immediately after the numerals (with which it can clearly be associated) (Anon. 1601: 23v), although one of the authors refers to the Vocabulary shortly afterwards (Anon. 1601: 24v).

6. Descriptive resources Despite the fact that the Sambal Grammar differs in structure from other Philippine grammars, the same descriptive resources are used as in this grammatical tradition. In other words, the author(s) resolves the main descriptive problems in exactly the same way as those dealing with Tagalog, Visayan, Ilocano or Pampango and, in addition, they use the same technical terminology, which in many cases is specific to this grammatical tradition. An initial problem facing the missionaries describing Philippine languages is the existence in Tagalog, Ilocano, Pampango, etc., and also in Sambal, of certain functional elements through which a syntactic relationship is established between the nucleus and its modifier, i.e., between a noun with a group of roots which can be nominal, verbal, adjectival or even adverbial, and can be modified (Schachter & Otanes 1972: 118; 120–131). These particles are used to introduce modifications to the noun through one or more adjectives, but they must also be used in almost any extension of a nominal or verbal nucleus (Himmelmann 2005a: 161; 2005b: 359). When a noun is modified using a verbal root, these carry out a function very similar to those of relatives in Latin or Spanish. In the very first works, which still survive, these elements are recognised as a specific grammatical category which, perhaps in line with a tradition tracing back to Aztec grammar, is known as ligatura or ligadura [ligature] (Ridruejo 2005). In the Sambal Grammar these functional components are also described without any alternations. Here they are not given any particular section in the grammar, but they are referred to on several occasions. When describing forms of third person pronouns (Anon. 1601: 3v) 8

The type of relationship between relative predicates and commercial activities (comprar [buy] vs. vender [sell], pagar [pay] vs. cobrar [charge], etc.) made them ideal for exemplifying the function of the different ―voices‖, a central descriptive resource in the Philippine grammatical tradition.

18

Emilio Ridruejo

the author indicates that the y in sigay and the n in sinan are ligatures. Later, he specifies their morphological variation: La n y la y se siguen a vocal y quando precedio n se suele conuertir en y, pero si se acabo en otra letra consonante se pone a, Vg. si lucas anambyon parri, nacoya saybali, lucas el que pilo el arroz fue a manila… También suelen no yr solas estas letras a, n, y, sino que van acompañadas con la consonante que pre[ce]dio multiplicandola, y es muy usado en la lengua de Bolinao. Vg. pacanennamabitil, donde en na auia de ser a… [The n and the y follow a vowel and when an n goes before it normally becomes a y, but if it ends in another consonant an a is used, e.g. si lucas anambyon parri, nacoya saybali, Lucas who piled up the rice went to Manila… Also these letters a, n, and y do not normally appear alone, but are used with the consonant which went before, multiplying it, and it is used commonly in the language of Bolinao. E.g. Pacanennamabitil where en na would have been a...] (Anon. 1601: 8v–9) As a ligature is used in order for a verbal root to modify a nominal root, the author indicates that ligatures are the equivalent of the relative, but he also compares them with the Latin conjunction ut, perhaps insofar as this conjunction is used to introduce noun subordinates which modify a verb: Advierte que el relativo a quien corresponde ut se haze con una de las tres letras n, a, y y también con la partícula na (Anónimo 1601, 9). Suple el romance que y a qui, quid, quod y el sum, es fuy y el significado ser tal y sirven de aposición que en lengua latina es estar los sustantiuos juntos en un nmismo caso... aqui es si Pedro, asi sin ligadura, pero otras veces se junta algunas destas: si Francisco i capitan (Anónimo 1601: 12) [Note that the relative which corresponds to ut is made with one of the three letters n, a, y and also the particle na (Anon. 1601: 9). It replaces the Spanish que [what] and qui quid, quod and sum, it is fui [I was] and ―to be‖ just as in the apposition in Latin when two nouns are placed together in the same case…here it is si Pedro, without ligature, but on other occasions some of these are placed together: si Francisco i capitan] (Anon. 1601: 12) It is also suggested in the Sambal Grammar that the ligature acts in a similar fashion to the Spanish preposition de [de] in order to modify a noun: en cosas de peso y medida y en las que son materia de que se haze algo suplen el romançe de. Vg. sanchinan tan atilin, vna chinanta de zera [in weights and measures and in materials used to make something they replace the Spanish de. E.g. sanchinan tan atilin a chinanta9 of wax (Anon. 1601: 12v). This works in a similar fashion for cardinal numbers. 9

A unit of measurement used in the era in the Philippines, equivalent to 6.326 kg (approx. 14 lbs).

The first grammar of a Philippine language?

19

While the notion of ligature is introduced by Philippine grammarians in order to describe a specific category for the languages they study, in other cases they must deal with grammatical categories which are nothing new to the grammar of the languages they know, but they do so using different terminology. The expression of causatives is common in many languages, including, for example, Hebrew among the classical languages studied by the clerics, and also in some Mexican languages. This category can also be found in Philippine languages, where there are forms such as pa, in Ilocano (Lopez 1928: 113; Rubino 2005: 342–343) and pa-…-an, ipag in Tagalog (Schachter & Otanes 1972: 327–329) which are used to denote indirect action or cause (Himmelmann 2005b: 342). The missionary authors of Grammars use the term facere facere for these constructions, for which the generalised abbreviation is ff. Furthermore, some grammarians, in order to justify this abbreviation, which is without a doubt strange to the Classical grammar of Greek and Latin, allude to a tradition dating back to the first Philippine grammarians: Esta regla de Facere Facere, que este nombre inventaron los primeros Padres Ministros de muy loable memoria, que primeros que todos trabajaron en la lengua, para denotar que la persona que hace, ya en nominativo en la activa, ya en genitivo en la pasiva, no hace la acción signficada por la tal raíz, sino que la manda hacer a otro... [This rule of Facere Facere, was the name invented by the first Fathers of praiseworthy memory, the first to work in the language, to denote that the person or subject, both in nominative in the active, and in genitive in the passive voice, does not carry out the action in itself, but causes another to do it...] (Blancas 1997 [1610]: 80) In the Bolinao Grammar this same terminology can be found. What is more, our author finds it so common that, unlike other grammarians, he does not bother to justify or explain its meaning, but seems to consider it something well known among the potential readers of the text. In Chapter 5 of the fifth section on verbs, the author describes the particle mama: mama es 1º ff desta lengua, en actiua antepuesta a las raizes significa mandar hazer aquello que significa la raiz a quien se junta, a las raizes de um se antepone sola, a las demas se le puede posponer los verbales a cada vna, vg mamasurrat mandar escriuir, mama pagsenita, manda hablar pretérito Namcasurrat namapag sanita... [mama is the 1st ff of this language, in the active voice preceding the roots it means to order someone else to do whatever the root means, it can be placed alone before the um roots, in the case of others the verb can be placed behind each one, e.g. mamasurrat to command to write, mama pagsenita to be sent to speak past tense Namcasurrat namapag sanita…] (Anon. 1601: 94v).

20

Emilio Ridruejo

The constructions referred to in the Sambal Grammar as ff seem to be few, and it only refers to those which mean ordering the process indicated in the root to be done. In this way, other causative combinations are excluded from this classification of ff: Algunas vezes [mama] no significa ff sino permitir o dexar que se haga aquello o causarlo postivamente si esso cave en aquel de quien se habla, vg. mamopalav ay dexar amaneçer, mama sarrita dexar hablar, mamaliti embiar rayos… [Sometimes [mama] does not mean ff, but to permit or let something to be done or positively cause it to happen if it is possible on the part of the speaker, e.g. mamopalav ay to let (someone) wake up, mama sarrita to let (sb) speak, mamaliti to send rays… (Anon. 1601: 95). Also, later in the work, when describing the combinations of the form ma with others, the author specifies that placing pa after ma is ff: mapi significa acçion de humildad y sujecion Vg. mapimisa oydor de misas asi frequentemente parece que es contrario de ff etc. como mapa, pero estos principalmente van a poder y a frequentacion. [mapi means action of humility and subjection e.g. mapimisa hearer of Mass, thus it often seems to be the opposite of ff etc. like mapa, but these mainly refer to power and attendance] The author of the Bolinao Grammar provides other example of pama: [llos compuestos de pama] muchas vezes no tienen esta signficaçion de ff. Vg. pamaliyan modo que tienen de vivir los de una cossa, pamatorriran sarrita vel pamorriran con que endereçar el habla [often does not have this meaning of ff. e.g. pamaliyan the way of life of things, pamatorriran sarrita vel pamorriran to harden the speech (Anon. 1601: 95v). Finally, on p. 113, in a fragment written in another hand containing a description of particles it is explained: Mipa, mapi, mapa todas tres son lo mismo, solo se distinguen por las raizes que se anteponen que los piden assi, son frequentativos y son ff quando la açion sale de la persona que manda, de suerte que son ff de açion, a diferencia de pa que es de persona y de paque es de cossa [Mipa, mapi, mapa, all three are the same, they are only distinguished by the roots which go before that require them to be like this, they are frequentative and ff when the action comes from the person who is ordering, so they are ff of action, unlike pa that refers to a person and paque which refers to things] (Anon. 1601: 113).

The first grammar of a Philippine language?

21

7. The verbal paradigm The first chapter of the fifth section on verbs has the title ―On the substantive verb sum, es, fuy‖. As is commonplace in missionary grammars, here the copular verb forms are presented, or, if this does not exist, alternatives are given for expressing copula, identity or existence. In this case, the author indicated that in Sambal the copular verb is included in nouns and pronouns and only there, rejecting all other parts of the sentence: Vna sentencia de que no ay sum, es, fuy sino que esta embeuido en los nombres y pronombres de qualquier modo que sean y no en otra parte ninguna de la oracion, porque el verbo significa acion o pacion, el aduerbio significa el modo de la açion, la pripossi[ci]on no tiene mas que servier como reuete o flueco de algunas partes de la oracion, pues por si sola no puede estar; la interjeçion no haze mas que notar los diuersos afictos del animo, la conjunçion serue de atar o trauar las demas partes de la oracion o las oraciones diuersas. Y asi solo queda el nombre y pronombre donde puede estar embeuida la significaçion ser, pues no ay palabra que siempre absolutamente serua de ser, es fuy ni el partecipio incluye esta significacçion, porque rigurosamente hablando no ay mas que el del primitiuo que uiene es lo mismo que nombre adjectiuo [A sentence where there is no sum, es, fuy, and instead it is contained in the nouns and pronouns of any type and not in any other part of the sentence because the verb is active or passive, the adverb describes the action, the preposition only has to act as an addition to some parts of the sentence, as it cannot stand alone; the interjection only denotes the different effects of mood, the conjunction serves to join different parts of the sentence or link the sentence with other sentences. And so there only remains the noun and pronoun to hold the meaning of ser (‗to be‘), as there is no word that always means ser, es, fuy, nor does the participle contain this meaning because, strictly speaking, there is nothing else but the primitive one, which is the same as the adjective noun] (Anon. 1601: 40; Rubino 2005: 40v ) This same argument, formulated in almost identical terms, can be found in the López Grammar on Ilocano: No ay palabra particular que corresponda al verbo (sum, es fui) en quanto significa (seer). Lo segundo digo (para abreviar) que absolutamente está embebido o subintellecto en los Nombres i Pronombres de qualquier genero i modo que sean, i no otra parte ninguna de la oracion. Dos fundamentos ay para prueva de esta mi Proposicion. El primero es la diferencia de significar que cada parte de la Oracion tiene. I el otro, es los exemplos sin numero (que son la tercera parte, por lo menos, de todo quanto en la lengua se puede hablar (hablando a buen ojo, con los quales, por Ascenso se colige la prueva de lo dicho. El primer fundamento es claro. Porque el verbo significa accion o pasion.

22

Emilio Ridruejo

El Adverbio significa el modo de la accion. La Preposicion no tiene mas que servir como ribete o flueco de algunas partes de la oracion (pues por si sola no puede estar). La Interjeccion no hace mas que denotar los diversos afectos del animo. La Conjuncion sirve de atar o travar las demas partes de la oracion o las oraciones diversas. De suerte que solo queda el Nombre o Pronombre (que todo es una cosa, quanto a esto), pues no ay palabra particular que la tenga de suyo como el verbo (sum, es, fui) [There is no particular word for the verb (sum, es fui) in terms of meaning (ser). The second thing I say (to be brief) is that it is certainly contained or understood in the nouns and pronouns of any gender and form, and not in any other part of the sentence. There are two pieces of evidence to demonstrate my theory. The first is the difference in meaning of each part of the sentence, and the other is the numerous examples (which form a third, at least, of everything which can be said about the language – speaking as an estimate in order to increase the strength of the argument. The first fundamental is clear. Because the verb is active or passive. The Adverb describes the action. The Preposition only serves as an addition to some parts of the sentence (as it cannot stand alone). The Interjection only denotes the different effects of mood. The Conjunction serves to join or link the other parts of the sentence or different sentences. This means that there only remains the Noun or Pronoun (which for this purpose is the same thing), as there is no particular word which acts as the verb (sum, es, fui)]. (López 1617: 63–64). From the above texts it can be concluded that there is a connection between the Sambal Grammar and the Ilocano Grammar by Francisco López, although this does not help to deduce which came first, as it could even be that both used the same previous source. In the Sambal Grammar, this is followed by an indication that the verb sum has ―no more than 3 tenses, present, preterite and future, and the others are formed by circumlocution‖ (Anon. 1601: 40v) and examples are given of pronouns, alone or with other nouns, to which a translation is given attributing the verb ser (to be) in different persons and tenses, given that the mere juxtaposition of a subject and predicate gives rise to an attributive structure (Schachter & Otanes 1972: 62).10 The second observation in the grammatical works of the missionaries is that there are different elements which can be used to replace the verb sum. One suggestion is the possibility of using a which ―replaces the meaning of ser and also this particle mag: marunung a tauo good man‖ (Anon. 1601: 41). The unique feature of the Sambal Grammar is that, among other possible replacements, one that allows ereng is indicated:

10

In reality, in Philippine languages there is no clear difference between equative and narrative structures, as there is no distinction between noun and verb roots (Himmelmann 2005a: 141).

The first grammar of a Philippine language?

23

que es sum es fuy en quanto a todas significaciones y solo se distingue en que para sum no es necessario poner ego, para aqui si es necesario el pronombre por las palabras invariables, pero podemos deçir que en otros casos excede a sum es fui [which means sum es fuy for all meanings and is only differentiated in that for sum it is not necessary to use ego, here it is necessary to use the pronouns for invariable words, but we can say that in other cases this exceeds sum es fui] (Anon. 1601: 41v).11 The author makes use of this form ereng in order to introduce the full paradigm of the conjugation, ―for all tenses, and also serving as a rule hereafter. With this we do not conjugate actives and passives‖ (Anon. 1601: 42). The paradigm of tenses and moods presented here contradicts what was previously stated about there only being three tenses, and that the rest are created using circumlocutions, a contradiction which would be adequately explained if this were indeed the sum of fragments of grammars from different sources. Presente actual [present]: manereng aco, yo estoy siendo [I am being] Presente ymperfecto y perfecto [present imperfect and perfect]: sin manereng aco, quando estaba o estuue siendo [when I was being] Presente ordinario [ordinary present]: ereng aco, vel sicoy ereng, yo soy vel estoy siendo [I am, I am being] etc. Preterito imperficto y perficto [past imperfect and perfect]: sin ereng aco vel sinsicoy ereng vel erengaco etc., yo era vel fuy [I was] Preterito plusquamperfecto [pluperfect]: nayari acoyna in un deng fiscal vel nag fiscal, ya yo auia sido fiscal [I had already been a public prosecutor] Futuro imperfecto [future imperfect]: Ereng a coly, yo sere, vel si coy erengly vel vndeng a coly fiscal, yo sere fiscal [I will be public prosecutor] Futuro Perfecto [future perfect]: Nomayari a coly vndeng fiscal vel mag fiscal payanauaco, quando aure acabado de ser fiscal [When I will have stopped being...] Imperatiuo afirmatiuo [affirmative imperative]: Vndengca , se tu [be (you)] Imperativo prohibitivo [prohibitive imperative]: ande ca vmdeng no seas o no tengas ser o ayas ser [don‘t be] In Sambal three present tenses are proposed: present, present imperfect and perfect and ordinary present; a past imperfect and perfect and pluperfect. There are two future tenses, imperfect and perfect. 11

There is a note written by another hand: ―es el mayor disparate que se puede imaginar y discurro que ni en el mundo ay lengua a quien sirva de sum, es, fui ereng [it is the greatest folly one can imagine and I reflect that there is no language in the world for which sum, es, fui ereng can be of use]‖ (Anon. 1601: 42).

24

Emilio Ridruejo

In regards to mood, two imperatives are described, one in the affirmative and another as a prohibitive; and an optative, a subjunctive in the present, imperfect, past perfect, pluperfect tenses and a future are introduced. In the infinitive there is also a present, a pluperfect and a future. Furthermore, three gerunds are added (ending in -di, -do, and -dum), in line with the Latin forms and, finally, something particularly interesting, a form which seems to denote aspect in the egressive phase and which is described as ―acabadito‖:12 ―Capiereng, acabadito de ser; capopalauay, acabadito de amaneçer [Capiereng, acabadito of to be; capopalauay, acabadito of to dawn]‖ (Anon. 1601: 46v). The complexity of the proposed paradigm is greater than what is usually found in printed Philippine grammars and is only comparable with that of López (1617), who also introduces a complex conjugation model with two present tenses, several future tenses and imperative and prohibitive moods. In all likelihood, the author is attempting to use the multiplication of present and past tenses in order to reflect the difference in meaning, which he has identified and today we would call aspects. Indeed in Philippine languages, for example in Tagalog (Himmelmann 2005b: 363–364) or in Ilocano (Rubino 2005: 338) the category of grammatical aspect is functional and can be expressed both inflectionally and derivationally. However, the Sambal Grammar does not manage to break away from the tense distinctions of the Latin verb, which should, nevertheless, not be applied to the language in question.

8. Voices In the first grammars of Philippine languages, by Blancas (1997 [1610]), López (1617), Méntrida (1559?–1637) (1818 [1618]), Coronel (1621), etc., and those that followed, the grammatical model used has a central core consisting of verb descriptions, with specifications about tense and person and, above all, using the most important component, the voices, in particular the passive voice. Each of these passives, with different morphological marking, plays a different semantic role attributed to the noun in question.13 With these, grammarians do not try to reflect, as in Latin, variations in diathesis, but they want to report the fact that in Philippine languages, there are different affixes for verbs, whose function is to specify the semantic role (agent, beneficiary, instrument, locative, etc.) of the argument represented by the noun used as a topic.14 By associating certain morphological marking with each of these supposed passives, grammarians are indicating that the thematic noun, considered as the subject, is not an 12 13

14

See the use in Spanish of acabar de to express the meaning ‗to have just done something‘. This noun is commonly known as subject, even if it has peculiar features different from subject in the Indo-European languages (Himmelmann 2005a: 155–6). In the sentence, in addition to the topic, there can be nominal phrases which are nontopic and which can be marked with case morphemes.

The first grammar of a Philippine language?

25

agent but plays a different semantic role (as happens with the passive in Latin). This descriptive procedure is also admitted in the Sambal Grammar, although with certain peculiarities which distinguish it from the Philippine tradition. The function of voices is introduced in the notes at the beginning of the section devoted to verbs and before dealing with verb categories. Initially, there is an observation that is presented in all grammars, which explains that the supposed passives appear only when there is a specific noun reference, as the specification plays a key role in the distribution of these markers (Himmelmann 2005a: 145–147). The Sambal Grammar starts out by establishing the different verb functions depending on whether they refer to ―cossa indiferente, indeterminada o en general [indifferent, non-determined or general things]‖ or ―quando la oraçion es de cossa determinada en particular o que es sola en aquella especie [when the sentence is about something in particular, or is one of a unique kind]‖. In the first case, which is described in some detail, the sentence is taken as active, ―aunque a vezes no estara mal de otras por passiva [although sometimes it may be passive]‖ (Anon. 1601: 38). For the second case ―se hablara por pasiba (y nota que es mas elegante las mas de las vezes hablar por passiva que por active) [it will be spoken in the passive (and note that it is more elegant to speak more in the passive than in the active)]‖. Later, in note six of the same section, the author states that there are ―tres modos de significar de los verbos: acción transitiva, intransitiva o neutra y pasibamente transitiva [three meanings for verbs: transitive, intransitive or neuter and passively transitive action]‖ (Anon. 1601: 38). This classification is rather strange, as it combines the genera verbi of Latin grammar with the types of construction, as can be found, for example, in the Nebrija Grammar,15 yet neuter verbs are identified as intransitives. (Anon. 1601: 39). However, it is from chapter 13 of the fifth section ―on passives and the first affix en‖, where the author explains the use of the three affixes that make up the supposed passives in Sambal: en, y, an. Explicare mas en. La açion que mira primario alguna cossa es por en, la que mira secundario, que viene ser lugar es por an y el con que por y. Vg. para hazer vna cossa ha de auer con que la hazer, y esso es por y, y quien la haga, y si mi intento va a quien la haga que es por na es por en y ha de auer adonde se haga y si ua ay mi intento sera por an [I will explain en some more. The action which first contemplates something is with en, if it is secondary, it is normally with an and the ―with‖ concept is with y, e.g. in order to do something there must be something to do it with, and that is with y, and if my intention is for someone to do something, it is with na, and with en and there must be somewhere to do it and if there is ay my intention will be with an (Anon. 1601: 97v).

15

―Quottuplex est construction? Duplex, Transitive, Intransitive‖ (Nebrija 1996 [1488]: 127).

26

Emilio Ridruejo

The difference between the Sambal Grammar and other Philippine grammars consists fundamentally in that it considers that the meaning of the verb is given by the affix and not, as stated in other works, that each affix is added to the verbal roots according to the precise meaning of the verb and, consequently, there is a search for specific semantic classes of verbs that can be linked to each of the passives, i.e. with each functional affix. La partícula en se pospone a las raizes y significa ser hecha la cossa ad intra como atrayendolo assi y como indeterminante, a difirencia de y, que significa ser hecha la cossa ad extra como echándolo hazia afuera y determinante y a diferencia de an que significa lugar y cosa determinada y la tal cossa de en, aunque sea persona se ha como cossa y pone en nominatiuo ... [The affix en is placed after the root and means that something has been done ad intra as if extracting it as an indeterminate, unlike y which means the thing has been done ad extra in an outward and determinate fashion and unlike an which denotes a certain place and thing and that thing is an, even if it refers to a person doing something and is placed in the nominative…] (Anon. 1601: 96) Although the connection between each of the passives and the meaning of the verb is already established, there is still not even a minimum attempt at a semantic classification of verbal roots, in order to create rules for the origin of each of the passives, as Méntrida did very successfully in 1618 (Méntrida 1818 [1618]: 60v), as this model spread rapidly. The meaning of each affix is emphasised in the following chapters, 14 and 15, when dealing with an and y: Las raizes que de suyo tienen la cossa, con an pospuesto significan quitarselo, y las que no la tienen seno ab extrinseco significa ponerselo Vg. Bonoton, quitar la estopa al coco, quisquisan quitar el cauello vel oldogan, frotalan, poner frontal al altar [The roots which use the an suffix mean ―remove it‖, and those which only have it ab extinseco mean ―place it‖, e.g. Bonoton, to remove the husk from a coconut, quisquisan to remove the hair from the oldogan, frotalan to place a frontispiece at the altar] (Anon. 1601: 100–100v). Con raizes que significan miembro corporal significa tener grande el tal miembro. Vg. ycoyan, de grande cola, sarrauan, de grandes curenos etc. [With roots which mean a part of the body it means that the part of the body is large. E.g. ycoyan with a large tail, sarrauan with large horns etc.‖(Anon. 1601: 100v). Although the author of the Sambal Grammar describes combinations with an, en and y as passives, these combinations are actually some of those which can establish each verbal root, and can also be combined with other groups of affixes. For this reason, the fifth section continues to cover the meaning of other affixes and the meanings resulting from their combinations. The description becomes extremely complex, and at the same time impre-

The first grammar of a Philippine language?

27

cise, as the grammarian, who does analyse the affix combinations correctly, does not, however make generalizations about their functions but rather attempts to describe the multiple meanings produced by each affix in combination with the different roots.

9. Adverbs In the organisation of the grammar in line with the sequence of parts of speech, we find section 6 on adverbs (Anon. 1601: 119v). This section precedes the one dealing with interjections and conjunctions. The opening presents a general outline. It is stated that adverbs are not necessary in speech and, above all, they are defined in a wider sense than that seen in the classic tradition of Greek and Latin. Considering that adverbs are forms which, when blended with the verbal root, alter its meaning, the author treats a great many particles as adverbs. Los mas de los adverbios que significan modo de acion no los usan en su propia forma de adverbio como fortiter...sino que los hazen verbo desta manera açota blandamente a esse niño planapalnaenmay panĝanĝanla [Most adverbs, which describe a means of action, are not used in the form of an adverb such as fortiter…but they make a verb, thus ‗he softly hit that boy‘ planapalnaenmay panĝanĝanla (Anon. 1601: 119v). Although the establishing of classes of adverbs, given the great variation, tends to be an excellent means of identifying grammar sources, in the Sambal Grammar, the author establishes such a large list that it far surpasses the extent of Latin or Philippine grammars. This list coincides with some of the traditional classes (adverbs of place, time, negation and similarity, etc.), but there are many other types which are established according to the function of the sentence in discourse: indicative adverbs, adverbs for honour, for prohibition, reprehension, quality, gratitude, calling, etc. Of all the Philippine grammars, only that of Francisco López on Ilocano seems to bear any resemblance which could confirm the existence of any connection. López (1617) also considers the most common classes of adverbs: place, time, similarity, doubt, etc., but he adds a large section with other adverbs with all elements which do not fall under the previous classification, adverbs for talking about general things, for increase in price, lamenting a loss, etc. The difference between López‘s work and the Sambal Grammar lies in the fact that the latter does not limit itself to covering all these particles in a single section, but instead tries to establish similarities to the traditional types among them.

28

Emilio Ridruejo

10. Sounds We have already mentioned the strange position in this manuscript of the description of letters, the Sambal alphabet, and pronunciation, as they appear as if they are an opening note which has subsequently been placed in medias res. It deals with the Latin letters used in the transcription of Sambal texts and, although the term ―letter‖ in most 16th and 17th century grammars can refer both to the written letter and the phonemes they represent, in this text the initial list shown is of the written letters, as it includes , , and , which both in Sambal and Spanish writing are used to represent the same phoneme. The same is true at least with and and even perhaps with and . The list also includes the unique letter , which must represent the nasal velar commonly found in Philippine works, and even in other parts of the Sambal Grammar, as the hatted . As is the case in other Philippine grammars the symbols of the indigenous alphabet are listed, which during the early years of colonisation must still have been in use. The symbols are included with their phonological equivalent and the name given to each one: ―la B la llaman baba y la escriuen assi O, la c llaman caca y la escriuen asi – Z... [the B is called baba and they write it like this , the c is called caca and they write it like this – Z…]‖ (Anon. 1601: 36) The range of Sambal phonemes is clarified in the observations found after some of the letters. As is common in missionary grammars, it is stated that certain letters ―are confused‖: ―la z se escriue como la s. La e es lo mismo que y. Y también usan indiferentemente o por u y de u por o [the z is written like an s. The e is the same as y. And o and u, and u and o are used interchangeably]‖ (Anon. 1601: 36). Furthermore, after the description of the phonic features of the alphabet, the author alludes to the pronunciation he calls ―cortadilla [shortened]‖. This term is used in a number of Philippine grammars to try to reflect the existence of something which the missionaries perceived clearly, but which was unknown to them in Spanish or in Latin. It is the glottal stop // which exists as a phoneme in Philippine languages and allows minimal pairs to be opposed (Himmelmann 2005a: 117). The author of the annotations to the fifth section does not proceed to describe this sound, but merely states that ―los vocablos que se pronuncian con cortadilla los enseñara el uso y vease el vocabulario [the words pronounced with cortadilla are learned through the use of the language, and seeing the vocabulary]‖ (Anon. 1601: 36v). And indeed, the vocabulary at the end of the manuscript has the peculiarity – not expressly formulated – (which leads us to believe that the paragraph that explains the conventions used is missing) of introducing a virgule, or forward slash, after the syllable where this sound is found.16 In addition to referring to the ―cortadilla‖ in the section devoted to sounds, an16

In other grammars, e.g. in Father Lopez‘s, it was marked by a hyphen: malag-an ‗cosa liviana‘ [light thing]. (López 1617: 5)

The first grammar of a Philippine language?

29

other page in the manuscript, p. 161, which comes after the vocabulary, includes further information on the pronunciation of this sound. The page starts with two lines including examples of a previous rule, which is missing, to which is added ―con el vso se sabran los demas que no ay aquí [with use the ones not included here will become clear]‖. There then follows examples of cortadilla and the following characterisation is provided: No consiste la cortadilla en parar entre silaba y silaba, sino que la consonante que esta entre las dos vocales vaya con la primera de suerte que en el punto que se pronunçia la tal vocal, se ha de pronunciar tambien la tal consonante, sin reparar en que yera alguna siguiente vocal al modo que se pronunçia la conjunçion /pues/ en la lengua castellana... [The cortadilla does not consist of stopping between two syllables, but the consonant which lies between the two vowels goes with the first so that when the vowel is pronounced, the consonant must also be pronounced, without concern for affecting the following vowel in the style that the conjunction pues is pronounced in Spanish...] (Anon. 1601: 161). The example given in Spanish is rather unfortunate,17 although the most accessible means of representing a sound like the glottal stop was to suggest something similar to an interruption in speech, marking a syllabic boundary, this interruption does not take place in the example with -s at the end of a word.18

11. Conclusions The manuscript on the Sambal language preserved in Valladolid groups together several fragments from different works with the intention of being unified. According to what 17

18

The lack of skill or knowledge of the author of the Sambalic Grammar is obvious if compared to the much neater description of Ilocano by López: Llévese también notado que en esta lengua como en la tagala, hay muchos vocablos que siendo una dicción sola, requiere para su natural propia y congrua pronunciación lo que en nuestro español requieren en muchas ocasiones dos palabras distintas para que no parezcan una sola y de muy diferente significado: v. gr. si estas dos palabras mal lino las pronunciamos de corrida parecerá que tratamos de algún espíritu maligno, no tratándose sino de lino malo. De manera que para que el que oye entienda que tratamos de lino malo, hemos forzosamente de cortar una palabra de otra, pronunciando cada una de por sí, v. gr. mal lino... [It should also be noted that in this language, as in Tagalog, there are many words which are pronounced as one, requiring for their own natural and congruous pronunciation what in Spanish often requires two distinct words in order for them not to seem as one word of a very different meaning: e.g. if the two words mal lino are pronounced together it would seem like we are talking about some kind of evil spirit (maligno = evil, malign in Spanish), rather than poor quality linen (in Spanish lino malo). In order for the listener to understand that we are talking about poor linen, we must purposely separate one word from the other, pronouncing each one individually, e.g. mal lino] (López 1617: 5). This would only have a satisfactory explanation if in the Spanish dialect spoken by the author there was pre-aspiration of the -s as seen today in Southern dialects.

30

Emilio Ridruejo

is stated in the work itself, on an unspecified date in the 17th century, there were a number of grammars on the Sambal language at the Augustinian monastery in Bolinao. These included, at least, a so-called Arte grand, a Vocabulary and perhaps other partially completed works. At a given time, several fragments from these works were grouped together and a compiled Grammar was attempted. This compendium received a prologue and a colophon in 1780, written by Fr. Agustín María de Castro. The result is the manuscript which we conserve today, an unusual work within those on Philippine grammar, even when compared with those which were not printed either. It is not the first grammar of a Philippine language. It falls within an existing tradition of Philippine grammars both in terms of the descriptive resources used, the passive voice or ligatures and by the terminology it uses: facere facere, cortadilla, etc. However, the author (or authors) does not cite any previously existing sources, despite the fact that it was customary to do so in Philippine grammars, even unpublished ones. The organisation and description differ in several ways from what is usually found in Philippine grammars: there is no clear separation between grammatical paradigms and particles, nor is there any distinction drawn between lexicon and grammar. All this reveals that it is a less detailed and more primitive work than the first printed grammars. It is therefore probable that some of the elements – the most systematic – constitute an early attempt to represent the grammar of the Sambal language, and that this took place in the first few years of the Spanish presence in the Bolinao region. The date, 1601, which appears on the front cover may not, therefore, be an impossible date, but neither is it certain, nor did the monk that wrote it justify this date as correct for an original manuscript or some of its parts.

Bibliography A. Primary sources Anon. (1601): Arte de La lengua Sambala y Española compuesto por un Religioso de la orden de Nuestro P. S. Agustín Ministro que ha sido por muchos años en ella; para ayuda de los Nuevos Operarios que vinieren a la Administración. Dedicado al gran Apostol de España Santiago. Año de 1601. Blancas, Francisco de S. Joseph (1997 [1610]): Arte y reglas de la lengua tagala. En el Partido de Bataan, por Thomas Pinpin Tagalo. Tercera edición, impreso en la imprenta nueva de don José María Dayot, por Tomás Oliva, Madrid, 1832. Edición facsímil de la edición princeps con Estudio de Antonio Quilis. Madrid: Ediciones de Cultura Hispánica, 1997 Coronel, Francisco (1621): Arte y reglas de la lengua pampanga, compuesto por el Padre Predicador – del Orden de N. P. S. Agustín y Prior Ministro del Convento de Macabele, manuscrito. [A lápiz] Acabado de copiar el 1677. Biblioteca de PP. Agustinos (Valladolid). 45 páginas de un volumen facticio con textos en tagalo López, Francisco (1617): Arte de la lengua ilocana. Manila, En el Colegio i Universidad de S. Thomas de Aquino, por Thomas Pipin.

The first grammar of a Philippine language?

31

Méntrida, Alonso de (1818 [1618]): Arte de la lengua bisaya hiliguayna [sic] de la isla de Panay, Manila, Oficina de los Jesuítas. Segunda edición, Manila: Imprenta de don Manuel Memije, por don Anastasio Gonzaga, 1818. Nebrija, Elio Antonio de (1996 [1488]): Introduciones latinas contrapuesto el romance al latín. (Materialien zur Geschichte der Sprachwissenschaft und der Semiotik, 7). Edición e introducción de Miguel Angel Esparza Vicente Calvo. Münster: Nodus.

B. Secondary sources Adelaar, Alexander & Himmelmann, Nikolaus P. (eds.) (2005): The Austronesian languages of Asia and Madagascar. London/New York: Routledge. Blust, Robert A. (1991): The Greater Central Philippines hypothesis, in: Oceanic Linguistics 30(2), 73–129. Calvo Pérez, Julio (ed.) (2001): V Jornadas Internacionales de Lengua y Culturas Amerindias. Contacto interlingüístico e intercultural en el mundo hispánico. Valencia: Instituto Valenciano de Lengua y Cultura amerindias, Universitat de València, 519–533. Castro, Agustín María de (1954 [1780]): Misioneros agustinos en el Extremo Oriente, 1561–1780 Obra inédita que con el título ―Osario venerable‖ compuso el Agustino P. Agustín María de Castro, edited by Manuel Merino, O. S. A. Madrid: C. S. I. C. Constantino, Ernesto (1971): Tagalog and other major languages of the Philippines, in: Sebeok, Thomas A. (ed.), Current trends in linguistics, 8. Linguistics in Oceania, I. The Hague: Mouton, 112–154. Corrales, Cristobal; Dorta Luis, J.; Corbella Díaz, D.; Torres González, A. N.; Plaza Picón, F. M.; Izquierdo Guzmán, L.; Martín Díaz, M. A.; Medina López, J. & Hernández, B. (eds.) (2004): Nuevas aportaciones a la historiografía lingüística. Madrid: Arco/Libros S.L., II. Hernández, Policarpo (1990): Los agustinos y la filología en Filipinas, in: Archivo Agustiniano 74, 199–214. Himmelmann, Nikolaus P. (2005a): The Austronesian languages of Asia and Madagascar: typological characteristics, in: Adelaar, Alexander & Himmelmann, Nikolaus P. (eds.), The Austronesian languages of Asia and Madagascar. London/New York: Routledge, 110–181. Himmelmann, Nikolaus P. (2005b): Tagalog, in: Adelaar, Alexander & Himmelmann, Nikolaus P. (eds.), The Austronesian languages of Asia and Madagascar. London/New York: Routledge, 350–376 Li, Charles N. (ed.) (1976): Subject and topic. San Francisco: Academic Press. Lopez, Cecilio (1928): Comparison of Tagalog and Iloko. Hamburg: J. J. Augustin. McKaughan, Howard (1971): Minor languages of the Philippines, in:Sebeok, Thomas A. (ed.), Current trends in linguistics, 8. Linguistics in Oceania, I. 155–167. Niederehe, Hans-Josef (1999): Bibliografía cronológica de la lingüística, la gramática y la lexicografía del español (BICRES). II, 1601–1700. Amsterdam/Philadelphia: Benjamins. Ridruejo, Emilio (2001): Las primeras descripciones gramaticales de lenguas filipinas, in: Calvo Pérez, Julio (ed.), V Jornadas Internacionales de Lengua y Culturas Amerindias. Contacto interlingüístico e intercultural en el mundo hispánico. Valencia: Instituto Valenciano de Lengua y Cultura amerindias, Universitat de València, 519–533. Ridruejo, Emilio (2004): Del ff: historia de una noción gramatical, in: Corrales Zumbado, Cristóbal et al. (eds.), Nuevas aportaciones a la historiografía lingüística. Madrid: Arco/Libros S.L., 1387–1398. Ridruejo, Emilio (2005): Las ligaturas en las gramáticas misioneras filipinas del siglo XVII, in: Zwartjes, Otto & Altman, Cristina (eds.), Missionary Linguistics II/ Lingüística Misionera II. Orthography and phonology. Amsterdam/Philadelphia: Benjamins, 227–245.

32

Emilio Ridruejo

Ridruejo, Emilio (2006): Una gramática desconocida de la lengua pampanga fechada en 1694, in: Roldán Pérez Antonio et al. (eds.), Caminos Actuales de la Historiografía Lingüística. Actas del V Congreso Internacional de la Sociedad Española de Historiografía Lingüística. Murcia, 8-10 de noviembre de 2005. Murcia: Universidad de Murcia, 1343–1360. Ridruejo, Emilio (2007): El problema de la descripción del ―sujeto‖ o tópico nominal en las primeras gramáticas filipinas, in: Zwartjes, Otto; James, Gregory & Ridruejo, Emilio (eds.), Missionary Linguistics III/ Lingüística Misionera III, Morphology and Syntax. Selected papers from the Third and Fourth International Conferences on Missionary Linguistics, Hong Kong/Macau, 12–15 March 2005, Valladolid, 8–11 March 2006. Amsterdam/New York: Benjamins, 325–340. Rodríguez Rodríguez, Isacio & Álvarez Fernández, Jesús (1992): Diccionario biográfico agustiniano. Provincia de Filipinas. 2 vols. Valladolid: Ediciones Estudio Agustiniano. Rodríguez Rodríguez, Isacio (1978): Historia de la provincia agustiniana del Santísimo nombre de Jesús de Filipinas, vol. XIV. Manila: Arnoldus Press. Roldán Pérez, Antonio (coord.) (2006): Caminos Actuales de la Historiografía Lingüística. Actas del V Congreso Internacional de la Sociedad Española de Historiografía Lingüística. Murcia, 8-10 de noviembre de 2005. Murcia: Universidad de Murcia, 2. Rubino, Carl (2005): Iloko, in: Adelaar, Alexander & Himmelmann, Nikolaus P. (eds.), The Austronesian languages of Asia and Madagascar. London/New York: Routledge, 326–349. Santiago Vela, Gregorio de (1913–1935): Ensayo de una Biblioteca Ibero-americana de la Orden de San Agustín. Obra basada en el Catálogo bio-biográfico agustiniano del P. Bonifacio Moral. Vol 1, 1913; Vol 5, 1920. Madrid: Imprenta del Asilo de Huérfanos del Sagrado Corazón de Jesús. Schachter, Paul & Otanes, Fe (1972): Tagalog reference grammar. Los Angeles: University of California Press. Schachter, Paul (1976): The subject in Philippine languages: topic, actor, actor-topic or none of the above?, in: Li, Charles N. (ed.), Subject and topic. San Francisco: Academic Press, 491–518. Sebeok, Thomas A. (ed.) (1971): Current trends in linguistics, 8. Linguistics in Oceania, I. The Hague: Mouton. Stone, Roger (2006): The Sambalic languages of Central Luzon, in: Tenth International Conference on Austronesian Linguistics. 17–20 January 2006. Linguistic Society of the Philippines and SIL International. Puerto Princesa City, Palawan, Philippines. Online website: http://www.sil.org/asia/philippines/ical/papers.html (August 2008). Sueiro Justel, Joaquín (1997): La lingüística colonial y misionera española: El Archivo de los PP. Agustinos Filipinos de Valladolid, in: Romanistik in Geschichte und Gegenwart 3, 213–240. Sueiro Justel, Joaquín (2003): Historia de la lingüística española en Filipinas (1580–1898). Lugo: Editorial Axac. Zwartjes, Otto & Altman, Cristina (eds.) (2005): Missionary linguistics II/Lingüística misionera II. Selected papers from the second international conference on missionary linguistics). Amsterdam/Philadelphia: Benjamins. Zwartjes, Otto; James, Gregory & Ridruejo, Emilio (eds.) (2007): Missionary Linguistics III/ Lingüística Misionera III, Morphology and Syntax. Selected papers from the Third and Fourth International Conferences on Missionary Linguistics, Hong Kong/Macau, 12–15 March 2005, Valladolid, 8–11 March 2006. Amsterdam/New York: Benjamins.

JOHN U. WOLFF (NEW YORK/USA)

The Vocabulario de Lengua Tagala of Fr. Pedro de San Buenaventura (1613)

Abstract The Vocabulario de Lengua Tagala by Fr. Pedro de San Buenaventura is the earliest Tagalog dictionary to have survived. It is a remarkably full compilation of Tagalog language and culture at the time of the Spanish conquest. A study of Spanish thought at the time and a reading of the missionaries‘ writings provides an insight into the process and motivations for Christianization and interaction between Tagalog and the Christian Spanish culture. The dictionary provides descriptions of Tagalog material culture and notions, even those thought to be of satanic origin and an impediment to Christianization, drawn from a first-hand experience of many years lived isolated among the natives. This is fuller than that of any subsequent dictionary. Most interesting to the linguist is the method the author found for handling morphological forms and distinctions untranslatable into Spanish, unequalled by subsequent dictionaries until modern times. These characteristics of the dictionary are discussed in the light of the role this dictionary was meant to play in the Christianization of the Tagalog population.

1. Introduction In 1613 a Franciscan Friar, Pedro de San Buenaventura, published in Pila, Laguna Province, south of Manila, an extensive dictionary of Tagalog, Vocabulario de Lengua Tagala. Little is known about the author other than what he tells of himself – namely, that he resided in the Philippines for ten years before beginning this work and then within seven years completed a manuscript to be delivered to a printer. The contents are remarkable for their fullness and clarity of presentation. It is without question the finest dictionary of any colonial language to be published in pre-modern times and in some ways, superior to any other Tagalog dictionary that was published during the Spanish colonial period, both in scope and in accuracy. It is remarkable all the more because it was produced under arduous and primitive conditions in the early years of the Spanish conquest, while the author was serving the spiritual needs of several villages located on the shores of Laguna de Bae, south of Manila. I will talk more about the contents later.

34

John U. Wolff

1.1. Goals of the Spanish domination of the Philippines In this paper I would like to discuss the peculiarity of presentation and what this tells us about the purposes of this dictionary and the author‘s conception of what a dictionary is. To do this we need some background on the Spanish conquest of the Philippines and how the preparation of this work fits into that. Missionaries accompanied the Spanish expeditions, for there was a dual purpose in Spanish colonization in the orient: first, to control the spice trade and exploit natural resources and second, to save the souls of the inhabitants (Phelan 1959: chapter 3). These two cannot be separated, for both depended on domination of the local population: the Spaniards needed to subjugate the population – that is, make the population subjects of the Spanish crown, in order to achieve their economic goals and also in order to convert them on Spanish terms, for in the Spanish view, a subject of the King of Spain had to be a Christian. Thus, conversion on Spanish terms meant subjugation (Rafael 1988: 18–19).1 The Philippines never fulfilled the hope of enriching the Spanish coffers – in fact, the colony operated at a substantial deficit throughout the first century of its existence (Phelan 1959: chapter 8), but the Spanish crown took seriously the endeavor to save the souls of the natives and gave full support to the missionary activities throughout the early years of the colony. The missionary brothers risked their lives and volunteered for the dangerous travels and arduous work out of a burning desire to save souls.2

1.2. Sixteenth Century Spanish philosophy of language and its role in lexicography San Buenaventura arrived in the Philippines in 1594, most likely originating in Spain, as did the vast majority of the missionary brothers sent there, and he most certainly had received a Spanish education. In the Spanish schools of the time, the Diccionario LatinCastellano of Antonio de Nebrija, published in 1492, and his grammar of Castilian were central to the curriculum. For Nebrija, and generally for humanistic thought of the Renaissance, grammars and dictionaries were made to codify a language. By codification was understood a statement of the grammatical rules and the meanings of forms that provide the correct rendering of the signs and significations created by God. This grammar was thought to be manifested in Latin by the writers of the Golden and Silver Ages. Latin writings of later times were deemed to be corrupt insofar as the authors departed from the forms employed by the Classical Latin authors. The humanists be1

2

Cf. the expulsion of the non-converted Jews in 1492 and the expulsion of the Moriscos in 1609. In the Spanish view, conversion was a necessary concomitant to conquest. Although the Friars also believed that subjugation and resettlement of the population was necessary to conversion, they mostly took the part of the population against the lay Spaniards, whose efforts at gaining wealth led to impoverishment, exploitation, and murder of the natives (cf. Phelan 1959: 10).

The Vocabulario de Lengua Tagala

35

lieved grammarians and educators needed to clear away the corruption that had afflicted Latin writing after the classical age and restore Latin to its perfect ability of rendering God‘s creation. Castilian, in Nebrija‘s view, also needed to be codified – that is, cast into the mould of Classical Latin grammar. Once it had been endowed with a grammar, it would become a language, capable of representing the perfection of Latin. The existence of a grammar for Castilian was thought to prevent decay through common usage (Perona 2004: 148). (In the hierarchy of perfect languages, Latin, Greek, and Hebrew were at the summit, but Castilian was in a quasi-equal position for its capability of rendering Latin once the rules of its grammar had been brought into the order and harmony of Latin.) It was appropriate to codify Castilian in this way, in Nebrija‘s view and in the view of the Spanish intellectuals of the Spanish courts, because Castilian was the closest to Latin.3 Nebrija‘s codification consisted of a grammar, the Gramática de la Lengua Castellana, and a Latin-Castilian dictionary, which was entitled Diccionario Latin Castellano. He also produced a Castilian-Latin dictionary, which he called Vocabulario Castellano-Latin.4 Nebrija was also motivated in his linguistic studies by the view that Castilian, because it had been perfected by codification, was the language in which the Empire was to be anchored: where the Castilian language reigned, there reigned the King and Queen of Spain and vice versa. In his dedication to Queen Isabella of his Gramática de la Lengua Castellana, Nebrija adumbrates the political role of a grammar and dictionary in extending imperial authority.5 Later authorities made this view applicable as well to the Spanish conquests and colonizations. In Nebrija‘s view there was a hierarchical relationship among languages. Other languages were subordinate to Castilian, and this subordination was iconic of the political order.6

3

4

5

6

Rafael (1988: 25) points out that sixteenth century theologians advocated the use of the vernacular in devotional literature and translations of sacred texts. Castilian was deemed perfected enough to convey the word of God. It is not clear how the Spanish grammarians conceived of the difference between diccionario and vocabulario. San Buenaventura calls his work vocabulario and not diccionario. Length and fullness can have nothing to do with the difference between the two terms, for San Buenaventura‘s Vocabulario is a copious work with illustrations far greater than works of the time entitled diccionario. Perhaps the difference has to do with the thought that a work that aimed to elucidate Latin, the most perfect language, merited the appellation diccionario but one that elucidated any other language should be relegated to the term vocabulario. Nebrija wrote, ―Language is the perfect instrument of empire…One thing I discovered and concluded with certainty is that language was always the companion of empire; therefore it follows that together they begin, grow, and flourish, and together they fall‖ (quoted by Rafael 1988: 13). Rafael summarizes Nebrija‘s arguments as follows (Rafael 1988: 26): Castilian grammar could be ordered on the structural rules of Latin – this implies the possibility of subordinating other languages to Castilian thus rendering Castilian suitable as a language of empire. Cf. Kamen‘s analysis (2003: 494): ―Castilians assumed that they were superior and insisted that the imposition of order had to be done through their language, which was the only means of communication that they accepted‖. Kamen argues that the inability of the officials of the Spanish Empire to

36

John U. Wolff

1.3. The decision to use Tagalog for christianization All this implies that Castilian should be the language of subjugation. Subjugation, including Christianization, meant Hispanization. However, in the Philippines, as in the New World, Hispanization and conversion through Castilian turned out to be impossible. The missionary was most often alone among a population speaking other languages vastly different from Spanish. The missionaries had no choice but to try to communicate in the native language. This is the background for a decision taken by the Custodial Chapter of the Franciscan Order in 1580 to produce dictionaries and grammars of Tagalog.7 Linguistic studies were seen as a necessary part of the missionary work. In any case, San Buenaventura‘s dictionary was composed in response to it.8 Although at least one other dictionary,9 possibly two, were produced prior to that of San Buenaventura‘s, his is the earliest to have survived, and indeed he states in his introductory remarks that his is the first dictionary. It is possible that San Buenaventura never had access to either of these works, if indeed they were ever prepared. He does mention in his introduction (¶2, under his introduction ―Notables‖) that the morphological analysis that determined the presentation in the dictionary is based on the grammar of the friar Juan de Oliver, which is no longer extant and which is said to have been based on the grammar of Juan de Plasencia. In short, San Buenaventura was working within a tradition of Tagalog translation and analysis.10

7 8

9

10

speak languages other than Castilian was largely responsible for the many failures that this enterprise encountered in its history. Que se escribiese una gramática y un vocabulario de la lengua Tagala (Santa Inés 1892). This is made clear by the Approval (Aprobación), published as part of the Evaluation (Tasado), published as a forward to the work. The Franciscan Friar, Juan Portocarerro, known as Juan de Plasencia, the author of the first Doctrina Cristiana and of several other Tagalog books, was specifically charged in the publication of the 1580 decision of the Franciscan Chapter to produce these documents. In 1585 he wrote to King Philip II that he had completed an Arte de la Lengua Tagala and was compiling a Vocabulario: En la lenga mas general que hay en estas Islas tengo escrito algunas cosas, como es el Arte de la lengua y declaracion de toda la doctrina xtiana, y agora voy haciendo el Vocabulario. Son cosas muy necesarias para todos los ministros si se imprimiese. Seria particular merced que V.M. nos haria, hacernos merced mandalla imprimir en Mexico a costa de su Real hacienda, y para esto enviarme su Cedula, que sería de grandísima utilidad para estas almas. [I have written some things in the language that is most widely used in these Islands, to wit: the grammar of the language and the whole Christian doctrine. Now I am in the process of preparing a vocabulary. These are very necessary things for all of the ministers, and if it were to be printed, it would be an especial grace that Your Majesty would grant us, to do us the favor of ordering it to be printed in Mexico at the cost of your Royal purse, and for this (to happen) to send me your written order, which would be of the greatest utility to these souls]. As Rafael (1988: 27) points out, Tagalog grammar was presented in terms of the system of Latin. Latin grammar was deemed the universal grammar created by God, and Tagalog forms were described as having the grammatical properties of the Latin forms that they render.

The Vocabulario de Lengua Tagala

37

1.4. The need to replicate Castilian concepts in Tagalog for the natives to know God As we can read in the Approval (Aprobación), printed as part of the prefatory matter (Taso), Pedro de San Buenaventura received the assignment to prepare a dictionary. As one educated on the basis of Nebrija‘s writings, San Buenaventura brings to his dictionary a conception of Castilian that was able to render Latin perfectly and therefore, render the signs and significations created by God. Having been taught through Nebrija‘s dictionaries, San Buenaventura understood a dictionary to be a repository of correct language, but not as we understand a dictionary nowadays to be: he did not think of a dictionary as a documentation of what people said but rather as a repository that represents divinely created language.11 Whereas Nebrija‘s pedagogical aim was to inculcate correct Latin, the missionary friars thought the grammatical and lexicographical studies as a way of finding a way to communicate: inasmuch as it was impossible for the missionaries to bring the natives to understanding in terms of Castilian, a means had to be found whereby they could bring the perfection of Castilian to the natives by means of Tagalog terms, in the language that the people did understand. In short, the Vocabulario was compiled to teach the missionaries what Tagalog forms replicate the Castilian concepts.

1.5. The concept of language perfection Before I go on to explain this further, I should take a little excursus to think further about what perfection in a language could have meant to a Spanish missionary friar of the sixteenth or seventeenth centuries. Their writings indicate a belief in the power of language (as God created it, not as corrupted) to invoke God‘s intercession that was total, without limit and without scope. The words of the Latin ritual had the magical power to transform bread and wine into the body and blood of Christ. Latin, as the language of Christian ritual and sacred texts, per se was a language of the perfection that could effect magic, and Castilian, as already mentioned, had the power to reflect this perfection.

11

Admittedly, in the English-speaking world the dictionary is regarded as the source of information on correct usage, but ―correctness‖ nowadays is seen as based on common usage. Language is seen as something changeable, and correctness is determined by what forms are commonly used. In our contemporary way of looking at things, the purpose of a dictionary is to document the language entirely. For Nebrija and the humanists of his time, a dictionary documented something immutable, already perfect.

38

John U. Wolff

2. The missionaries‘ conception of translation and its purpose Now let us return to the notions of the missionaries‘ conception of translation. First, they understood that a translation of Holy Scripture and ritual texts into uncorrupted Castilian, as codified by Nebrija in his grammar and dictionaries, could also have the magic power of Latin in invoking God‘s intercession. The decision to devote missionary activities to linguistic research was based on the belief that the natives could and needed to understand God‘s created signs and significations through the words of their language, interpreted to them from the Castilian and delivered to them orally. The missionaries could only teach them if they the priests knew the Tagalog forms that can render the Castilian of God‘s created language.12 Let us look at what the missionaries themselves said about this matter. First, we take San Buenaventura, who in fact says little about his motives. He only states that he undertook this arduous work for the sake of God and for the sake of the Blessed Virgin, to whom he writes a lush and fulsome dedication.13 He says nothing about the humans for whose sake he wrote the Vocabulario other than to say that he was moved to compile the Vocabulario for the sake of souls, and by that he is referring to the souls of the Tagalogs among whom he worked. We have to go to statements from other contemporary missionary grammarians and lexicographers to understand what is behind this brief reference. Perhaps the clearest statement of the purpose of these linguistic studies is found in the opening prayer of Fr. Blancas de San José, which appears in his grammar of Tagalog, published shortly before San Buenaventura‘s work. Blancas puts in his introduction the following prayer, which I translate:14 Señor Dios nuestro zelotissimo amador de las almas, que para sacallas de los errores de la ydolatria y dalles conocimiento de vos verdadero criador y padre suyo henchistes repentinamente de celestial sabiduria los coraçones de vnos homres ydiotas, y les distes don de lenguas, conque pudiessen hablar todas las q. ellos no sabian; porque assi pudiessen comunicar a todos vuestra dotrina celestial y con ella (recibiendo la ellos) vuestro amor y gracia, para despues dalles 12

13

14

Rafael (1988: chapter 3) discusses the connection between translation and Filipino subjugation and how translation fits into the political conceptions of the population. ...deterimine prosiguille y sacalle a luz solo por Dios y por vos que segun las adversidades, contrastes y contradicciones sean atrauesado, sola vos despues de Dios fuera bastante a proseguille yo. Deposito de la graçia de Dios Reçiua V.A. Ilustrissima esta peq.ña ofrenda, humilde y pobre presente deste el mas invtil e indigno sieruo suyo como por algun genero de agradeçimiento a lo mucho que a V.A. deuo [I was determined to carry it out and bring it to light for God and for You, who, as I passed through the adversities, resistance, and contretemps, You alone after God were enough for me to carry on with this work. Store of God‘s grace, receive this small offering, a humble and poor gift of this most useless and unworthy of Your servants, as a token of gratitude for the great debt I have to You]. Rafael (1988: 30) quotes this prayer.

The Vocabulario de Lengua Tagala

39

tambien vuestra bienauenturança: suplico os Señor mio con quanta humildad yo puedo querays ayudar a este pobre ygnorante ministro vuestro y dale con que el pueda ayudar a estas almas destos pobrecitos, para que os conozcan y amen. Vos dispusistes que por el oydo entrase vuestra saludable dotrina que nos ha de hazer sabio [Lord our God, most jealous lover of souls, in order to save them from the errors of idolatry and give them knowledge of You the true creator and their father, I pray that You fill the hearts of some idiotic men immediately with heavenly wisdom, grant them the gift of tongues, with which they might speak all the languages that they did not know, because in that way, they may communicate to all Your celestial doctrine and (when they have received it) Your love and grace, so that thereafter You may grant them the joy of heavenly bliss. I implore You, my Lord, with as much humility as I am able, that You may wish to help Your poor ignorant minister and give him the means wherewith he might help the souls of those pitiful people, so that they may come to know and love You. May You dispose of matters such that the ears may receive your healthful doctrine,15 which will make us wise]. This prayer asks God to help the priest find the forms of Tagalog that can convey the celestial doctrine by which the souls of the parishioners may come to know and love God, receive God‘s grace and enjoy the fruits of heavenly bliss (bienaventuranza). In other words, the purpose of these linguistic works is to expose to the missionary the rendition into Tagalog that corresponds to the Castilian forms that reflect divinely created language and have the magic powers of such language. Blanco‘s grammar, as well as Buenaventura‘s Vocabulario, are aimed at the priests who teach the population. The priest must have the right language to enable the people to know God and save their souls.16

15

16

With the word saludable ‗healthful‘, Blancas meant to refer to the belief (in Philippine as well as Christian tradition) that certain religious rites have curative powers. Cf. the quotation of Plasencia‘s references, given by Rafael (1988: 93). Although the purpose of the full and detailed Castilian-Tagalog section of the book is to teach the missionaries, a Tagalog-Castilian register, occupying 95 pages of the 707-page work, is added to ―help the natives learn Castilian‖ (…porque los naturals puedan de prender con mas façilidad a hablar la lengua Castellana). Judging by the briefness of the register, San Buenaventura did not accord teaching Spanish to the natives high priority. Not all the missionaries gave teaching Spanish to the natives as little importance as San Buenaventura. Another Tagalog-Spanish dictionary, Vocabulario Tagala, was prepared soon after San Buenaventura‘s and was apparently aimed at the natives (Postma 2000: 12). Two somewhat different manuscript copies of this work still exist (Postma 2001: 3). These manuscripts were edited and published (San Antonio 2000, Postma 2000). By meticulous examination of the contents, Postma concluded it was a derivative of San Buenaventura‘s and was most likely prepared by a Franciscan friar, Francisco de San Antonio O.F.M. (Postma 2000: xii).

40

John U. Wolff

3. The importance of accuracy in San Buenaventura‘s view Accuracy and fullness thus become a matter of saving souls. An error could have serious consequences, and San Buenaventura devotes much of his introduction to explaining that his work has errors and why it should be so. He fears the criticism of others and says simply that he is aiming this dictionary at those who know nothing (meaning less than what is in the Vocabulario), who may learn from it, not because the work itself is perfect. Implied but unstated: that they may have the right language whereby to lead their flock to salvation.17

4. The contents of the dictionary Now let us go on to discuss the contents of this dictionary. The purpose is to show how to render the concepts of Castilian into Tagalog, and appropriately, the dictionary lists Castilian forms first, normally in a citation form and not inflected. This is followed by a Tagalog form that translates it, and if a certain Castilian form has several senses that can be rendered by different words in Tagalog, as is almost always the case, each different Tagalog rendition gets a different entry. The Tagalog root is followed by inflected Tagalog words, used in a sentence and then translated. Often there is a Castilian phrase or sentence containing an inflected form of the Castilian entry or a special way it is used in Castilian followed by the Tagalog rendition. This is given, not to illustrate how the Castilian is used, but rather to show how this particular idiom can be rendered by Tagalog. There seems to have been an attempt to give a complete listing of Castilian roots (those expressing signs created by God, that is), but Tagalog forms are given only insofar as they have the power to render Castilian concepts. I once estimated that the Tagalog root is capable of more than five hundred affixational forms, but the idea that any portion of them should be cited for a given root did not come to mind in the case of the Vocabulario, for this is in fact not a dictionary in any modern sense of a lexicon that expounds a language. If one wishes to know whether this or that affixational form occurs with a given root, the Vocabulario will only provide the information in the event that that particular form happens to translate a certain Castilian concept. If that particular form does not happen to be chosen to translate a Castilian concept, there is no way of knowing from the Vocabulario whether or not it exists. 17

Solo suplico humilmente [sic = humildemente] se reçiua mi voluntad y deseo del prouecho de las almas q esto me a mouido, y el ver la falta q desta obra auia en esta tierra. No le comunico a los doctos y peritos en Tagalog pues no tienen neçesidad de el, siendo ellos cada vno de por si un copiosisimo vocabulario. Ni tampoco a los curiosos en poner faltas en trabajos ajenos, sino a los pobrecitos q poco sauen, y q con humildad se quieren sujetar al parecer ajenos, a estos le ofrezco y ruego lean en el que quiça se aprouecharan, no por lo que la obra es sino por su humildad que Dios nuestro señor. Humilibus dat graçiam.

The Vocabulario de Lengua Tagala

41

4.1. Rendition of native concepts unknown to Spanish and rendition of Christian concepts unknown to Tagalog This is not to say that the dictionary ignores native concepts that were unknown to the Spaniards. Words for articles or things that are not found in Spain are given in the entry for Spanish things that correspond in some sense. For example, tubaᾣ, the word for the fermented water from the coconut blossom, is given as the translation for ―mosto‖, which refers to grape juice that is to become wine. Indeed, tubáᾣ can be distilled into hard liquor, but its primary use is as a beverage of daily consumption, as opposed to the Spanish ―mosto‖. In other words, the Tagalog forms are only an approximation of the Castilian. This is no problem for San Buenaventura, for his aim is to find whatever Tagalog can render the Castilian, not give an accurate representation of Tagalog on its own terms. As Rafael points out (1988: 29), for certain of God‘s divinely created signs, an appropriate Tagalog rendition was not available (for all candidates had devilish implications).18 These words were left untranslated. The Castilian form reflecting these signs were deemed appropriate to convey Christian concepts to the Tagalogs. Nevertheless, the Vocabulario includes words that impinge on religious beliefs and superstitions, though the Spaniards believed Tagalog religion to be a form of devil worship. One way to obviate the satanic aspect of Tagalog forms was to ignore some of their religious and spiritual meanings. For example, núnoᾣ is translated as aguelo ‗grandparent‘, which is indeed a meaning the form had, but in contemporary Tagalog and surely in the Tagalog of the sixteenth and seventeenth centuries, this word is used only in the context of a revered and feared ancestor with supernatural powers. Another way to handle forms with devilish implications is to list them with a definition, but state up front that the word has a devilish implication. The treatment of bathálaᾣ, the Tagalog word for ‗god‘, is instructive. San Buenaventura states his belief that the word bathalaᾣ is improper to render the concept of ‗God‘ because bathalaᾣ is said to be the greatest of the anitos, a kind of devil. There are two entries for ‗God‘: Dios and Bathala: Dios) Dios (pc) C. señor n.ro a diferençia de sus Dioses, nagpapaDios 5.ac. dejarse a la voluntad de Dios; diniDios.1.P. ser algo tenido por Dios, imp: magpaDios ka, t, palooban mo siya, dejate a D. y a su voluntad, hovag mong Diosin ang ditotoong Dios, no tengas por Dios al q. no lo es [God …our Lord in contrast to their Gods...leave oneself in God‘s hands...Put yourself in God‘s hands...be something taken to be God. imperative: …Leave it to God and to His will,...Do not take anyone to be God who is not God]. Dios) Bathala (pp) ansi llamava antes estos a N.S.D. q. auq. adorauan sus Anitos Lichas y Diablillos confesavan auer vn solo Dios y señor de todo, al cual lla18

Rafael (1988: 29) makes the important point that the notion that certain forms could not be translated into Tagalog was a further indication of the superiority of some languages over others in the communication of God‘s word.

42

John U. Wolff

mauan Bathalang maycapal sa lahat. I. Dios que todo lo hiço, aunq. segun otros, se toma por el mayor de sus Anitos, set non congrue (sic = sed non congruet) [God …that is the way they formerly called our Lord God, who although they adored their ―anitos‖, ―likhas‖ and minor devils, avowed their faith in a sole God, lord of all, whom they called… ―God, who made all‖, although others say that he is considered the greatest of the Anitos, but that does not figure].

4.2. Transcription of Tagalog As for the transcription, there is no thought of writing the Tagalog forms in the native syllabary: the Vocabulario is aimed at Spanish missionaries, not at the native population, though they are the humans for whose sake it was composed. The transcription follows that of Oliver‘s grammar and documents that had been printed. This is not entirely phonemic. Tagalog at the time (as opposed to currently) had only three vowels, but Spanish linguistic thought at the time, supposed that the five vowels of Spanish and Latin were universal (God given). De Buenaventura simply assumes five vowels and writes u as well as o, choosing one or the other for a particular form, but not on any discernable principles. He transcribes /i/ normally with i, and only in a very small number of cases writes e. Further, Tagalog has long and short vowels, a fact recognized by the dictionary19 and in fact signaled by the letters ―pp‖ and ―pc‖ written after every citation, which can indicate the length of the root vowels: pp, being long penult and pc, being short penult. This system does not allow for the indication of length on derived or inflected forms, although San Buenaventura notes that long vowels may occur elsewhere in the word as well as in the penult. The Tagalog of San Buenaventura‘s time also had glottal stops that occurred freely as any other consonant did, and although Blanco‘s grammar recognizes the existence of this sound and endeavors to describe it, San Buenaventura makes no mention of it, and omits it entirely in his Tagalog entries and citations. It does not occur in Latin or Greek, nor does it occur in Spanish. Presumably, it was not one of the divinely given sounds.

5. The excellence of San Buenaventura‘s presentation I should go back to the point I made in the introduction, namely, that I consider San Buenaventura‘s Vocabulario the best lexicographical work of the Spanish colonial times. To be sure, the Vocabulario of Noceda-San Lucar, the final edition of which came out in 1860, was considered in its time, and still is, the most complete Tagalog dictionary, and this reputation is well deserved. Noceda-San Lucar endeavored to com19

Length was not understood to be a contrast of vocalic duration, but a matter of stress (acento).

The Vocabulario de Lengua Tagala

43

bine the results of all the dictionaries of Tagalog that had been made up to their times, and their dictionary contains a far larger number of entries than San Buenaventura‘s (or any other before or after them). Nevertheless, Noceda and San Lucar were outsiders to Tagalog culture much more so than San Buenaventura, who with his fourteen years of intimate life among the Tagalogs, during which he had minimal contact with his fellow missionaries or any other Spanish colonists, came to know the Tagalogs from the inside. This shines through in his presentation, despite his focus on conversion and finding ways to express Christianity in Tagalog. We could take as an example his treatment of the term kudyápiᾣ, which refers to a musical instrument and compare it with that of Noceda-San Lucar. By 1860 the kudyápiᾣ was probably no longer known in the Tagalog regions, and it is likely that San Lucar never had any contact with it. He defines it simply as guitarra suya ‗their guitar‘. From San Buenaventura‘s treatment, we can get a notion of what the instrument looked like, how it was made, and how it was played. Admittedly, this information is scattered in several places in the dictionary, and it is clearly given in the context of telling the reader how to convey certain notions in Tagalog. Here it is: Guitarra) Coryapi (pp) de esta tierra que casi es como timpano con dos cuerdas de alambre, tañese como guitarra con la mano, nagcocodyapi 2ac tañerla, nangongodyapi.3.ac. estar siempre tañendo, cqinocodyapi 1.P. ser tañida la guitarra, cqinocoryapian, serle tañida Vi. Tañer [Guitar A local instrument that is like a small drum with two strings of wire, that are played with the hand like a guitar. ... play it...always be playing...for the guitar to be played...be played for him Cf. Tañer] Tañer) Coryapi (pp) biolon o rabel, imp) magcoryapi ca tañe con biolon, coryapian mo aco, tañeme [Play Coryapi cello or ravel imperative: ...play a cello] Rabel) Codyapi (pp) ouiolon que tocan con arco, nagcocodyapi. a.ac. tocarle, cqinocodyapian .i. P. el por quien se toca, imp) magcodyapi ca, toca el rabel, codyapian mo aco, tañe me un rabel [Ravel Codyapi a cello that they play with a bow imperative... play it...the person whom it is played for… play the ravel... plan a ravel for me]. Let me illustrate these points with some further entries. A term I opened at random hechar has no less than 39 entries, and an additional eleven entries for hecharse. Incidentally, many Spanish words beginning with /e/ are listed with initial /h/.Let me illustrate all this with the 44th of the 50 entries for hechar. Hecharse Hinay (pp) el viento calmando. Vi. Abonçar) cun baga mahangin pa, y, pahinayin ninio, si haçe a un viento, esperad a que se heche. Vi. Anojar (sic = Arrojar.) [Drop) hínay for the wind to drop. Die down) …if there is wind, wait until it dies down Cf. Hurl] Abonançar) Hinay (pp) el tiempo o el viento hungmihinay.1.ac. abonançar, pinahihinay 7.P.h deaar [sic—i.e. dejar] que abonançe, imp) pahinayia mo mona

44

John U. Wolff

ang hangin, aguarda que abonançe el viento, hingmihinay na, ya esta abonançible (aunque alguno diga que este vocablo es Bisayo, usase en todas partes generalmente) [Clear up) Hínay for weather or wind hungmihinay. Clear up, pinahihinay Allow to clear up ……Wait until the wind dies down...it has cleared up now (although someone said that this word was Bisayan, it is everywhere in general use)] [Explanation: ―(pp)‖ means that the accent is on the final syllable—i.e., hináy ―1.ac, 7.P.h , etc‖ refer to the verb class as given in Oliver‘s (and Blanca‘s) grammar.] From this reference we can get a notion of the detail into which this dictionary goes in an endeavor to present the way the Castilian notion can be rendered into Tagalog. The Castilian is taken to be a known, and no effort is made to explain the Castilian forms other than to give the sense of the Tagalog form that renders the Castilian.

5.1. The Tagalog-Castilian register There is a Tagalog-Castilian register. This lists Tagalog forms and Castilian entries under which these forms are found. However, by no means are all Tagalog forms listed in the entries found in the register. The list is partially by root, but a large number of affixed forms are listed. (The roots of the affixed forms are not necessarily given, however.) This register was created to help Tagalogs learn Spanish (p. 619). There are a number of false references – forms in the register that do not occur in the body of the dictionary, some of which in fact do not even exist nowadays, and probably never did exist. The register is not a reverse dictionary. It often makes reference to Spanish forms that in no way could translate the Tagalog. For example under one of the listings of manuc, which meant ‗chicken‘ in San Buenaventura‘s time but must have meant ‗bird‘ at an earlier time, the reference is to agorar ‗practice augury‘, and indeed a verbal derivative of manuc, at that time meant to practice augury. To give an idea of the detail and complexity, I give an example from the Register: the word for ‗fear‘ or ‗be afraid‘ tákot. This is listed three times Tacot.pp) Temor o Miedo ―Fear‖ Tacot.pp) Temer.l.Atemoriçar ―Fear, Frighten‖ Tacot.pp) Amenaçar 1.Espantar ―Threaten, Frighten‖ This yields six entries under which this root is treated. And in fact, this root is also listed a seventh time under ―temeroso‖. There is a great deal that one can learn about Tagalog and how tákut may be affixed and used in sentences from these listings.

The Vocabulario de Lengua Tagala

45

5.2. Difficulties in finding Castilian equivalents for Tagalog forms Let me focus on a couple of points. There are several forms that are pretty much the same, translated into Spanish or English, but in Tagalog they mean something entirely different. The Vocabulario makes an attempt (not invariably successful) to come to grips with the various nuances. However, in comparison with other dictionaries completed before modern times, the definitions here are far more precise and understandable than those of later works. kinatátakútan ‗be afraid of something‘ San B.: ‗ser temido‘ e.g. (my example), I am afraid of the ghost kinatátakútan ko ang núnoʔ ikinatátákot ‗[so-and-so] fills [genitive] with fear‘ San B: ‗causa de temer‘. e.g., (my example) What happened frightens me, Ikinatátákot ko ang nangyári San Buenaventura‘s example: anong ikinatátákot mo? De que te espantas? ipinanánákot ‗[so-and-so] fills [genitive] with great fear‘ San B: ‗atemorizado‘ katakut-tákot ‗frightening‘ San B: ‗cosa de temor‘ katakut-tákut sabihin ‗It is scary to talk about it‘ (San B‘s example) itinátákot ‗used to frighten [direct object] with‘ San B.: lo con que amenaza, walang diʔ itinákot sa ákin, San Buneaventura‘s translation: amenaçome con mil cosas. ‗They frightened me with a thousand things.‘ (My translation of San Buenaventura‘s example: ‗They frightened me with everything.‘ (Literally, ‗there was nothing they didn‘t frighten me with.‘) Here are the entries that these examples are taken from: Temer) Tacot (pp) qualquier cosa, tungmatacot , 1ac temer, natatacot. S. estar con temor, cqinatatacotan 4.P. ser temido, ycqinatatacot 4.a causa de temer, imp) hovag cang matacot, no temas [Fear tákot something... to fear…be afraid…be feared..reason for fearing imperative don‘t be afraid] (This is followed by 3 other entries of temer giving various Tagalog roots similar in meaning to tacot.) Miedo) Tacot (pp) q. vno tiene a otro. Vi Temor) ang takot co oy niyong acoy habulin nang castila tube mucho miedo quando vi al español que yba tras mi corriendo [Fear tákot that one has of another cf. Temor ...I was very much afraid when I saw the Spaniard running after me] Temor Tacot (pp) que vno tiene de algo, Catacot-tacot, cosa de temor, catacottacot sabihin, es cosa temerosa el deçillo, ano caya ang tacot mo? de que es tu temor? may tacot ka tienes temor? [Fear tákot that one has of something…a fearful thing ...it is a fearful thing to say it...what is it you fear?...Are you afraid?]

46

John U. Wolff

(This is followed by six other entries for temor that may be rendered by various Tagalog synonyms or other forms that actually translate nuances of temor [e.g. nganib ―danger‖, where the derived form nanganganib can be rendered ―temer algun peligro‖]) Atemoriçar) Tacot (pp) a alguno, nagpapanacot. 5.ac. atimoriçar [sic]20 a alguno con figura o espantajo ipinananacot. 3.P. ser atemoriçado, imp) magpanacot ca sa manga bata Atemoriça a los niños con algun espantajo panacotin mo ang manga bata. Vi. Amenaçar. y. Temer [Frighten) Tákot someone...to instill fear in someone with an image or a scarecrow] Amenaçar Tacot (pp) con palabras o fieros, tinatacot i.P. la persona que es amenaçada, ytinatacot.1. lo con que amenaça, walang di ytinacot sa acqin, amenaçome con mil cosas imp) tacotin mo na, amenaçale. Vi Temer.y. Atemoriçar) [Menace Tákot with words or horrors…the person who is menaced...that with which one menaces...scare me with a thousand things imperative threaten him Cf. fear and frighten)] Espantar Tacot (pp) de qualquiera manera, tinatacot. i.P. ser vno espantado. Vi Temer) tacotin mo, espantale, anong ycquinatatacot mo, de que te espantas? Vi amenaçar) [FrightenTákot frighten in every which way ...be someone made afraid cf. fear ...frighten him...what is it that is frightening you? cf. menace] (This is followed by 14 other entries for espantar and five entries for espantarse.) Temoroso Matacotin (pp)21 que de nada teme. Ver 5. de tacot. l.malagimin, de lagim, malagimin kang lubha, muy temeroso eres, magtacot tacotan ka ha? ‗Haras como q. estas temoroso?‘ [Fearful Matakutín one who fears for no reason Cf. meaning 5 of tacot , synonym: malagimin from lagim, ...you are very fearful...are you going to act as if you were fearful?] (This is followed by 14 other entries for espantar and five entries for espantarse.)

6. Final words: how the Vocabulario was conceived and what it reveals In short, the Vocabulario was conceived as a document that explained how Spanish concepts could be expressed in Tagalog. In our contemporary view almost every Tagalog form renders only a portion of the meaning of a given Castilian form, and vice versa almost every given Tagalog form requires several Castilian forms to render its range of 20

21

The misspelling of i for e illustrates the difficulties Tagalog speakers had distinguishing e and i. The type was set by Tagalogs. This is an error on the part of San Buenaventura. (pp) implies a long penult matakútin, but currently and in all probability in San Benuaventura‘s time, as based on comparison with other languages, the penult of this form was short matakutín.

The Vocabulario de Lengua Tagala

47

meanings. San Buenaventura recognizes this implicitly and comes to grips with it, but the fact that Castilian does not express the totality of Tagalog is not viewed as an imperfection and no attempt is made to elucidate Tagalog forms entirely. Failure of Tagalog to express the totality of Castilian forms is a proof of its subordinate status, but San Buenaventura makes an effort to find a Tagalog way to convey the totality of the Castilian meaning. San Buenaventura does this by multiple entries. By giving a large number of Tagalog forms for almost every root, the Vocabulario makes reference to a wide range of Tagalog culture at the time, including much unlike anything known in Spain.22 I found more than twenty words referring to witchcraft, sorcery, and healing. There probably are more. Thus the Vocabulario is a treasure trove of invaluable but heretofore largely unexploited information about early Philippine culture, much of which is alive to this day. Further close study of the Vocabulario can undoubtedly reveal a host of information important for understanding the Philippines.

Bibliography A. Primary sources Blancas, Fray Francisco de San Joseph de la Orden de S. Domingo Predicador General en la Provincia de N. Señr del Rosario de las Islas Filipinas (1610): Arte y Reglas de la Lengua Tagala. Bataan: Thomas Pinpin, Tagalo. Facsimile edition with commentary by Antonio Quilis. n.d. Ediciones de Cultura Hispánica, Agencia Espanola de Cooperación Internacional, Avda Reyes Católicos, 4, 28040 Madrid. San Antonio, Francisco de, OFM (2000): Vocabulario Tagalog, edited by Antoon Postma. Quezon City: Pulong, Ateneo de Manila. San Buenaventura, Fr. Pedro de (1613): Vocabulario de Lengua Tagala. El Romance Castellano Puesto Primero. Primero y segunda parte. Pila: Thomas Pinpin and Domingo Loag, Tagalos. Facsimile edition published by Cayetano Sanchez Fuertes, of the Arcivo Franciscano Ibero-Oriental, Madrid. Servicio de Reproducción de Libros, Librerías ―Paris-Valencia‖ Pelayo 7, 46007, Valencia. Santa Inés, P. Francisco de (ed.) (1892): Crónica de la Provincia de San Gregorio Magno de Religiosos Descalzos de N.S. San Francisco in las Islas Filipinas, China, Japon, etc. Manila: Chofre. 1892. First published in Manila, 1676.

B. Secondary sources Costa, Horacio de la, S. J. (1961): The Jesuits in the Philippines, 1581–1768. Cambridge: Harvard University Press. 22

As an example, William Henry Scott (1982) studied a large number of forms referring to Filipino technology of the time. On the basis of a meticulous study of the Vocabulario, Scott was able to trace methods of agricultural production, hunting and fishing, textile preparation, construction, navigation, and others.

48

John U. Wolff

Guerrero Ramos, Gloria (1995): El Léxico en el Diccionario (1492) y en el Vocabulario (¿1495?) de Nebrija. Sevilla: Universidad de Sevilla. Kamen, Henry (2003): Empire: How Spain became a world power 1492–1763. New York: Harper Collins. Noceda, Juan Josel de (1860): Vocabulario de la Lengua Tagala, revised by Pedro San Lucar SJ. Manila: Ramirez y Giraudier. Perona, José (2004): Antonio de Nebria 1442–1522, in: Domínjuez, Frank A. & Greenia, George D. (eds.), Dictionary of literary biography, vol. 286, Castilian Writers 1400–1500. Detroit: Gale, 142–155. Phelan, John Leddy (1959): The Hispanization of the Philippines: Spanish aims and Filipino responses, 1565–1700. Madison: University of Wisconsin Press. Postma, Antoon, SVD (ed.) (2000): Introduction to the Vocabulario Tagalo of Francisco de San Antonio. Quezon City: Pulong, Ateneo de Manila University. Postma, Antoon, SVD (2001): Tagalog Vocabularios. Paper published on line by the Pila Historical Society Founction, Inc. blog.bayangpinagpala.org/2006/11/vocabulario-de-lengua-tagala.html Quiles, Antionio de (n.d.): Commentary to Fr. Francisco Blanco de San José‘s Arte y Reglas de Lengua Tagala (see above). Rafael, Vicente L. (1988): Contracting colonialism. Translation and Christian conversion in Tagalog society under early Spanish rule. Ithaca: Cornell University Press. Rosales, Antonio-Maria, O.F.M. (1984): Study of a 16th century Tagalog manuscript on the Ten Commandments: Its significance and implications (Juan de Oliver‘s ―Declaracion de los Mandamientos de la Ley de Dios‖) Quezon City: University of the Philippines Press. Sanchez-Fuertes, Cayetano (n.d.): The Vocabulario de Lengua Tagala by Pedro de San Buenaventura, OFM: Historic and linguistic value of an important Spanish-Filipino Incunabulum (Draft). Paper published on line by the Pila Historical Foundation, Inc. http://pinagpala.googlepages.com/ Cayetano English.pdf. Scott, William Henry (1982): Sixteenth-century Tagalog Technology from the Vocabulario de la Lengua Tagala of Pedro de San Buenaventura, O.F.M., in: Carle, Rainer (ed.), Gava‗: Studies in Austronesian Languages and Cultures dedicated to Hans Kähler. (Veröffentlichungen des Seminars für Indonesische und Südseesprachen der Universität Hamburg 17). Berlin: Dietrich Reimer, 523–537.

REBECA FERNÁNDEZ (VALLADOLID/SPAIN)

Hispanic Philippine historiography: a brief description of the structure of the Calepino Ylocano (18th C.)

Abstract In this article we present a brief and general study of the macro- and micro-structure of an 18th century Hispano-Philippine vocabulary manuscript: the Calepino ylocano o vocabvlario de yloco en romance written and compiled by Spanish Augustinian missionaries. We will examine the historical and linguistic background to the text, and identify its authors and primary and secondary sources. We will also describe the most distinctive aspects of the inner structure of the vocabulary: entries and types of lexicographic definition. We will then describe the internal and external structure, together with examples, phraseology, Hispanisms and Americanisms included in the vocabulary. However, the main purpose in this article is historiographical.

1. Introduction The legacy of missionaries in the Philippines is widespread and notable in terms of the number, complexity and value of their works Sueiro Justel (1999: 131) lists the number of lexicographic or grammatical works completed for the main Philippine languages: Tagalog – twenty-eight grammars and twenty-five vocabularies, Visayan – twenty-six grammars and eighteen vocabularies – followed by Ilocano – twelve grammars and ten dictionaries, and fewer for Bicol or Pampango. These works were mainly intended for use by missionaries learning the indigenous languages, and we thus find not only grammatical and lexicographic information, but also translation theories and points of catechism, as well as habits and customs. The Calepino Ylocano o vocabvlario de yloco en romance compuesto por diferentes padres ministros antiguos, diestros en este ydioma y últimamente corregido, y añadido segun lo que aora se vsa y de vltima mano, por el padre fray Pedro Bibar, de la orden de nuestro gran padre san Augustin, ministro en esta provincia, visitador, que ha sido de ella, examinador sinodal de este obispado de Nueva Segovia y prior vocal del comvento de Bantay [Calepino Ylocano or Ilocano-Spanish vocabulary composed by different ancient Fathers fluent in this language, and finally amended and updated according to current use, by Father Pedro Vivar, of the Order of our great Father Augustine, a Minister and visitor in this province, who has acted as educational inspector of this diocese of Nueva Segovia and Rep-

50

Rebeca Fernández

resentative Prior of the Convent of Bantay], by Pedro Vivar (1730–1771),1 is a correction and extension of a manuscript, now lost, entitled Thesauro (c. 1700), compiled by the Augustinians José Carbonel (1665–1711) and Miguel Albiol (ý 1710).2 The Calepino Ylocano is extremely important, as all subsequent vocabularies of Ilocano were based on it. We can therefore affirm that it is comparable, in terms of significance, with the Philippine lexicographic works completed by Pedro de San Buenaventura (ý 1627) for Tagalog, and Alonso de Méntrida (1559–1637) for Visayan.

2. Background and sources Some authors (Villoria 1997: 124, Sueiro Justel 2003: 80), quoting Elviro Pérez, state that the Augustinian friar Pedro de la Cruz Ávila (ý 1617), about whom we know almost nothing, wrote the first vocabulary in Ilocano around 1600. However, it is certainly true that the first printed work in Ilocano was the Doctrina Cristiana [Christian Doctrine] (1621), by the Augustinian friar Francisco López (ý ca. 1730),3 who also published the first grammar, Arte de la lengua yloca [Grammar of the Ilocano language], in 1627. The Calepino Ylocano suggests that López also wrote a Vocabulario de romance al iloco [Spanish-Ilocano Vocabulary] although it was not printed due to his premature death. Given that the López vocabulary was probably unfinished; two other Augustinians, the above-mentioned Carbonel and Albiol decided to complete it in their work known as Thesauro [Thesaurus]. However, the vocabulary contained so many errors that the leader of the Augustinian Order commissioned Vivar to recompile and rework it. In order to do this, the different letters were distributed among the most talented missionary linguists and they corrected, added and ordered everything which was incorrect. So, once Vivar had put everything in order, the vocabulary was ready to be sent to Manila for printing. The Calepino Ylocano obtained printing licences in 1760 and 1761. However, it was never actually printed in the form set out by Vivar. For this reason, another Augustinian friar, Andrés Carro (1733–1806),4 altered it by moving 1

2 3 4

Pedro Vivar was born in Logroño in 1730. He was ordained in 1750 at the Augustinian Hospicio de Sto. Tomas de Villanueva, Mexico. He arrived in the Philippines in 1752 and was appointed parish priest of Candon on Oct. 31, 1752. In 1754 he was appointed as a missionary among the Igorots of Tanglo and Benguet, and from 1757 until his death he was appointed parish priest of several towns of the Ilocos region: Bacarra on May 3, 1759; Batac 1762; Bantay on April 25, 1765 and Paoay on April 10, 1769. He died in Batac in 1771 (Hernández 1998: 59). He is also referred to as Rivar by some authors (Quilis 1997). He is sometimes referred to as Arbiol and as Alviel. Francisco López was born in the province of Toledo (Spain). He arrived in the Philippines in 1598. Andrés Carro was born in Pedrosa del Príncipe in the province of Burgos in 1733, and died in Manila in 1806. He joined the order in Valladolid, and arrived in the Philippines in 1759. From 1762 to 1792 he worked as a missionary in the Ilocos region, where he acted as Prior, Provincial Visitor, Educational Examiner for the diocese of Nueva Segovia, and Minister of Masingal.

Hispanic Philippine historiography

51

lexical collocation information from the final appendix to the relevant entry. Words out of place were moved into their exact alphabetical order: ―Assimismo los adiciones de muchos terminos, que tenia al fin, se han distribuido, y puesto en las letras, q[ue] les correspondia. [Also many terms which were listed at the end have been distributed and placed under the relevant letters.]‖ (Vivar 1797: 37) The title of the work indicates that the author was Pedro de Vivar: Vela (1931: 346) believed that Vivar had been assumed to be the author as a result of the unfinished version left by Carro in 1792. However, he stated that it is possible that Vivar had prepared the vocabulary for printing in 1760, as stated in the prologue, despite having spent only three years working in the Ilocano language, as his task was actually limited to putting the finishing touches on the approved material passed on to him. Retana (Vela 1931: 194), meanwhile, studied the work and drew some unfavourable conclusions for Carro, whom he believed to have stolen the lexicographic glory from Vivar, the true compiler and expander of the work completed by the eminent López. The key lies in the prologue, where we find a paragraph which could not have been written by Vivar, as he only worked as a missionary for fourteen years in the Ilocos region and not thirty as stated in the vocabulary: ―Lo que yo hè notado en treinta años de exercicio con estos hermanos ylocos… [What I have noticed in thirty years of work with these Ilocano brothers...]‖ (Vivar 1797: 17). On the other hand, thirty years is the exact period of time spent there by Carro. As regards acknowledgements, the prologue mentions the invaluable help of the leaders of the village of Masingal where Carro was posted: ―Para todo lo qual me han ayudado, y valido, ocho algunos principales de este pueblo de Magsinĝal [For all the help they have given me, which has been very useful, some eight leaders of this village of Magsinĝal]‖ (Vivar 1797: 37–38). Unfortunately, we do not know if Vivar worked there. We can therefore conclude that the manuscript in question is the compiled work of Vivar, with amendments by Carro. As regards the date, despite attempts by some authors (Vela 1931, Sueiro Justel 2003) to place it in 1771 or 1792, there is no clear evidence in favour of either of these. However, we are inclined to date it at 1797, as this is the latest date mentioned in the work.

3. Description of the vocabulary The Calepino Ylocano or Vocabulario de yloco en romance consists of four sections: introduction, prologue, notes and vocabulary. The introduction, entitled A los muy reverendos padres ministros evangelicos de la provincia ylocana assi seculares como regulares, salud Jesuchristo, que es la verdadera salud [May the most reverend evangelical fathers in the Ilocano province, both secular and regular, be granted the health of Jesus Christ, which is the true health] covers eight sheets in one single column. The second part, some 26 sheets long, is the prologue, which contains a range of information and rules about figures, accents, pronunciation and measurements of time, among oth-

52

Rebeca Fernández

ers, abstracted from the Arte de la lengua yloca (1628) by López, to which it refers for clarification. The last page before the vocabulary is devoted to notes summarising the features of the language and certain issues regarding the organisation of the vocabulary and printing licences. The final part is the actual Ilocano-Spanish vocabulary, with 8040 entries on 725 sheets. It is in alphabetical order: A, B, C, D, E, F, G, H, I, L, M, N, O, P, Q, R, S, T and U. It should be noted that the following consonants are not included: J, K, V, W, X, Y and Z. This order is similar to that used by Méntrida in his Visayan vocabulary: A, B, C, CH, D, E, F, G, H, Y, I, L, M, N, O, P, R, S, R, approximant V and liquid approximant V, although Méntrida included CH and Y, possibly as it was a two-way vocabulary: Spanish-Visayan and vice versa. One should also note that Vivar did not use either K or W, nor did he use an NG digraph unlike modern Ilocano dictionaries (Rubino 2000) which use the order A, B, D, E, G, H, I, K, L, M, N, NG, O and U, P, R, S, T, W, Y. Vivar, however, did differentiate between O and U, despite the fact that this merely represented a phonetic variation of what was one single vowel in the baybayin phonological system, the preSpanish syllabic writing system.

3.1. Description of the micro-structure The entries consist of an Ilocano term, followed by the abbreviated accent, either P.P. (penultima producta) – word stress falling on the penultimate syllable – or P.C. (penultima correcta) – word stress falling on the final syllable; the equivalent term(s) in Spanish, with a definition or explanation; different illustrative phrases, generally with their translations into Spanish; grammatical information on word formation (passive, preterite and future); and lexical collocations (sayings, teachings, adages, and riddles) with a Spanish or Latin translation and, on occasion, etymological information, biblical quotations and synonyms. The lexical entries are usually complex. However, not all entries contain this descriptive complexity: over a third include only the abbreviated accent marking and the meaning, as in (1) and (2).5 (1)

Cogmès. P.C.: sico + ropa mojada. [Cogmés. P.C.: sico + wet clothing.] (1797: 274)

(2)

Combabuy. P.P.: cochinillas. [Combabuy. P.P.: cochineal insect.] (1797: 276)

They do not tend to include grammatical meta-language except in the case of adverbs and verbs where the tense – passive, preterite or future – is given, as in the following examples: (3)

5 6

Alay-ày. P.P.: + an, (echo ya verbo pasivo) fuego mancho, cocer alguna cosa a fuego lento, + an, mang + el que v.g.6 el cosinero; San Lorenzo, y otros martires, que fueron assi quemados por la fe de Dios. V[éase] sarabàsab. [Alay-áy. P.P.:

All accent marking in the examples in this article faithfully reproduce the marking in the source. v.g. = verbi gratia

Hispanic Philippine historiography

53

+ an, (passive verb) a gentle flame, to cook something on a slow fire, + an, mang + he who, e.g. the person who cooks; San Lorenzo, and other martyrs who were thus burned for faith in God. See sarabásab.] (1797: 61) (4)

Cagdipas. P.C.: ma +; en el ayre se acaba (modo) na + pretérito. [Cagdipas. P.C.: ma +; it is finished in the air (manner) na + preterite.] (1797: 234)

(5)

Tonocuà. P.C.: despues; adverbio de t[iem]po; alargar, ò dejarlo de un rato, para otro, + en, l, dipdipaien, luego de aqui a un rato, +, l, no madamdama, l, despues, l, de aqui à un rato. [Tonocuà. P.C.: after; adverb of time; to grow longer or to do it some other day, + en, l, dipdipaien, after a while, +, l, no madamdama, l, after a while] (1797: 730)

The gender or number of the entries is never indicated. Nevertheless, archaisms and terms derived from other languages are indicated: (6)

Alonquin. P.C.: termino antiguo, y q[ue] no usan aora; significaba cosa de moldura, no escultura. V[éase] gurit. [Alonquin. P.C.: ancient term, which is not used now; it used to mean something moulded, not sculpted. See gurit.] (1797: 75)

(7)

Campilan. P.C.: catana, alfanje, ò machete, saure, ò sable vos visaya. [Campilan. P.C.: cutlass, sword, or machete, blade, or saber, a Visayan word.] (1797: 250)

(8)

Botónes. P.P.: boton de vestido, l, broches abotonar v.g. jubon + an ti jubon, agi + botonarse, l, botine voz española, muy usada ent[re] estos. [Botónes. P.P.: a button on clothing, l, button broaches e.g., jubon + an ti jubon, agi + to button up, l, botónes a Spanish term, commonly used among these people.] (1797: 209)

3.2. Description of the definitions In the definitions we find an ―x = y‖ pattern, where ―y‖ refers to the linguistic expressions in the entry. (9)

Abàda. P.P.: Es el pexe mulier que llaman v[éase] dicyung. [Abáda. P.P. This is what they call the ‗dugong‘.7 See dicyung.] (1797: 40)

(10)

Abacá. P.C. Es una yerba llamada assi y de que hacen cordeles, y cables muy fuertes. V[éase] pagúa. Es una especie de plantano (sic) cuyo tronco sirve de cañamo. [Abacá. P.C. This is a grass known by this name and which is used to make very strong cords and ropes. See pagúa. It is a species of plantain whose trunk is used as hemp.] (1797: 40)

7

A marine mammal of the order Sirenia, found in waters of the Far East and Indonesia, similar to the manatee of Latin America and the Caribbean.

54

Rebeca Fernández

Occasionally, the definitions are introduced using comparisons with reference to known terms, as synonyms: (11) Omoc. P.P.: nido de paxaro, ò de otro ave; cosa anudada; ò enredada, como v. g. madexa de algodon; na +, v[éase] cosocosò. [Omoc. P.P.: nest of a small bird, or another bird; a knotted or tangled thing; like e.g., cotton threads; na +, See cosocosò.] (1797: 480) At other times, reference is also made to the Spanish term using comparisons: (12)

Pacac. P.C.: arbol assi llamado; la fruta es como calabazas y las pepitas como abas. [Pacac. P.C.: a tree known by this name; the fruit is like squash and the pips are like beans.] (1797: 488)

Or expressions such as a manera de [by way of] are used in order to describe objects: (13)

Cotocot. P.P.: lanza a manera de punzon, con que agujeras (etc.), lanzillo de hierro como clavo, v[éase] colòcol. [Cotocot. P.P.: a lance with a pointed shape, which you use to make holes (etc.), a small iron lance like a nail, See colòcol.] (1797: 280)

(14)

Lambòn. P.C. vestido largo a manera de sotana, ò havito. [Lambón. P.C. a long outfit in the style of a cassock or a habit.] (1797: 407)

On only one occasion is the expression hechura de [made from] used, preceded by the verb ser [to be]: (15)

Àrip. P.P.: gàlip, P.P. l, cadit, P.P. cuchillo corbo, que usan estos, para capar los cocos, quando hacen vino, es de hechura de hoz: y assi, para segar qualquiera cosa de yerba; dizen, gapasen de gapàs, P.P. el cochillo con que siegan el arros, ò le cortan la espiga, à su modo llaman raqèm, P.C. el termino, gàlip, l, càdit, es la mas comun; y en esta àbag[atan], no entienden el àrip. [Árip. P.P.: gálip, P.P. l, cadit, P.P. curved knife, used by these people to cut coconuts, when they make wine, it is made from a sickle: and thus, for cutting any grasses; they say, gapasen de gapàs, P.P. the knife used to cut rice, or they cut the ear in their way they call it raqèm, P.C. the end, gálip, l, càdit, this is the most common; and in this àbag[atan], they do not understand árip.] (1797: 120)

Synonyms are also introduced using the verbs llamar [call], decir [say] or parecer [seem]: (16)

Capirote: ydem, de di[s]ciplenantes; los viejos dicen caparosi; castellano. [Capirote: the same as flagellant; the elders say caparosi; Spanish.] (1797: 256)

(17)

Lolonglòngan. P.P. pescado, que parece ser el erizo de la mar, y hiere. [Lolonglòngan. P.P. fish, which looks like a sea urchin, and causes pain.] (1797: 434)

Hispanic Philippine historiography

(18)

55

Bulong-onas. P.P. pescado que llaman espada. [Bulong-onas. P.P. fish which they call ‗sword‘.] (1797: 223)

In certain definitions, reference is made to the similarities or differences with other well-known concepts in order to explain terms: (19)

Adùnget. P.P.: um + agùur. P.P. Llorar à voces, dando gritos, muy semejante al dung-aò, que es el llanto de los yndios, a sus difuntos, otros dicen, anùngur, como en Dingras, y Bangui. [Adùnget. P.P.: um + agùur. P.P.: To cry out, shouting, very similar to dung-áw, which is the call of the local people to their dead; others say anùngur, such as in Dingras, and Bangui.] (1797: 51)

(20)

Tòcàc. P.C.: rana, à otras llaman battubattùg, v[éase] pilat; semejantes en parte al sapo; a la rana comun dicen +. [Tòcàc. P.C.: frog, others are called battubattùg, see pilat; similar to toad; a normal frog is called +] (1797: 726)

The verb ser [to be] is used to indicate size or shape; in addition to including descriptive definitions regarding material, size and colour: (21)

Carabùyo. P.P.: concha de almeja, con que se arquean los pelos, tabog, es grande y colorada; quitarse la barba con ella, ag +; agiming + an, quien se la quita el otro; aripit; es como tenazilla de hierro; ag +; + an. [Carabùyo. P.P.: clam shell, used to cut hair, tabog, it is large and coloured; for removing the beard, ag +; agiming + an, the person who removes from another person; arípit; like an iron crimper; ag +; + an.] (1797: 258)

The lexicon in the Calepino Ylocano covers the daily Christian needs of the Ilocano people and their missionaries. There are definitions of herbs and plants and their medicinal remedies, animals, agricultural terms – mainly for the cultivation of rice – together with parts of the human body and daily objects, such as clothes, tools, and games, among others. (22)

Singitan. P.P.: la regaliga; yerva medicinal, para granos majada y embuelta en ojas, su asada en el rescoldo emplastarla. [Singítan. P.P.: liquorice; a medicinal herb, crushed into grain and wrapped in leaves, spread out for cooking on the embers.] (1797: 652)

(23)

Tàyab. P.P.: dongdong, olla, en que cuesen las yerbas, y el pescado; cazuela vasija en que se guisa, +, l, caldera. [Tayáb. P.P.: dongdóng, a pot used for cooking herbs, and fish; a china pot for stewing, +, l, cauldron.] (1797: 687)

56

Rebeca Fernández

3.3. Lexical collocations One of the main features of the vocabulary, as mentioned above, is the inclusion of lexical collocations in the definitions: a) Ilocano proverbs and sayings, sometimes with a translation into Spanish, which usually refer to physical and biological defects, such as illness and death, and reprehensible social conduct, such as rage and greed, among others; (24)

Alanà. P.C.: um + ti carayàn. Viene el rio medio de avenida. Refran: No carayàn ti um +, mangngalap ti aggongonà a rio rebuelto, ganancia de pescadores. [Alaná. P.C.: um + ti carayàn. Coming from the midstream of the river. Saying: No carayàn ti um +, mangngalap ti aggongonà troubled waters, fisherman‘s gain.] (1797: 62)

(25)

Lusùlos. P.P.: ag + escusarse, l, renunciar; ò descabullirse de su lugar. Ref[ran] yt, dinapai rebbèng, qet ag +, pacalaan baddùt. El escusarse antes de la ocacion, es culparse. Excusatio non petita, àcusatio (etc.). [Lusùlos. P.P.: ag + take leave, l, to renounce; or to slip away from one's place. Saying yt, dinapai rebbèng, qet ag +, pacalaan baddùt. Unwanted excuse means manifest accusation. Excusatio non petita, àcusatio (etc.)] (1797: 442)

b) Teachings and adages, normally without a translation into Spanish, on religious issues – parts of the Sacred Scriptures, sins, martyrs – negative traits and actions – death of a father, wealth, sadness – and moralistic resources – if you want to be wise, you must know yourself. (26)

Bulsec. P.C.: ciego de ambos ojos; demmang à +, cegar de uno quebrar los ojos del animal; + en cegarle, inchazon de gallina, ò ave, agbu +, l, ceguera de gallina = agbul +, jugar a la gallina ciega. Sentencia; iti + abulsèc muet ti manguyog agpadada à mailabùg, si cecus cecum ducit, (etc.). If the blind lead the blind… [Bulsec. P.C.: blind in both eyes; demmang à +, blinding by removing an animal‘s eyes; + en to blind a chicken or bird, agbul +, l, blindness in a chicken = agbul +, to play blindman‘s bluff. Teaching: iti + abulsèc muet ti manguyog agpadada à mailabùg, si cecus cecum ducit, (etc.)] (1797: 223)

(27)

Baroconĝ. P.P.: pecho generalmente, de hom[bre] o de animal, sentencia, auan ti naimbag à idaton, à cas iti nadalùs a +. [Baroconĝ. P.P.: chest, generally man or animal chest, saying, auan ti naimbag à idaton, à cas iti nadulùs a +] (1797: 186)

c) Riddles, which include a question in Ilocano and an answer in Spanish. (28)

Bocò. P.C.: artejo, ò conjuntura de los dedos, l, nido de cañabojo, o conjuntura de pie. Adivinanza: cumarayàm à +; el caracol. [Bocò. P.C.: knuckle, finger joint, l, a reed bed, nest, or foot joint. Riddle: cumarayàm à +; the snail.] (1797: 205)

Hispanic Philippine historiography

(29)

57

Bulding. P.C.: tuerto de un ojo, que le falta; tener nube en un ojo, busà adivinanza: natay adi icutcut, cadua qet ti nabiag, acacuycuyog, el tuerto [Bulding. P.C.: the one-eyed, he has only one eye; to have a cloud in one eye, busà. Riddle: natay adi icutcut, cadua qet ti nabiag, acacuycuyog, the one-eyed] (1797: 220)

The copyist tried to help the reader by adding the words proverbs, sayings and riddles in the margin of the vocabulary where these can be found.

4. Latin The author of the Calepino Ylocano used Latin to highlight certain points, to add a teaching or saying, or perhaps for reasons of style for cross referencing with other entries: a) unde, vide; b) v.g. or verbi gratia; and c) etc. On other occasions, Latin is also used for didactic or descriptive reasons, as it is normally used for examples or to indicate the pragmatics of some expressions; as well to explain morally incorrect issues, or those with scatological euphemisms related to genitals or sexual practices. (30)

Saporiding. P.P. (véase monng, (sic)) verbum impudicum, quo nominatur pars verenda mulieris. [Saporíding. P.P.: (See muting) verbum impudicum, quo nominatur pars verenda mulieris.] (1797: 627)

(31)

Ùbèt. P.P. culo; el mismo ojo de lavia; dicenlo tambien, de suelo de cualquier vasija; llamanlo tambien puring girit gigiritan. Solar de cesto, ò arca (etc.), ubtàn; suelo de cesto, ò cantaro, olla, ò arca escrivania, petaca + peccatum nefandum. Mang +. [Úbet. P.P. bottom; the very eye; also used for the base of any pot; also called puring girit gigiritan. The floor of a basket or chest (etc.), ubtàn; floor of a basket or pitcher, pot, or scribe‘s chest, or flask + peccatum nefandum. Mang +.] (1797: 750)

When using Latin, for the most part, the author does not bother to translate or explain the meaning as it was considered that the main readers were other missionaries who were assumed to be well-versed in this language.

5. Hispanisms The introduction of Hispanisms in the Philippine languages was a necessity in the light of realities unknown until that time by the indigenous people, making it easier to opt for the Spanish term than to choose an indigenous word to name new objects, foods, flora and fauna. The Calepino Ylocano contains numerous Hispanisms: 135 entries and 75 Spanish terms in the explanations, 38 of which are no longer in use (Rubino 2000). These mainly refer to: a) religion: bendición [blessing], comulgar [take communion],

58

Rebeca Fernández

confesar [confess]; b) European culture, mainly animals: cabayo [horse], baca [cow], toro [bull]; fruits and vegetables: coles [cabbage], ubas [grapes]. The Hispanisms we have found as entries are the following: Albahàca [basil], alomigas [ants], andan [to walk], aritos [little ring], articulo [article], asar [to roast], asùcar [sugar], bàca [cow], bacalao [cod], balsa [raft], bandera [flag], bàra [stick], barbaro [barbarian], basar [to go down], basta [vast], bendición [blessing], bendita [blessed], bigote [mustache], biscocho [sponge cake], bisti [dress], bola [ball], boticario [apothecary], botisa [pitcher], botones [buttons], votos [vows], brea [pitch], bruto [careless], vuelta [turn], cabál [fine], caban [to dig], cabayo [horse], cabeza [head], cama [bed], camisa [shirt], campana [bell], candela [candle], candelero [candlestick], canela [cinnamon], canta [to sing], capilla [chapel], capirote [pointed hood], capitan [captain], capote [cloak], capuyo [bud], casuela [casserole], cinta [belt], clavos [nail], cocinero [cook], cofrades [member of brotherhood], cofradia [brotherhood], cola [tail], colacion [lunch], colchon [mattress], coles [cabbage], comedia [comedy], comostaca [how are you?], comulgar [to take communion], confesar [to confess], confirmar [to confirm], corona [crown], corral [farmyard], correa [belt], corredor [corridor], cosina [kitchen], cota [doublet], crimen [crime], cruz [cross], cruzcruz [cross], cuentas [bead], cuerdas [ropes], culantro [coriander], cusel, custodia [monstrance], cutsara [spoon], champan [champagne], charidad [charity], chinelas [flipper], chirimías [moaner], chrisma [consecrated oil], christal [glass], christiano [christian], dado [dice], daga [dagger], diablo [devil], Dios [God], dispencero [larder], domingo [Sunday], entremete [to meddle], estancia [room], evangelio [gospel], evangelista [Evangelist], fiestas [day of obligation], fino [fine], firma [signature], fluecos [clog], freno [brake], frontal [frontal], ganso [goose], gargantia [necklace], gloria [glory], golosa [sweet-toothed], hoy [today], imbido [envy], lacasa [house], lamesaàn [table], limus [alms], missa [mass], moldi [cast], oliva [olive], onda [wave], ondo [deep], oras [hours], pàdre [father], padron [register], pantocos [flippers], pepino [cucumber], pesos [money], pila [sink], pinta [dot], posta [relay], rama [branch], rapas [to shave], robo [to steal], sacar [to take out], salta [to jump], sintas [ribbon], sugar [to play], templò [temple], tilde [accent], tiña [filth], toca [to touch], trinidad [trinity], tumba [tomb], ubas [grapes] We believe the missionaries included these terms in the entries because the Ilocano people already used them in their daily conversations since Hispanisms represented the purity of the Christian doctrine. Nevertheless, there were other Hispanisms in the explanations: Acta [minutes], alar [wings], altar [altar], angeles [angels], catre [bed], conserva [cans], convento [convent], cuaresma [Lent], demonios [devil], diamante [diamond], dispensa [larder], doctrina [doctrine], escrivania [writing desk], espada [sword], espiritu [spirit], filisteos [Philistine], fiscal [fiscal], frasco [bottle], gi-

Hispanic Philippine historiography

59

gante [giant], governador[ governor], habito [habit], imagen [image], isla [island], jamon [ham], jarro [pitcher], Jueves santo [Maundy Thursday], justicia [justice], laghuerta [garden], leon [lion], letra [letter], libro [book], lobo [wolf], manga [sleeve], martires [martyr], monjas [nun], mostasa [mustard], obispo [bishop], olandes [Dutch], olla [pot], oracion [pray], organo [organ], paño [cloth], pelota [ball], picota [pillory], prior [prior], prophetas [prophet], provincial [provincial], puñal [dagger], purgatorio [purgatory], relicario [reliquary], romero [rosemary], sachristan [sacristan], sachristia [sacristy], sacramento [sacrament], salve [Hail Mary], santissimo sacramento [Most Holy], señora [Lady], soldado [soldier], tambor [drum], teclas [key], tinta [ink], torete [little bull], toro, [bull], baca [cow], virgen [virgin], visita [religious area], vísperas [vespers], infierno [hell], zapatos [shoes]. We also find 70 Spanish and Judeo-Christian proper nouns among the explanations:

a) Evangelical names: Santa Maria Virgen, Jesus Nazareno, Jesu Christo, Jesús, Jesuchristo, Christo, Adan, Eva, Espiritu Santo, San Juan Bautista, San Lucas, Nuestra Señora la Virgen, Pasqua, Job, Padre Costales, David, Jerusalem, Santo Thomas, San Nicolas, Dios, Santa Clara, San Lorenzo, San Juan de Dios, San Agustin, Santo Domingo, Domingo de ramos, Dios Nuestro Señor, padre Ribera and Goliat; b) Non-religious toponyms and names: Castila, Castilla, España, Juan de Salcedo, Juan Pablo de Tholentino, Pedro, Juan, Andres, Francisco, Alonzo, Jesus, Pablo, Augustin, Antonio, Manila, Nueva Segovia, Montoya, Lopez, Guerrero, Funtanillo, Nuñez, Marzan, Davalos, Joseph, Carbonel, Miguel, Alviol, Filipinas, Belarmino, Velarmino and Mexico. From the 16th century, the missionaries introduced Hispanisms into Philippine languages, on the one hand to prevent the use of indigenous terms which may have led to confusion between pagan and religious terms and, on the other, in order to introduce new concepts. This strategy also responded, therefore, to the pedagogical and didactic interests of the religious orders. Hispanisms could be modified by expanding or reducing their meaning, with or without a corresponding phonetic transcription. However, the missionaries also introduced loan terms or terms followed by an explanation.

6. Americanisms There are many terms from American Indian languages in Philippine languages, generally reflected in new concepts brought by the Manila Galleon, which carried products and materials unknown on the Islands until that time, originating in Europe and the New World. The Calepino Ylocano includes a total of 39 of these terms, of which only half a dozen have their own entry, while the rest are used by way of explanation or clarification of another term. They are of diverse origin: primarily Náhuatl, Arahuacan, Taine,

60

Rebeca Fernández

Quechua or Antillean. They are as follows: Atole [hot maize drink], barbacoa [barbecue], batea [flat-bottomed boat], bejuco [liana], cacao [cocoa], camote [sweet potato], canoa [canoe], caiman, coyote, chichigua [pittance], chile, chocolate, Cuzco, enaguas [petticoat], guayaba [guava], guayabo [guava tree], inga [Inca], yguana, loco [mad], maiz [corn], mecate [string], mexicano [Mexican], México, mote [maize], papagayo [parrot], papaya [papaya tree], petaca [hipflask], petate [kit bag], quilite, sabana [sheet], tabaco [tobacco], tiburon [shark], tinguianes, tocayo [namesake], uracan [hurricane], zacatal [grassy area] and zacate [grass]. Only the following are also included by Méntrida in his Visayan vocabulary: canoa, caimán, bejuco, maíz, papagayo, huracán, tabaco, tiburón, iguana, papaya, bate and guayaba. In many cases, there is a difference in spelling, for example in huracán [hurricane] and iguana. Lope-Blanch (1990: 161), meanwhile, highlights the following Americanisms in the Covarrubias Tesoro de la lengua castellana o española (1674 [1611]): acal, araucana, cacique, caimán, canoa, coca, Cuzco, hamaca [amok], huracán, inga, maíz, Michoacán, mexicano, México, mico, Motezuma, nopal, Perú, perulero, pita, Tenochitclán, tiburón and tuna [prickly pear]. Only nine of these appear in the Calepino Ylocano, which does include bejuco, unlike Covarrubias. The inclusion of Americanisms by the missionaries was due to the fact that these terms already formed part of habitual Spanish vocabulary. We can therefore conclude that the Philippine lexicographical work constituted a step forward and a series of improvements of Meso-American models, which, in turn, were adaptations of works originating in Spain. Today, only twenty Americanisms still remain in the Ilocano language (Rubino 2000).

7. Conclusions We have tried to provide an overview of the micro- and macro-structure of an 18th century Ilocano-Spanish vocabulary, the Calepino Ylocano, a highly complex vocabulary which includes basic grammatical information and a large lexicon. The grammatical meta-language is very scarce, although highly innovative, due to the use of the term cortadillo instead of saltilllo to refer to the glottal stop, and the use of penultima producta and penultima correcta for the two types of word stress. The authors were very sensitive when it came to including information relating to the life and customs of the Ilocano region and of the Islands in general, with the inclusion of terms from other Philippine languages. Similarly, they were happy to include Hispanisms and Americanisms which expanded concepts on the islands with the aim of avoiding religious equivocation. The Augustinian Missionaries described lexical collocation and common sayings of the Ilocano people. It broke with the Nebrija tradition, and followed in the footsteps of the Calepino in its lexicographical style, including examples of use, sayings, teachings and riddles. As a result, this vocabulary of a Philippine language, just like the American Indian ones, is an incredible source of information.

Hispanic Philippine historiography

61

Bibliography A. Primary sources Blancas de San José, Francisco (1997 [1610]): Arte y reglas de la lengua tagala por el padre f. fray Francisco de S. Ioseph de Orden de S. Domingo Predicador General en la provincia de N. Señora del Rosario de las Islas Filipinas en el partido de Bataan. Thomas Pinpin Tagalo. Estudio y edición de A. Quilis. Madrid: Ediciones de Cultura Hispánica, Agencia Española de Cooperación Internacional. Carbonel, José & Albiol, Miguel (ca. 1700): Thesauro. (ms.) Carro, Andrés (1848 [1888]): Vocabulario de la lengua ilocana, trabajado por varios religiosos del Orden de N. P. S. Agustín, coordinador por el M. R. P. Predicador Fr. Andrés Carro, y últimamente añadido y puesto en mejor orden alfabético por dos religiosos del mismos Orden. Primera edición. Manila: Establecimiento tipográfico del Colegio de Santo Tomás, a cargo de Don Manuel Ramírez. Manila: Manuel Ramírez. de la Cruz Ávila, Pedro (ca. 1600): Arte, vocabulario y catecismo. (ms.) (not found) de Castro, Agustín María (1954 [1780]): Misioneros agustinos en el Extremo Oriente 1565-1780. Obra inédita que con el título Osario Venerable compuso el agustino P. Agustín María de Castro. Año de 1780. Edición preparada por Manuel Merino. Madrid: CSIC. López, Francisco (1628): Arte de la lengua yloca compuesto por el padre Fr. Francisco López del orden de N. P. S. Aug[usti]n con las licencias de los Superiores en Manila en la Compañía de Jesús por Raym[un]do de Peñafort. Año 1628. (ms) [1792, 1793, 1895] López, Francisco (1621): Libro á naisuratán ámin ti bagas ti Doctrina Cristiana nga naisúrat iti libro ti Cardenal á Agnagan Belarmino, Ket ináon ti P. Fr. Francisco López padre á S. Agustin, iti Sinasan toy. Impreso en el convento de San Pablo de Manila por Antonio Damba i Miguel Seixo. Año de 1621. Manila. Méntrida, Alonso (2004 [1637]): Vocabulario de la lengua bisaya, hiligueina y Baraya de la isla de Panay y Sugbú y para las demás islas. Edición y estudio del Dr. Joaquín García-Medall en 2004. Valladolid: Universidad de Valladolid. [1818, 1841, 1894] San Buenaventura, Pedro (1994 [1613]): Vocabulario de lengua tagala. El romance castellano puesto primero. Primera, y segunda parte por fr. Pedro de San Buenaventura, inútil e indigno religioso franciscano descalzo. Dirigido a D. Ivan de Silva cavallero del orden de Santiago governador y capitán general de estas islas, y presidente de su Audiencia y Chancillería Real. Con licencia impreso en la noble villa de Pila, por Thomas Pinpin, y Domingo Loag tagalos. Año de 1613. Valencia: París-Valencia. Vivar, Pedro (1797): Yloco en romance compuesto por diferentes padres ministros antiguos, diestros en este idioma y últimamente corregido, y añadido segun lo que ahora se uso y de ultima mano, por el padre fray Pedro Vivar, de la orden de nuestro gran padre san Agustin, ministro en esta provincia, visitador, que ha sido de ella, examinador synodal de este obispado de Nueva Segovia y prior vocal del comvento de Bantay. (ms)

62

Rebeca Fernández

B. Secondary sources Hernández, Policarpo (1998): The Augustinians in the Philippines. Makati: Colegio San Agustín. Henríquez Salido; do Carmo, Mº & Esparza Torres, Miguel Ángel (eds.) (1999): Estudios de historiografía lingüística hispánica ofrecidos a Hans-Josef Niederehe, Tréveris, 16 a 17 de junio de 1997. Hamburg: Helmut Buske Verlag. Lope, Blanch (1990): Los indoamericanismos en el Tesoro de Covarrubias, in: Estudios de historia lingüística hispánica. Madrid: Arcolibros, 153–174. Quilis, Antonio (1997): Sobre las costumbres de los antiguos Philippine, in: Homenaje al profesor A. Roldán Pérez, Volumen II. Murcia: Universidad de Murcia, 477–488. Rubino, Carl (2000): Ilocano dictionary and grammar. Honolulu: University of Hawai‗i Press. Sueiro Justel, Joaquín (1999): Panorama de la lingüística española de Filipinas, in: Estudios de historiografía lingüística hispánica ofrecidos a Hans-Josef Niederehe, Tréveris, 16 a 17 de junio de 1997. Hamburg: Helmut Buske, 113–121. Sueiro Justel, Joaquín (2003): Historia de la lingüística española en Filipinas (1580–1898). Lugo: Axac. Vela, Gregorio de Santiago (1931): Ensayo de una biblioteca iberoamericana de la orden de san Agustín. El Escorial: Imprenta del monasterio. Villoria Prieto, Carlos (1997): Un berciano en Filipinas: Alejandro Cacho de Villegas. León: Universidad de León.

OTTO ZWARTJES (AMSTERDAM/NETHERLANDS)

Oyanguren de Santa Inés‘s grammar of Tagalog (Tagalysmo Elucidado 1742): towards a reconstruction of 18th century reflections on comparative typology

Abstract This paper analyses descriptions of three Asian languages: Chinese, Tagalog and Japanese by the Spanish Franciscan Melchor Oyanguren de Santa Inés. Although the author did not aim at a systematic analysis of these three languages, we find a great number of examples and analysis, where the author attempts on the one hand to contrast Asian languages with European languages, such as Spanish, Latin, Greek, and on the other hand he demonstrates that there are also important typological differences between these Asian languages. For the period in question, it is noteworthy that the author sees similarities between Japanese and his own native tongue, Basque, both agglutinative languages. This paper demonstrates that Oyanguren did pioneering work on the field of comparative typology antedating the achievements of Lorenzo Hervás y Panduro and Wilhelm von Humboldt.

1. Introduction Melchor Oyanguren de Santa Inés (1688–1747) was born in Salinas (Guipúzcoa, Spain) in 1688. He arrived in the Philippines in 1717. After having worked as a missionary in Southeast Asia (Cochinchina) he returned to the Philippines in 1725, where he became ‗doctrinero‘ in the village of Los Baños de Aguas Santas in 1726 and ‗Minister‘ in Saryaya until 1736. For health reasons he decided to return to Spain, but never arrived there. On his journey back to Europe he passed Mexico where he became President of the Hospicio del Convento de San Agustín de las Cuevas (today in Tlalpan). He died there in 1747. His Japanese grammar entitled Arte de la lengua japona, dividido en quatro libros segun el arte de Nebrixa, con algunas voces proprias de la escritura, y otras de los len

I wish to acknowledge Thomas Stolz, the University of Bremen and the Amsterdam Center for Language and Communication (ACLC) of the University of Amsterdam for the support I received which enabled me to participate at ICHOLS XI. This article is an elaborated version of two papers presented at the Second International Conference on Missionary Linguistics in São Paulo (March 2004) and the XIth ICHOLS in Potsdam (August 2008). Participation at the University of São Paulo was financed by the University of Oslo and the Norwegian Research Council (Norges Forskningsråd). This article was completed while holding a fellowship at the Netherlands Institute for Advanced Studies (NIAS) in Wassenaar, the Netherlands. Thanks to Prof. Lawrence A. Reid this paper has been much improved, though needless to say, I alone will accept responsability for any infelicities which could remain.

64

Otto Zwartjes

guages de Ximo, y del Cami, y con algunas perifrases, y figuras was published in Mexico (Joseph Bernardo de Hogal) in 1738 and his grammar of Tagalog appeared in 1742 under the title Tagalysmo elucidado, y reducido (en lo posible) a la Latinidad de Nebrija. Con su sintaxis, tropos, prosodia, pasiones & c. y con la alusión, que en su uso y composición tiene con el Dialecto Chinico, Mandarin con las lenguas Hebrea y Griega, also printed in Mexico (Francisco Xavier Sánchez). Two works are lost, his grammar of Basque and a trilingual dictionary Tagalo-Castilian-Basque (Zwartjes 2010: 9). The most important sources for his Japanese Arte were the grammar of Japanese written by the Spaniard Diego Collado (ý1638) – who was inspired mainly by the work of the Portuguese Jesuit João Rodrigues (1562–1633) and the anonymous Vocabulario de Japón declarado primero en Portugués por los padres de la Compañia de Jesús de aquel Reyno, y agora en castellano en el Colegio de Santo Thomás de Manila, printed in 1630 in Manila (Colegio de Santo Tomás), a Spanish translation of the Vocabulario da Lingoa de Iapam com a declaração em Portugues, feito por alguns Padres, e Irmãos da Companhia de Iesu completed in 1604 in Nagasaki. As the titles indicate, Antonio de Nebrija (c. 1441/44?–1522) served as the linguistic framework which shaped his analysis of the Japanese and Tagalog data. Before Rodrigues, the Jesuit Duarte da Silva (1536–1564) wrote an Arte da Lingua Japoneza and a Vocabulario da Lingua Japoneza which have been lost (Streit 1929, 380). After the expulsion of the Christians from Japan many Christian books were burnt, although some of them were possibly brought to the Philippines. Even before the expulsion, Japanese teaching manuals were brought to Manila, as we can read in the letters written by Alessandro Valignano (1539–1606) (cf. Moran 1993: 157). The Tagalysmo elucidado, mentions some predecessors who served as the main sources of his data, sometimes called ―los primeros Tagalistas‖ (1742: 11), and in other cases they are mentioned by name: Francisco Blancas de San José (1560–1614), Oliber de N.H.S. Gregorio de N. H. Sta Anna, not identified yet, and Gaspar de San Augustín (1646–1724) who wrote a short compendium of Tagalog. Blancas de San José has the nickname ―El Cicerón Tagalog‖1 since he wrote one of the most important pioneer works of Tagalog, although he was not the first. We have reports that around 1580, an Arte del idioma Tagalog was written by Juan de Plasencia Portocarrero (ý1590). According to others, the first grammar of Tagalog was written by Agustín de Albuquerque (ý1581) (Sueiro Justel 2007: 84), and other works are written in the period before Blancas by Bernardino de Jesús (ý1604), Juan de Quiñones (1551–1587) (Streit 1929: 332, 363, 366). We cannot exclude the possibility that Oyanguren also had access to these works, and that these authors are some of the ―primeros Tagalistas‖.2

1

2

In other sources he is called Aquiles de esta dificultad [The Achilles of this difficult (language)] (San Augustin 1787 [1703], Prólogo al lector), or Demosthenes de esta lengua [The Demosthenes of this language] (1787 [1703]: 153). There is documentation that more grammars were composed in this period. See Sueiro Justel (2007: 77–87) for more details.

Oyanguren de Santa Inés‘s grammar of Tagalog

65

As we have demonstrated in previous studies, Oyanguren‘s grammars are particularly interesting since he includes a great number of different languages in his examples and linguistic ‗theories‘. Contemporary authors, such as San Augustín, only describe Tagalog and no other Asian languages are mentioned. Oyanguren‘s work was not written exclusively for the learning of one particular language; he obviously had the purpose of explaining to learners of Japanese, Chinese, and Tagalog some important features of the other languages. Nevertheless, Oyanguren was not the only polyglot missionary. It has been documented that Juan de San Pedro knew Visayan and Tagalog. Jordán de San Esteban, a missionary of Sicilian origin, was educated in Spain and knew his native tongue (Sicilian) Italian, Spanish, Chinese, Latin, Greek, and the Philippine language spoken in ―La Nueva Segovia‖. Father Joseph de Madrid knew Cebuano, Visayan, Tagalog, the ‗language of Ytuy‘, Chinchea (Southern Mĭn-Chinese), Italian and Portuguese, and Esteban Ortiz spoke the ―Mexican language‖ (probably Nahuatl), Tagalog, Ilocano and Chinese sufficiently well to enable him to confess and preach in them (Quilis 1997: 14). As we can see, it was not at all uncommon for missionaries to speak or understand more than one indigenous language, but unfortunately no grammars or dictionaries written by these priests survived. Linguistic typology or comparison was generally not considered an object of serious study nor as a useful tool in the process of language learning. The Japanese grammar of Oyanguren does not have any indication that the author benefited from native speakers of the Japanese minority in Manila; probably he took all his linguistic data from written sources, mainly the grammar of Collado and the Spanish version of the Vocabulario,3 creating a completely different work with a different structure and approach. For the Chinese material, we have shown (Klöter & Zwartjes 2008) that Oyanguren relied heavily on the Mandarin grammar of Francisco Varo (1627– 1687), since we find his name in a reference where he advises readers, particularly the more advanced learners, to consult his work as we can read in the following citation: Por quanto en este Arte Tagalog, vá como encajada la substancia del Arte Chinico, ó Mandarin; declinaciones, composicion nominal, y conjugacion, y aver en él muchos vocablos, debajo de dos advertencias se insinúa su prolacion, por no dilatarme, pues se puede recurrir á los Artes del P. Varo, de la Sagrada Religion de N.P. Santo Domingo: ò à los del P. la Piñuela, ó Fr. Basilio Glemona Recoletos de Nuestro Padre San Francisco [Chinese or Mandarin grammar is incorporated in this grammar of Tagalog; i.e. declensions, nominal composition, and conjugation. Since there are many words included in it, I only allude to the meaning of them in a handful of observations, in order not to dwell on these matters, because one can fall back on the grammars of the Dominican Father Varo, or those of the Recollect Franciscan Fathers la Piñuela or Basilio Glemona] (Oyanguren 1742: 8) 3

―… porque este libro sigue el methodo del Vocabulario en las activas, y passivas …‖ [because this book follows the method of the Vocabulario in its treatment of active and passive …] (Oyanguren 1742: 140).

66

Otto Zwartjes

The sections on Chinese are embedded in the Tagalog grammar but the author does not tell his readers explicitly that he checked the linguistic data with native informants of the Chinese minority in Manila. The Tagalog linguistic data were also based on written sources, but opposed to the Japanese and Chinese data, the Tagalog material was checked using native speakers of the languages, as we can infer from the following citation: Los Indios (à quienes yo he preguntado, y examinado sobre este assumpto, y se suelen enbobar, respondiendo ya cesta, ya ballesta) fuera de tener raïzes generales, especificas, e individuales para semejantes oraciones, necessitan de mucha reflexion para la intelegencia de tanta reduplicacion [I have asked and examined this matter with the Indians, but they are usually ignorant; they give me answers to my questions haphazardly. Their language consists of ‗general‘, ‗specific‘ and ‗individual‘ roots for similar clauses and the Indians reflect lengthily on these matters in order to comprehend so many redundancies] (Oyanguren 1742: 141). In Antiquity, Greek grammarians were usually not interested in languages other than Greek. They developed a descriptive model with the objective of describing their own language. In the tradition of Latin grammars, some isolated approaches to languages other than Greek and Latin are documented,4 but Roman grammarians were also mainly interested in their own language, Latin. The difference between Greek and Roman grammarians is the fact that the latter added another dimension to their grammatical theory. Although Latin-Greek bilingual grammars were not common, Greek served as the main source of inspiration, so that Latin grammars were no longer strictly monolingual, as the Greek grammars were. There is documented evidence that during the Middle Ages, missionaries were interested in foreign languages. Some grammars and dictionaries survive as manuscripts, such as the Codex Cumanicus, describing a Turkic language, and Persian. Humanists in Renaissance Europe added another dimension to their linguistic interests, mainly the study of Semitic languages. In the 16th century, grammars of Aramaic, Hebrew and Arabic appeared and around the same period, missionaries started to describe a great number of non-Indo-European languages all over the world. Linguistic typology was not an established discipline, but authors often attempted to explain similarities and differences between languages from a broader perspective. In pioneering studies on linguistic typology, the tripartite division between isolating, agglutinative and inflectional languages was introduced by August Wilhelm von Schlegel (1767–1845), the brother of the Sanskritist Karl Wilhelm Friedrich von Schlegel (1772–1829), according to Auroux (2000: 18), who divided languages into those that are ‗organic‘, like Sanskrit, Persian and those of Europe, and on the other 4

Cf. Marcus Terentius Varro‘s (born 116 B.C.) section in his De lingua latina, where the languages of the Phoenicians, the Egyptians and the Gauls are mentioned. According to Varro, these languages have words with ‗only one case-form‘ (―Quare si essent in analogia, aut Phoenicum et Aegyptiorum vocabula singulis casibus dicerent‖) (Zwartjes 2008: 407).

Oyanguren de Santa Inés‘s grammar of Tagalog

67

hand ‗mechanical‘ languages, like Chinese, Coptic, Basque and Amerindian languages (Koerner 1990: 250, Errington 2008: 73). Wilhelm von Humboldt distinguished four types: (1) isolating (―with no grammatical structure‖), (2) agglutinating, (3) incorporating, or polysynthetic and (4) flectional (Errington 2008: 76). Melchor Oyanguren de Santa Inés‘s main purpose was no different from that of other missionaries of this period: the teaching of these languages to missionaries. Nevertheless, being a real ‗aficionado‘, his grammatical descriptions are accompanied by typological descriptions and comparative research. Even though Oyanguren made a clear distinction between the Indo-European languages he was familiar with (mainly Spanish) and Asian languages (Japanese, Chinese and Tagalog), yet he explained the most important typological differences between these Asian languages themselves, and in several sections he concluded that Japanese and his native tongue Basque – both agglutinative languages – shared some common typological features. Finally, we find observations on Greek, Hebrew, ―the language of Siam‖, and Nahuatl. In Zwartjes (2008) Oyanguren‘s views on Hebrew are analysed and in Klöter & Zwartjes (2008) we concentrated on his Chinese material (lengua chínica, dialecto chinico-mandarín, lengua caraya, mandarin, lengua de los sangleyes). In this article we shall analyse his observations on Tagalog, in relation to his theories on Asian languages in general. Zwartjes (2010) contains three volumes; a facsimile edition of the two grammars and a monograph.

2. Asian languages 2.1. Chinese and Japanese The Jesuit Francis Xavier (1506–1552) is one of the first missionaries who mentioned the difference between Chinese and Japanese. The languages were not mutually intelligible but he concluded that ―when they write they understand each other only by writing, for they know the signification of the ‗letters‘ [i.e. ‗characters‘], but the pronunciation always remains different‖ (Nachod 1922: 257,5 cited in Schreyer 1992: 5). In some sections of his grammar, Oyanguren points out that Asian languages are ―very different‖: 5

Oscar Nachod (1858–1933): ―Die ersten Kenntnisse chinesischer Schriftzeichen im Abendlande‖[The earliest acquaintance of Chinese characters in the West]. Hirth Anniversary Volume 1922, 235–73. London: Asia Minor. Compare also the observation of Nicolas Trigault (1577–1628) cited from the English translation entitled China in the Sixteenth Century. The Journals of Matthew Ricci: 1583–1610. 1953 [1615]. New York: Random House: ―For instance the Japanese, the Koreans, the people of Cochin, and the Leuhians have books which are common to all, but they differ so widely in their spoken languages that no one of them can understand the others. […] I have heard that over and above the symbols which they have received from the Chinese, the Japanese also have an alphabet and certain elements similar to our own, which enables them to write their own vernacular without the endless series of Chinese ideographs‖, cited in Schreyer (1992: 16).

68

Otto Zwartjes

En las mas de las lenguas assiaticas, aunque son muy distintas, el regimen de los verbos es muy semejante: el verbo substantivo xi, ser: rige dos nominativos como en la latinidad quando se los ponen: v.g. Francisco xi haò jîn. Francisco es buen hombre…[In most Asian languages – although they are very different from each other – the governance of the verb is quite similar; the substantive verb xi ‗to be‘ governs two nominatives, as occurs in Latin, if they are present: e.g. Francisco xi haò jîn, ―Francis is a good man‖]. (Oyanguren 1742: 200–201) Oyanguren‘s references to ‗Chinese‘ display some subtle, but nonetheless significant differences. In many instances, his wording is ―the language of China, or the Mandarin dialect‖ (la lengua de China, ó dialecto mandarín). But then he also mentions the ―Chinese dialects‖ (dialectos Chinos). This indicates that he had some idea of China‘s linguistic diversity. However, data from Chinese dialects other than Mandarin did not enter his analysis. According to Oyanguren, the Chinese language is a ‗brief and easy‘ language, although he admits that the use of the tones is difficult. … en este breve resumen, con un poquito de comunicacion con ellos, para la comprehension de las tonadas; puede con facilidad el Missionario hacerse capaz de su lenguage, y dâr á muchas Almas esse Consuelo; pues la Lengua Caraya es breve, y facil; aunque sus frases, y tonadas son dificultosas [… with this brief summary and some exercise in communicating with them (i.e. the Sangleyes‘, or Chinese speaking minority in Manila), the missionary will be sufficiently competent to understand the tones easily and able to give Consolation to these Souls; alter all, the Caraya (probably Oyanguren means ―Cathay‖) language is brief and easy, although its sentences and tones are difficult‘] (Oyanguren 1742: 3)

Oyanguren mentions some characteristic features of Japanese, comparing and contrasting this language with Chinese, Tagalog, and Basque. The following citation illustrates this strategy (for more details concerning Japanese see Zwartjes 2008, 2009 and 2010). La lengua Japona es invariable, y no tiene casos, en lo qual imita al dialecto Chino, ô Mandarino, que no tiene casos, aunque no le imita en las particulas, porque la lengua Japona tiene distintas, y muchas particulas; y la lengua China tiene pocas, ut sieng, seng: seng tien: tie, men &c. solo se advierte, que la lengua Japona en la mayor parte es subjuntiva como lo es nuestra lengua Vascongada. De qua egimus in sua arte [The Japanese language is invariable and does not have cases, resembling the Chinese or Mandarin dialect, which has no cases either, although it does does not resemble Chinese as far as the particles are concerned, since the Japanese language has many different particles, unlike the Chinese language, which has few, e.g. sieng, seng: seng tien: tie, men, etc. It has only to be observed that the Japanese language is predominantly subjunctive, as our Basque language, De qua egimus in sua arte (―of which we dealt in the grammar of this language‖)] (Oyanguren 2009 [1738]: 6)

Oyanguren de Santa Inés‘s grammar of Tagalog

69

Es el Dialecto Tagalog muy abundante, sonòro, elegante, y que en la mayor parte imita a la pronunciacion Europea, aunque no es su pronunciacion dental como es la Cantabrica, y tiene, y consta de pronunciacion guttural mas Sonora, que la Arabiga, y menos dificil que la China, ò Dialecto Mandarin, sin la molestia de sus tonadas: en lo qual imita el ñgna in gangoso del Hebreo [The Tagalog language is very abundant, sonorous and elegant. Its pronunciation resembles grosso modo that of European languages, although its pronunciation is not as dental as the Cantabrian (=Basque) language; its pronunciation is more guttural and more sonorous than Arabic, and less difficult than the pronunciation of Chinese or Mandarin, without the trouble of all those tones, resembling the nasal twanging of Hebrew as far as the sound ñgna in is concerned] (Oyanguren 1742: 1–2)

2.2. Austronesian languages According to Auroux (2000: 16) the birth of comparative linguistics in Austronesian languages has to be dated to the beginning of the 18th century, although the sources of this period are not mentioned. The most important works, according to Auroux (ibid.) are the 19th century works Ueber die Kavi Sprache der Insel Java (1830–1835) of Wilhelm von Humboldt, published by Buschmann, Dumont d‘Urville‘s (1790–1842) Apologie (1833–1834) and Pierre Louis Jean Baptiste Guassin (1821–1886)‘s Du dialecte de Tahiti des Isles Marquises et en general de la langue polynésienne (1853). Although the Spanish missionaries did not write comparative studies on the relationship between Austronesian languages, they report from the earliest period that the languages of the Philippines belong to the same linguistic ‗Malayan‘ family,6 although they were also aware that the languages they encountered were often mutually unintelligible, as we can read in the following citation: Las gentes de lugares poco distantes no se entienden los unos con los otros…. En algunas partes observamos que en la boca de un río se hablaua una lengua, y en el naçimiento dél otra. Cosa que es de grande estoruo para la conuersion y enseñanza de estas gentes [People who are living in close proximity do not understand one another… In some regions, we observed that one language was spoken in the estuary of a river, while in the regions along the upper course of the same 6

By comparison: One of the first sources which describe the great similarity between Semitic languages, such as Aramaic and Arabic is Yĕhudah Ibn Qurayš‘s (9th–10th century) Risāla, (cf. Sáenz-Badillos 2005: 431; Téné 1980: 363; Zwartjes 1997: 17). The famous talk of William Jones (1746–1749) of 1786 is often seen as the ‗birth of comparative philology‘ (Seuren 1998: 80), although two centuries earlier Thomas Stephens (ca. 1549–1619) had already discovered the similarities between the Indo-Aryan languages in India, Latin and Greek. Similarities between languages of the Dravidian family are first described by Robert Caldwell (1746–1794) and Francis Whyte Ellis (1777–1819) (cf. Sreekumar 2009).

70

Otto Zwartjes

river, another was spoken, causing many obstacles in converting and teaching these peoples] (P. Colín, 1651, cited in Quilis 1997: 12) Most authors of this period probably considered Chinese to be a difficult language, probably based on its tonal phonology and writing system. Compared to Chinese, the Philippine languages were often labeled as ‗easy‘. The local writing system called baybayin is not complex and the phonological system is indeed not so complicated, compared to Chinese. Compared to other Philippine languages, Tagalog was considered the most difficult language, as some authors conclude: I a la verdad estas lenguas no son muy difíciles, ni de aprenderlas ni de pronunciarlas. Mayormente agora que ai arte, i vocabulario; y mucho escrito en ellas. La más difícil es la de Manila (que llaman Tagala) [In fact, these languages are not very difficult to learn nor to pronounce, mainly because there exists a grammar and a dictionary, and much has been written in them. The most difficult one is the language of Manila, called Tagalog] (Pedro Chirino 1604, cited in Quilis 1997: 14) According to Blancas de San José, Tagalog was the most ‗universal‘ language of the Philippines, which does not surprise us, since it was the language spoken in the Manila area, which was the centre of political power and was the primary medium of communication between the colonial powers and the native population. (―Tagala, que es la mas universal desta tierra y la que se usa entre los indios alrededor de la ciudad por muchas leguas‖, cited in Quilis 1997: 17). Gaspar de San Agustin observed that the languages of the Philippines are actually ‗dialects‘, which are not entirely different from one another, since they all belong to the same family of ‗Malayan‘ languages. De todas me he valido, y de otras de lenguas de estas Islas, porque ciertamente no son totalmente diversas entre si; sino varios Dialectos de un a misma lengua, que dicen sèr la Malàya; como en la Griega los Dialectos Attico, Ionico, Eolico, y comun. Y de la Latina las lenguas Italiana, Española, Portuguesa, y Francesa. Y en el Norte los Dialectos de la lengua Godt, ò Teutonica [I have made use of all (grammars) of this language (=Tagalog), and even of other languages of these islands, since they are not totally different from one another, but they are different dialects of the same language, which they say is Malayan; as occurs in the Greek language, where Attic, Ionic, Aeolic and koiné varieties are distinguished…] (Gaspar de San Agustin 1787 [1703]: Prólogo al lector, no numbered pages) As Oyanguren mentions Gaspar de San Agustin as one of his sources, it is not impossible that the following fragment was inspired by him: Aunque de la Torre de Babel salieron tantas lenguas matrices, el dia de oy ay muchas mas por los distintos dialectos, que se han originado de las mesmas matrices, y por la comunicación de unas lenguas con otras: y para probar esto, no necessi-

Oyanguren de Santa Inés‘s grammar of Tagalog

71

tamos recurrir â dialectos Atticos, Jonico, Dorico, Eolico de la lengua griega, quando en el continente del dominio Español, tenemos dialectos del Vascuenze: en la America con la lengua Mexicana, y Peruana; en las Islas de Luzòn con la lengua Malaya, ô Tagala. Y finalmente lo mesmo podemos decir de los distintos dialectos que ay en el Imperio de la gran China, sus provincias, y Reynos adyecentes, en cuyos dialectos tan diversos experimentan los Missionarios Apostolicos el ―hoc opus, hic labor‖. El dialecto Nison, ô Japòn es dialecto, que procediò del Imperio Chinico, como tambien sus habitadores, segun consta de sus historias: es dialecto, que se divide en otros muchos dialectos, segun distintos Reynos, que ay en el Japòn: es mas puro, menos dificil, que los dialectos Chinicos, pues estos con el dominio Tartaro han dejado introducir en sus dialectos distintas voces, tomadas [sic], guturales, &c [Although so many languages originated from the Tower of Babel, there are nowadays many more languages because of the different dialects which are derived from the same matrix languages, and because of the communication between the speakers of these languages. It is unnecessary to prove this on the basis of Greek dialects, such as Attic, Ionic, Doric, or Aeolic, since we find such variation in the realm of Spain, where dialects exist of Basque, and in America we find Mexican and Peruvian dialects, and the same occurs with the Malayan or Tagalog language in the Islands of Luzon. And finally, we can say the same of the different dialects which exist in the Great Chinese Empire, in its provinces and contiguous Kingdoms, where the Apostolic Missionaries encounter a great variety of dialects, experiencing ―hoc opus, hic labor‖ (this work, this labour‖). The Nison, or Japanese dialect originates from the Chinese Empire, as the inhabitants of Japan, as has been established in the histories. It is a dialect which can be divided in several other dialects, according to different Kingdoms in Japan. Japanese is purer and less difficult than the Chinese dialects, since the Tartars have introduced several words, tones and guttural sounds in these dialects during the period of their occupation]. (Oyanguren 2009 [1738]: Prólogo al lector) And in his Tagalysmo elucidado we read also that Tagalog belongs to the ‗Malayan‘ family. El Dialecto Tagalog, ó Taga ylog es dialecto de la Lengua Malaya segun se infiere de las Historias, y Conquistas primeras, que escribieron los primeros Españoles: llamase este Dialecto Tagalog, por los Indios Tagalos, que la hablan, y porque es la mayor parte de los que habitan entre rios, que esto es lo que quieren decir estas dos voces, ò una voz compuesta de Taga, y de ylog; hoc est, natural del rio: este Dialecto se divide en otros muchos como en el de Camarines, Pampango, &c. y sirve como de Matriz à los otros Dialectos, no obstante las muchas diferencias de vocablos, y mutacion de letras, que ay en todos los Dialectos, y la introduccion de muchas voces nuebas, con la comunicacion de otras

72

Otto Zwartjes

Lenguas, y Dialectos, y olvido de las voces naturales, y proprias; calamidad de todas las Lenguas, que procedieron de la Torre de Babel, que unas se destruyeron del todo; otras se mezclaron con otras distintas, y otras se dividieron en distintos Dialectos [The Tagalog (or Taga ylog) dialect is derived from the Malayan language, as we can infer from the histories of the first Conquests, composed by the first Spaniards. This dialect is called Tagalog, a name derived from the Tagalog Indians who speak this language, and because its speakers live mainly between Rivers – which the two words Taga and Ylog in fact mean – forming the compound Tagalog; i.e. ‗naturals from the river‘. This dialect is divided into many others, such as the dialect of Camarines, Pampango, etc. and serves as a matrix language for other dialects, regardless of the large discrepancies between the words, the changing of letters – which occurs frequently in these dialects – and the introduction of new words, as a consequence of contact with other languages and dialects, and the oblivion of their own natural words, a catastrophe which affects all languages which derive from the Tower of Babel; some languages are extinguished entirely, other are mixed up with several others, and others are split up into different dialects] (Oyanguren 1742: 1) This means that the ‗discovery‘ of the Malayo-Polynesian language family (or at least the relationship between the western branches of the family) antedates the so-called ‗discovery‘ of the Indo-European language family (although it was still unknown that Malayo-Polynesian was actually a major branch of the Austronesian family).

3. Oyanguren‘s description of Tagalog 3.1. Word classes In Spanish (or in English) there is a clear correspondence between morpho-lexical and syntactic categories. Morphological properties of the noun are usually clearly distinguishable from those of verbs. Many Asian languages do not have such a clear-cut division. According to Himmelmann (2005: 361) content words in Tagalog ―do not have to be subclassified with regard to syntactic categories and they have all the same syntactic distribution, i.e. they all may occur as predicates, as (semantic) heads of noun phrases and as modifiers‖. Oyanguren‘s knowledge of the ‗root‘ from Hebrew grammar probably helped him to find a solution for this descriptive problem. According to traditional Greco-Latin grammar, nouns have declensions and cases, opposed to verbs which have conjugations and tenses. In Renaissance Western grammars of Hebrew, the concept of the root (Latin ‗radix‘) was introduced and the ‗radical‘ (also ‗substantial‘) letters determined the meaning (the ‗semanteme‘), as opposed to the ‗servile‘ (or ‗accidental‘) letters that do not have a ‗meaning‘ (see Zwartjes 2008). Oyanguren was not the first

Oyanguren de Santa Inés‘s grammar of Tagalog

73

who introduced the term raíz ‗root‘ in a grammar of Tagalog, since we find a detailed definition in the grammar of Blancas de San José. Nevertheless, the comparison between Tagalog and other languages are innovations by Oyanguren: La lengua Tagala imita en su verbalizacion á la lengua Hebraica, pues tambien suele verbalizar sus nombres, como consta del Cap. 9. Genes. Iaphet elohim leiaphet: Mas la lengua Tagala excede en este supuesto a muchas lenguas, pues apenas ay parte de oracion Tagala, que no se pueda verbalizar segun la significacion, que se le quisiere dar, atendiendo siempre â la genuina inventiva de la lengua, que es metaphysica, y muy discreta [Tagalog resembles Hebrew regarding its ‗verbalisation‘, since Hebrew also verbalises its nouns, as can be demonstrated by the example Cap. 9. Genes. Iaphet Elohim leiaphet. Howver, Tagalog exceeds many other languages in this respect, since there exists hardly any part of speech in Tagalog which cannot be verbalised, according to its meaning one wishes to give to it, maintaining the genuine inventiveness of this language, which is metaphysical and very discrete] (Oyanguren 1742: 72) Supongo lo tercero, que las vozes Tagalas son indiferentes para verbalizarlas, nominalizarlas, y adverbializarlas, …. y se acomodan, y asimilan â nuestra lengua Cantabrica Española, ó al Dialecto Japon, aunque aliter, & aliter [In the third place, I suppose that Tagalog words are indifferent (not morphologically marked as a noun, verb, etc.] and can be verbalised, nominalised or adverbialised, and they adjust and resemble our Spanish Cantabrian language or the Japanese dialect, although there are differences] (Oyanguren 1742: 174) In the following citation from Oyanguren we read: Es el verbo segun Nebrija una parte de la oracion, que tiene modos, y tiempos, y no tiene casos: esto se entiende del Verbo latino, porque en el Tagalysmo ay muchissimas raìzes indiferentes, (como tambien en el Dialecto Mandarin Chinico) y aunque algunas raízes ex parte modi significandi parezcan verbos, vt, labas pr. sacar, ò salir: labis pp. derramarse, &c [According to Nebrija, the verb is a part of speech which has modes and tenses and lacks case; this is what we understand from the Latin verb, because in Tagalog there are many indifferent ‗roots‘ (as occurs also in the Mandarin Chinese dialect), although some roots resemble verbs ex parte modi significandi (=by meaning) e.g. labas, ‗to take‘, or ‗to leave‘: labis ‗to drain away‘, etc.] (Oyanguren 1742: 71) In modern grammar, apart from the word ‗root‘, often the concept of ‗base‘ or ‗stem‘ is used. A corresponding term existed in the period in question (‗thema‘), but Oyanguren does not use it. Schachter & Otanes (1972: 355–356) distinguish 1) simple bases, 2) derived bases, 3) compound bases and 4) phrasal bases. The first class, the simple bases can generate different word classes, as in the following examples:

74

Otto Zwartjes

kurbata ‗tie‘ > magkurbata ‗wear a tie‘ ganda ‗beauty‘ > maganda ‗beautiful‘ > (Schachter & Otanes 1972: 355–356)

gumanda ‗become beautiful‘

Probably, Oyanguren had such morphological processes in mind when he claimed that ‗roots‘ can be verbalized, nominalized, etc. Derivation is no longer seen as a morphological process from one part of speech to another (nouns from verbs, verbs from nouns, etc.), but as verbs, nouns, etc., from ‗roots‘.

3.2. Inflection According to Oyanguren, Asian languages do not have the same inflectional morphology as Latin or Spanish. In the following sections we shall analyse a selection of definitions related to the Greco-Latin ‗nominal declensions and verbal conjugations‘. (Remember that the author is attempting to ‗reduce‘ Tagalog to Latin grammar, as far as this was possible: ―en lo possible‖).7 3.2.1. Nominal declension, according to Oyanguren 3.2.1.1. Case An important observation made by the author in the beginning of his grammar is that Tagalog does not have ‗cases‘, as occurs in Hebrew and Castilian, but, instead of them, in this language certain ‗particles‘ are used, which must be placed before the nouns: Declinacion no es otra cosa ex vi nominis, que el apartamiento de los casos obliquos, del recto, ò nominativo, á quien los Logicos conocen por verdadero nombre. Los Tagalos imitan â los Hebreos en no tener casos proprios; y se asemejan al Dialecto Castellano, en añadir anteponiendo unas particulas, ò articulos á los nombres, ò voces, para distinguir, y hacer sus casos [Declension is ex vi nominis nothing else than the setting apart of the oblique cases (=inflected) from the casus rectus, or nominative, which the Logicicians label as ‗the real noun‘. Tagalog nouns resemble their Hebrew counterparts in not having proper cases, 7

―El motivo de escrebir este Arte (aviendo tantos) es el nivelarlo en lo possible à la latinidad, poniendo reglas fixas, evitando con esto la confusion de tantos Artes, unos cortos, otros largos, otros muy confusos, pues à penas avrá Arte, que se parezca á otro; porque unos trasladan, y juntan de distintos Artes, otros mudan lo trasladado, y assi ay tantos Artes quantas cabezas, quot homines tot artes‖ [The reason why I have composed this gramar, among others, is the aim to equate this language to Latin, putting fixed rules, avoiding confusion which is present in so many other grammars, some of them brief, others voluminous, some of them confusing, since there hardly exists a grammar which resembles any other; some grammarians copy and put together from different grammars, others emend what has been copied, and thus, there are as many grammars as individuals. (―quot homines tot artes‖)] (Oyanguren 1742: 2).

Oyanguren de Santa Inés‘s grammar of Tagalog

75

and they are equivalent to the Castilian Dialect regarding the preposition of some particles or articles before the nouns, or words, for the distinction and formation of their cases] (Oyanguren 1742: 15)8 3.2.1.2. Number Nouns in Tagalog can be pluralized explicitly, but ―the pluralization of a noun need not, and in some cases, in fact, cannot be formally signaled if the context makes the plural meaning clear. Without context the sentence Nasa mesa ang libro means ‗The book is on the table‘. But the same sentence can be used to mean ‗The books are on the table‘ in a situation where both the speaker and the hearer know that more than one book is being referred to. (The insertion of the word mga before libro would make the plural meaning explicit...)‖ Schachter & Otanes (1972: 111). In Oyanguren‘s words: Aunque los Tagalos no tienen concordancias de nominativo, y verbo; por no tener los verbos inflexion de personas imitan a los Japones, y Chinos en no tener personas el verbo y suelen tener alguna indiferencia en la cõcordancia del sustantivo, y adjetivo; pues tal vez concuerdan; y tal vez las vozes son tambien indiferentes para singular, y plural [no obstante tienen articulos de pluralidad] [In the Tagalog language, concord between the nominative and the verb is non-existent, since their verbs do not inflect in person, resembling Japanese and Chinese, which do not mark person in their verbs either. In Tagalog there is generally some indifference in the concord between the substantive and the adjective, since in some cases they agree, and occasionaly words are also indifferent for singular or plural (although some articles of plurality do exist] (Oyanguren 1742: 34) Oyanguren compares this feature with Basque, where nouns are strictly invariable for number, for example, etxe means indifferently ‗house‘ or ‗houses‘ (number can be understood by the use or absence of quantifiers or determiners). In English there is obligatory agreement in number between the subject and predicate of a sentence: The man is a lawyer/The men are lawyers. In Tagalog there is, in general, no obligatory number agreement between the predicate and the topic.9 As we can see, Oyanguren‘s description is fairly correct.

8 9

Cf. his grammar of Japanese: (Oyanguren 2009 [1738]: 4). The term ‗topic‘ is used as defined by Schachter & Otanes (1972: 60): ―The topic expresses the focus of attention in the sentence. In many cases it corresponds to the subject of an English sentence … But there are also many sentences in which a Tagalog topic is not translatable by an English subject, and, conversely, in which an English subject is not translatable by a Tagalog topic‖. There are, however, a few cases of obligatory number agreement, as well as of obligatory absence of number agreement (cf. Schachter & Otanes 1972: 111ff.).

76

Otto Zwartjes

3.2.1.3. Gender One of the differences between European and Asian languages is the gender system. In Tagalog and in Mandarin, masculine, feminine, and neuter genders are not distinguished. Oyanguren postulates the existence of the dichotomy ‗grammatical‘ versus ‗philosophical genders‘, which is obviously an important and creative solution. Esta lengua es como nuestra lengua Vizcaìna, ó como la lengua tagala, ó como la lengua de China, ó dialecto mandarino, que no tienen generos masculinos, femeninos, &c. para las concordancias, como sucede en la lengua Latina, y su dialecto la castellana, ó en los dialectos de la lengua griega: por lo qual careciendo de generos grammaticales, algunas de estas lenguas tienen sus generos philosophicos, como sucede en la lengua Vascongada. En la lengua mandarina no obstante ser los adjetivos comunes de tres: distinguen á los machos por la particula kung, y â las hembras por la particula mù, y aun tienen otros nombres particulares (hablando de animales) distintos de los machos de las hembras: y lo mesmo sucede en esta lengua Japona, en la qual los generos son comunes de tres, ó comunes de dos [This language is like our Basque language, or Tagalog, Chinese, or the Mandarin dialect, which do not have masculine, feminine, etc. genders for the construction of concord, as occurs in Latin and the Spanish dialect, or in the dialects of the Greek language; since these languages (=Japanese, etc.) lack gramatical gender, whereas some of them do have philosophical genders, as occurs in Basque. In Mandarin, however, adjectives are common to three (genders); they mark males with the particle kung, distinguishing them from the female where mù is used, and they even have some other particular names when they speak of animals, distinguishing the male from the female; and the same applies for Japanese, where gendres are common for three, or common for two] (Oyanguren 2009 [1738]: 52) In his grammar of Japanese, we find a similar description: De los generos se puede discurrir lo mismo, que aunque no tienen generos grammaticales de masculino, y femenino, tienen unos generos logicos en algunas raìzes conjugadas: en las quales las que sirven a los del genero masculino, no pueden servir al femenino, y otras raízes son indiferentes para uno, y otro sexo, y esto succede en la lengua Hebrea, como lo tenemos probado en el Arte Cantabrico, que se escribió el año de 1715: en el Cantabrismo elucidado, v.g. al varon se le dice: Houag cang umagolo, no te amancebes: á la hembra: houag cang paagolo: indiferente: houag mong calunyain no te amãcebes; con él, ò cõ ella‘ [We can reflect upon the genders in the same manner, although grammatical genders, such as masculine and femenine are lacking, there exist some logical genders and some conjugated roots, some of them that are used for the masculine,cannot be used for the femenine, and other roots are indifferent regarding to one or another sex, and this

Oyanguren de Santa Inés‘s grammar of Tagalog

77

occurs in the Hebrew language, as we have proven in the Basque Grammar, which was composed in 1715 and in the Cantabrismo elucidado, e.g. to the male, one says: Houag cang umagolo ―do not cohabit‖; to the female: houag cang paagolo; and indifferently: houag mong calunyain, ―do not cohabit, with him, or with her‖] (Oyanguren 2009 [1738]: 70; Zwartjes 2010: 124) Basque, his own native tongue, was again of great help. There are no noun classes, no grammatical gender and no noun classifiers in Basque and there are no gender distinctions either made within the system of free personal pronouns (Saltarelli 1988: 200). 3.2.2. Verbal inflection, according to Oyanguren 3.2.2.1. As we have seen in the paragraph devoted to grammatical gender of nouns, Oyanguren also distinguishes, in both his Japanese and Tagalog grammars, between those tenses which he labels as ‗grammatical‘, as opposed to those which are ‗philosophical‘. Apparently, he was aware that not all traditional Latin tenses had inflectional morphology equivalents in Japanese and Tagalog. el verbo Tagalog no tiene modos: Mas á esto se podia responder, que la lengua Tagala tiene sus tiempos philosophicos con sus incrementos como se vè claramente en los presentes, y futuros con la geminacion de la primera sylaba de la raìz, ó particula [The Tagalog verb does not have modes. The question why this is the case can be answered as follows: this language has its philosophical tenses with its increments, as can be seen clearly in the present and future tenses, reduplicating the first syllable of the root or particle] (Oyanguren 1742: 71) 3.2.2.2. The term ‗affix‘ is used in Oyanguren‘s theory of verbal morphology: La syntaxis del Verbo passivo Tagalo es de nominativo de persona que padece, verbo, y genitivo de persona que haze… En los tiempos de preterito su conjugacion es adgetiva, y no necessita de afixo como nuestra lengua Cantabrica, ni de verbo substantivo, como las lenguas Latina, Griega, y China [The syntax of the passive verb in Tagalog the person which is affected is its nominative, the verb, the person which acts is its genitive. In the preterite tenses the inflection is adjectival, and an affix is not required, just as in our Basque language, neither is a substantive verb (the copula ‗to be‘) needed, as in Latin, Greek or Chinese] (Oyanguren 1742: 194) We have to interpret this passage as follows. Oyanguren argues that in Tagalog the formation of the preterite of what he calls ‗passives‘ (distinguishing three different types)10 is realized by the affixation of particulas modales, but it is not clear what forms 10

―In modern grammar, the Tagalog voice system consists of at least four voices. Actor voice marked by the infix -um-, (-ung- in Oyanguren) or the prefixes mag, or maN, patient voice by the suffix -in, locative voice by the suffix -an, and the conveyance voice by the prefix i-. The latter three voices share several semantic and formal characteristics, which makes it convenient to use undergoer

78

Otto Zwartjes

he had in mind when comparing Tagalog with his own language, Basque. Since he refers only to the preterite tense, he was probably not attempting to describe the agreement markers of subject/direct object and indirect object, but was probably alluding to the Basque tense marker -en (-e/i(n)…n). Synthetic verbs only have present and past forms, whereas periphrastic verbs exhibit a separate inflectional constituent. The past tense to which Oyanguren refers is marked by an auxiliary whose forms are distinct for the four basic predicate types (intransitive, intransitive-dative, transitive, and ditransitive) or by the infix -en- in the synthetic verbs. A characteristic feature in Basque is that all forms have the suffix -(e)n, as the following example illustrates: atzo presa-ka z-en–bil-tz-en yesterday hurry-ADV 2sA-PAST-walk-Ap.PAST ‗Yesterday you walked in a hurry‘ No comparable preterite marker exists in Tagalog. According to Oyanguren the first distinction to be made (when inflecting a Tagalog verb) is determining its ‗voice‘ (‗stem‘, or ‗root‘) from which other ‗tenses‘ and ‗moods‘ are derived. This fundamental distinction had been made earlier by Blancas de San José: Con solo seys tiempos se hazen todos los modos, y tiempos dellos, en todas quantas maneras de verbos ay, assí en actiua, como en passiua; y con una sola misma voz se hacen todas las personas y los dos números singular y plural. De manera que no hay amabam, amabas, amabat, etc. Sino una misma voz haze para todos [With only six endings, the modes and its tenses are constructed, both in the active and in the passive. And all the persons and the singular and plural numbers are formed using one single word, so that there are no forms, such as amabam, amabas, amabat (Latin, ―I loved‖, ―you loved‖, ―he/ she loved‖, etc.), but one single form is used for all] (Blancas de San José 1997 [1610]; Quilis 1997: 49–50) In Oyanguren‘s Tagalysmo we find a similar observation, but again he not only contrasts Tagalog with Latin, but he compares this language also with Chinese and Basque: Los Tagalos no tienen inflexion en los verbos, y assi se ponen las personas, por tenerlas el verbo, como succede en la latinidad, ò entre Dialectos Griegos: imita en no tener personas el verbo à los Dialectos Chinos, y á nuestros Dialectos Cantabricos, ò vascõgados: por lo qual el Tagalysmo lo mismo es, conjugar verbo que conjugar todos …[The Tagals do not have inflected verbs according to voices as a cover term for them‖ (Himmelmann 2005: 363). As we have seen, Oyanguren uses the term ‗affix‘ but the term ‗infix‘ was not known yet. This does not mean that he did not know the existence of infixes, since he describes their use in detail: ―Y lo particular de este tiempo en esta de ung, es que las raízes de incial consonante intercalan entre la consonante, y vocal el ung [como se dixo arriba]; y si la inicial es vocal, se antepone el ung‖ [a particular feature of the conjugation with ung is that ung is infixed when the first letter of the roots is a consonant, and when the roots begin with a vowel, ung is prefixed] (Oyanguren 1742: 76–77).

Oyanguren de Santa Inés‘s grammar of Tagalog

79

person, as in Latin or Greek, but they have person in the verb (i.e., person is marked on the verb). Tagalog resembles Chinese and our Cantabrian or Basque dialects in not having persons in the verb (i.e. verbs are not inflected for person), so that in Tagalog the conjugation of one verb is conjugating all of them…] (Oyanguren 1742: 81)11 With the particulas modales and the particulas transcendentales (1742: 81) the speakers of Tagalog construct their ‗modal conjugations‘ (conjugaciones modales). Verbs are not composed by prefixing ‗prepositions‘ (as in legere colligere, intelligere) as occurs in Latin, but with these ‗modal‘ and ‗transcendental‘ particles the meanings of the roots are modified, pluralized, and with these modal markers the verb can be given different meanings (imperative, permissive, etc.): Con las conjugaciones antecedentes habla el Indio Tagalo su lengua: y de estas modales se sirue como de proposiciones (por carecer de ellas) para modificar la significacion de las raízes, intentando, pluralizando, mandando, permitiendo, las cosas de intento, ó acaso, y otras phrases de distintas significaciones…[The Tagal Indian speeks his language using the preceding conjugations, and these modal forms correspond to propositions (which are lacking in this language), and they serve to modify the meaning of the roots, ‗attempting‘, ‗pluralizing‘, ‗commanding‘, ‗permitting‘, things as ‗intention‘, or ‗coincidence‘ and other phrases with different meanings] (Oyanguren 1742: 130) 3.2.3. Towards a theory which distinguishes agglutination from inflection From the 19th century, scholars were beginning to develop theories concerning the typological differences between agglutinative, inflectional and isolating languages. Probably, as far as we can trace, Oyanguren is the first in the history of linguistics who compared his native tongue Basque with Japanese, both of which he classifies as lenguas subjuntivas, an equivalent for ‗agglutinative‘. Oyanguren classifies Tagalog as a language which has ‗invariable words‘ (voces indiferentes); ‗absolute words‘ (los absolutos) are to be considered ‗simple roots without composition‘ (raizes simples y sin composicion). Tagalog is a language where ‗segments‘ are agglutinated (agregación de partes), although he admits that inflection in Tagalog is not totally non-existent:

11

Similarly, in another section of his grammar (Schemalexos Tagalidum), he says: ―Los Tagalos imitan à los Griegos, y latinos en las mas de las figuras, aunque aliter, & aliter por ser distintas las phrases, y accidentes de sus oraciones, carecer de concordancias, y inflexiones personales, todo lo qual se verifica en el Dialecto Chinico Mandarin‖ [Tagalog resembles Greek and Latin regarding the other figures of speech, albeit differently, since the phrasing of these languages is different as is the syntax of their clauses, lacking concord and personal inflection, which all can be verified in the Mandarin Chinese dialect] (Oyanguren 1742: 202).

80

Otto Zwartjes

Es un idioma compuesto por agregacion de partes, y assi la consideran los Tagalystas (no obstante que tienen su inflexion, y incremento los verbos) como los Chinos tambien hacen la misma consideracion de la suya [Tagalog is a language which is constructed by addition (‗affixation‘) of parts, and this is the way the grammarians of Tagalog put it (irrespective of the existence of inflection or increment of the verbs) as the Chinese grammarians observe, referring to their language] (Oyanguren 1742: 37) In conclusion, Oyanguren describes the use of prefixes, infixes and suffixes with astonishing detail, using the terms afixo, incremento and agregación and establishing the dichotomy between morphologically marked ‗tenses‘ (gramaticales), as opposed to those which were not morphologically marked (philosophicos/logicales). In modern descriptions, ‗tenses‘ in the Latin sense (i.e. present, imperfect, future, etc.) do not play a prominent role in the verbal paradigm, but rather ‗focus‘, ‗voice‘ and ‗aspect‘, as the table (from Himmelmann 2005) in the Appendix demonstrates. Oyanguren‘s description is far from comprehensive and we do not find a systematic analysis of these ‗aspects‘, but when comparing his approach to the paradigms we find in modern studies, we can conclude that Oyanguren was not strictly following canonical Latin paradigms; he was trying to fit the linguistic data into a new framework.

3.3. Syntax 3.3.1. Word order When Oyanguren describes word order in Japanese, he follows his predecessors, Collado and Rodrigues, although he adds that Japanese shares some features with Basque. He recognises that Japanese is an ‗SOV‘ language: Ponese el nominativo al principio de la oracion, y el verbo suele finalizar, ó terminarla, como se vee en muchos exemplos, que tiene el Vocabulario, el adverbio no suele seguir al verbo, antes le suele anteceder, las otras partes de la oracion se colocan segun cupieren, y el uso lo demuestra, aunque los adverbios de tiempo suelen posponer [The nominative has to be placed at the beginning of the sentence; the verb is generally placed at the end, as can be seen in many examples to be found in the Vocabulario, usually the adverb does not follow the verb, but generally it comes before it; the other parts of speech are placed at random, and language use demonstrates where to put them in the sentence, although time adverbs usually follow the verb] (Oyanguren 1742: 138) When Oyanguren describes Chinese word order, he concludes that in this language the active (‗or transitive‘) verb ‗governs‘ the accusative (= ‗direct object‘), which has to be placed after the verb (SVO):

Oyanguren de Santa Inés‘s grammar of Tagalog

81

El verbo activo Chino rige acusativo despues de si, aunque no tenga particula, que lo determine á dicho caso: v.g. gò gai tien chù, yo amo a Dios. [The active verb in Chinese governs the accusative which comes after it, although it does not have a particle which determines such case, e.g. gò gai tien chù, ‗I love God‘]. (Oyanguren 1742: 201) In his chapter on compound adjectives, Oyanguren concludes that Chinese follows the opposite order compared to Tagalog: El dialecto Chinico mandarin ordinariamente compone sus adgetivos al contrario de los Tagalos. [Mandarin Chinese composes its adjectives opposite to those of Tagalog]. (Oyanguren 1742: 49) 3.3.2. Agreement The non-existence of ‗accidents‘ such as gender and number also had implications for verbal agreement: Ordinariamente los Tagalystas dicen, que en este Idioma, ni ay concordancias, ni generos, y siendo una lengua de tanta reflexion, y tan metaphysica no avia de faltar en uno, ni en otro, aunque lenguajes ay en que no se dan generos grammaticales, sino es logicales, como succede en nuestra lengua Cantabrica en la conjugacion de los mismos verbos, y aun en la lengua Hebrea [Usually, the Tagalog grammarians maintain that concord and gender are non-existent in this language. According to them, this language should not have any shortcomings, since it is a language with so much reflection, and which is so metaphysical. However, there are languages where gramatical gender is also lacking, but instead of it, they have logical gender, as occurs in our Basque language as far as the conjugation of the verbs is concerned; this applies even to Hebrew.] (Oyanguren 1742: 69)

4. Conclusion It is obvious that Oyanguren‘s aim was not writing a study on linguistic typology. He never presents his linguistic observations and interlinguistic comparisons systematically, but if we situate these observations in its historical context, his pioneering observations are unique and worth to be studied, since they contain some crucial aspects of the phonology, morphology and syntax of three unrelated Asian languages, one agglutinating with mainly suffixes (Japanese), another agglutinating with mainly prefixes (Tagalog) and an isolating language (Chinese). It would be unfair to look only at Oyanguren‘s often inconsistent methodology and at his terminology which is not always transparent as seen through the eyes of a modern linguist, since in this article we have demonstrated that Oyanguren was trying to give a new dimension to descriptive linguistics, not only

82

Otto Zwartjes

by highlighting similarities between languages, but even more frequently, by pointing out differences between languages. This innovative approach was obviously not inspired by any previous grammarians. It was unique for a missionary grammarian to pay so much attention to linguistic typology. Some missionary linguists composed grammars of two or more different, even unrelated languages, as did, for instance, Luis de Valdivia (1561–1642), who wrote grammars of Mapudungun (Mapuche) and Millcayac/Allentiac. Nevertheless, we do not find any comparisons or cross references in these grammars between these languages. Oyanguren attempted to fit the linguistic data of unrelated languages into the categories of Nebrija‘s model, but in the subtitle of his Tagalog grammar the addition of the words en lo posible [―as far as possible‖], he demonstrated that he was fully aware that the linguistic facts encountered in Asia were so different from Latin, that some of them could not fit into Nebrija‘s framework. Oyanguren came to the insight that suffixation and agglutination cast new light on Latin grammar which was always considered to be ‗universal‘, but when ‗exotic‘ languages were discovered, such views had to be re-written.

Appendix Actor focus

Object focus

Root

Comp

Inc

Cont

Comp

Inc

Cont

paint (art)

pinta

nagpinta

nagpipinta

magpipinta

ipininta

ipinipinta

pipintahin

paint (on an object)

pinta

nagpinta

nagpipinta

magpipinta

pinintahan

pinipintahan

pipintahan

park (in the garage)

garahe

naggarahe

naggagarahe

maggagarahe

iginarahe

iginagarahe

igagarahe

park (parking lot)

parada

pumarada

pumaparada

paparada

ipinarada

ipinaparada

ipaparada

Table 1 Comp=Completed (Perfective) Aspect – action started and terminated Inc=Incompleted (Imperfective) Aspect – action started but not yet completed Cont=Contemplated Aspect – action not started but anticipated or contemplated

Oyanguren de Santa Inés‘s grammar of Tagalog

83

Bibliography A. Primary sources Anonymous (1972[1630]): Vocabvlario de Iapon declarado primero en portugves por los Padres de la compañía de IESVS de aquel reyno, y agora en Castellano en el Colegio de Santo Thmas de Manila. Manila: Tomas Pinpin, y Iacinto Magariua. Facsimile edition: Classica Japonica Facsimile Series in the Tenri Central Library. General editor: Makita Tominaga. Section I: Linguistics. Editor of the linguistic section: Hisanosuke Izui. Tokyo: Tenri Central Library. Collado, Diego (1632): Ars Grammaticæ Iaponicæ Lingvæ. In gratiam et adivtorivm eorum, qui prædicando Euangelij causa ad Iaponiæ Regnum se voluerint confeerre. Composita, & Sacræ de Propaganda Fide Congregationi dicata à Fr. Didaco Collado Ordinis Prædicatorum pewr aliquot annos in prædicto Regno Fidei Catholicæ propagationis Ministro. Romæ: Typis & impensis Sac. Congr. De Propag. Fide. Oyanguren de Santa Ynes, Melchor (2009[1738]): Arte de la lengua japona, dividido en quatro libros segun el arte de Nebrixa. Con algunas voces proprias de la escritura, y otras de los lenguages de Ximo y del Cami, y con algunas perifrases y figuras. México: Joseph Bernardo de Hogal. Edición facsimilar: Tenri Central Library, Japan, Classica Japonica, Facsimile Series. Ed. Makita Tominaga. Section I: Linguistics, vol. 5. Edition Otto Zwartjes. Madrid/Frankfurt am Main: Iberoamericana/Vervuert. Oyanguren de Santa Ynes, Melchor (1742):. Tagalysmo elucidado, y reducido (en lo posible) a la Latinidad de Nebrija. Con su sintaxis, tropos, prosodia, pasiones &c. Y con la alusión, que en su uso y composición tiene con el Dialecto Chinico, Mandarin con las lenguas Hebrea y Griega. México: La Imprenta de D. Francisco Xavier Sánchez. San Augustin, Gaspar de (1787[1703]): Compendio de la Arte de la Lengua Tagala. Por el Padre Fr.,–– Religioso de el mismo orden, Comissario de el Santo Oficio, Visitador de esta Provincia de Philipinas, y Prior de el Convento de Tambobong (Año 1703). Segvnda impression: Sampaloc: Imprenta de Nuestra Señora de Loreto, 1787. San José (San Joseph), Francisco [Blancas] de (1997[1610]): Arte y reglas de la lengua tagala. en el Partido de Bataan, por Thomas Pinpin Tagalo. Edición facsímil de la edición princeps con Estudio de Antonio Quilis. Madrid: Ediciones de Cultura Hispánica. Valignano, Alessandro (1944[1542–1564]): Historia del principio y progresso de la Compañía de Jesús en las Indias Orientales. Herausgegeben und erläutert von Josef Wicki. Roma: Institutum Historicum S.I. [Bibliotheca Instituti Historici S.I. vol. 2]. Valdiuia, Luys de (2008[1606]): Arte y gramatica general de la lengva qve corre en todo el Reyno de Chile, con vn Vocabulario, y Confessionario. Compuestos por el Padre – de la Compañía de Iesus en la Prouincia del Piru. Ivntamente con la doctrina Christiana y Cathecismo del Concilio de Lima en Español, y dos traduciones del en la lengua de Chile, que examinaron y aprobaron los dos Reuerendissimos señores de Chile, cada qual la de su Obispado. Lima: Francisco del Canto. [Biblioteca Nacional de Madrid, R/8826.]. Ed. Emilio Ridruejo. Madrid: Agencia Española de Cooperación Internacional. Valdiuia, Luys de (1943[1607a]): Arte y gramatica en dos lengvas de Indios Millcayac y Allentiac de las Ciudades de Mendoça y S. Iuan de la Frontera de la prouincia de Cuyo Cõ Cathecismos, Confessionarios, y dos breues Vocabularios en ambas lenguas. Compvesto por el Padre – de la Compañía de Iesus de la Prouincia del Peru. Lima: Francisco del Canto. Ed. F. Márquez Miranda. La Plata: Universidad de la Plata.

84

Otto Zwartjes

Valdiuia, Luys de (1943[1607b]): Vocabulario breve en lengva millcayac, en que estan los vocablos necessarios y suficientes para catequizar y confessar. Compvesto por el Padre – de la Compañia de Iesus, de la prouincia del Peru. Lima: Francisco del Canto. Varo, Francisco (2000[1703]): Arte de la lengua mandarina compuesto por el M.Ro. Pe – de la sagrada Orden de N.P.S. Domingo, acrecentado, y reducido a mejor forma, por N o H. fr. Pedro de la Piñuela P[redicad]or y Comisario Prov[incial] de la Mission Serafica de China. Añadiose un Confessionario muy vtil, y provechoso para alivio de los nueos Ministros. Canton: [publisher unknown]. W. South Coblin & Joseph A. Levi, eds.: Francisco Varo‘s Grammar of the Mandarin Language (1703). An English translation of ‗Arte de la lengua mandarina‘. With an Introduction by Sandra Breitenbach. Amsterdam/Philadelphia: John Benjamins. Varo, Francisco (2006[between 1677–1687]): Vocabulario de la lengua Mandarina con el estilo y vocablos con que se habla sin elegancia. Compuesto por el Padre fray – ord. Pred. Ministro de China consumado en esta lengua esriuese guardando el orden del A.B. c.d. Ed. W. South Coblin: Francisco Varo‘s Glossary of the mandarin Language. Vol. I: An English and Chinese Annotation of the Vocabulario de la Lengua Mandarina. Vol. II: Pinyin and English Index of the Vocabulario de la Lengua Mandarina. Nettetal: Sankt Augustin. [Monumenta Serica Monograph Series, vols. LIII/1 and LIII/2.]

B. Secondary sources Auroux, Sylvain (2000): Introduction. Émergence et domination de la grammaire comparée, in: Auroux, Sylvain (ed.), Histoire des idées linguistiques. Vol. 3. L‘hégémonie du comparatisme. Sprimont: Pierre Mardaga, 9–22. Auroux, Sylvain; Gilles, Bernard & Boulle, Jacques (2000): Le développement du comparatismo indoeuropéen, in: Auroux, Sylvain (ed.), Histoire des idées linguistiques. Vol. 3. L‘hégémonie du comparatisme. Sprimont: Pierre Mardaga, 155–171. Errington, Joseph (2008): Linguistics in a colonial world. A story of language, meaning, and power. Malden/MA: Blackwell Publishing. Himmelmann, Nikolaus P. (2005): Tagalog, in: Adelaar, Alexander & Himmelmann, Nikolaus P. (eds.), The Austronesian languages of Asia and Madagascar. London/New York: Routledge, 350–376. Klöter, Henning & Zwartjes, Otto (2008): Chinese in the grammars of Tagalog and Japanese of the Franciscan Melchor Oyanguren de Santa Inés (1688–1747), in: Histoire, Epistémologie, Langage (―Les langues du monde à la Renaissance‖) 30(2), 177–197. Koerner, Ernst F. K. (1990): The place of Friedrich Schlegel in the development of historicalcomparative linguistics, in: De Mauro, Tullio & Formigari, Lia (eds.), Leibniz, Humboldt, and the origins of comparativism. Proceedings of the international conference, 25–28 September 1986. (Studies in the History of the Language Sciences 49). Amsterdam/Philadelphia: John Benjamins, 239–262. Moran, Joseph F. (1993): The Japanese and the Jesuits. Alessandro Valignano in sixteenth-century Japan. London & New York: Routledge. Quilis, Antonio (1997): Francisco de San José. Arte y reglas de la lengua tagala. Vol. 2: Estudio. Madrid: Ediciones de Cultura Hispánica/Agencia Española de Cooperación Internacional. Sáenz-Badillos, Ángel (2005): Orígenes medievales de la semitística comparada, in: Aguadé, Jordi; Vicente, Ángeles & Abu-Shams, Leila (eds.), Sacrum Arabo-Semiticum. Homenaje al profesor Federico Corriente en su 65 aniversario. Zaragoza: Instituto de Estudios Islámicos y del Oriente Próximo, 421–437. Saltarelli, Mario Donato (1988): Basque. London: Croom Helm.

Oyanguren de Santa Inés‘s grammar of Tagalog

85

Schachter, Paul & Otanes, Fe T. (1972): Tagalog reference grammar. Berkeley/Los Angeles/London: University of California Press. Schreyer, Rüdiger (1992): The European discovery of Chinese (1550–1615) or the mystery of Chinese unveiled. (Cahiers voor Taalkunde 5). Amsterdam: Stichting Neerlandistiek. Seuren, Pieter A. M. (1998): Western linguistics. An historical introduction. Oxford: Blackwell Publishers. Sreekumar, P. (2009): Francis Whyte Ellis and the beginning of comparative Dravidian linguistics, in: Historiographia Linguistica 36(1), 75–95. Streit, Robert (1929): Bibliotheca Missionum. Vol. 5: Asiatische Missionsliteratur 1600–1699. Aachen: Franziskus Xaverius Missionsverein. Sueiro Justel, Joaquín (2007): Historia de la lingüística española en Filipinas (1580–1898). (Colección Ariadna, vol. 1, second edition). Lugo: Axac. First edition: 2003. Téné, David (1980): The earliest comparisons of Hebrew with Aramic and Arabic, in: Koerner, Konrad (ed.), Progress in linguistic historiography. Papers from the International Conference on the History of the Language Sciences, Ottawa, 28–31 August 1978. Amsterdam/Philadelphia: John Benjamins, 355–380. Zwartjes, Otto (1997): Love songs from al-Andalus. History, structure and meaning of the kharja. (Medieval Iberian Peninsula. Texts and Studies 11). Leiden/New York/Köln: Brill. Zwartjes, Otto (2008): El hebreo en las gramáticas del japonés y el tagalo de Melchor Oyanguren de Santa Inés (1688–1747), in: Döhla Hans-Jörg; Montero Muñoz, Raquel & Báez de Aguilar, González Francisco (eds.), Lenguas en diálogo. El iberorromace y su diversidad lingüística y literaria. Ensayos en homenaje a Georg Bossong. Madrid/Frankfurt am Main: Iberoamericana/Vervuert, 393–424. Zwartjes, Otto (2009): Introducción, in: Melchor Oyanguren de Santa Inés. Arte de la lengua japona (1738) dividido en quatro libros según el arte de Nebrixa, con algunas voces proprias de la escritura, y otras de los lenguages de Ximo, y del Cami, y con algunas perifrases, y figuras. Transcripción y edición preparada por Otto Zwartjes, con un prólogo de Toru Maruyama, 21–38. Madrid/ Frankfurt am Main: Iberoamericana/Vevuert. Zwartjes, Otto (2010): Melchor Oyanguren de Santa Inés: Arte de la lengua japona (1738), Tagalysmo elucidado (1742) y ―Arte Chinico‖ (1742). (3 vols.; facsimile edition of the work of Oyanguren + monograph). Madrid: Agencia Española de Cooperación Internacional para el Desarrollo.

PIERRE WINKLER (HAARLEM/NETHERLANDS)

Subject, Topic, Passive and Perspective in Functional (Discourse) Grammar and in Philippine Missionary Grammar

Abstract The missionaries in the Philippine area developed a functional linguistic method similar to modern Functional Grammar. Given this at that time new approach, it is puzzling why they kept using Latin case names and didn‘t invent new terms. It will be argued that they defined the meanings of case names (and of the passive as well) using the notion of perspective. Furthermore, it will be argued that they considered the syntactic level of Subject-Object irrelevant in their analyses of agentive sentences and that there is good reason for functionalists of today to follow the missionaries in this respect. Fourthly, it will be argued that in the discussion about the passives in Philippine-type languages it may be fruitful to consider that the main pragmatic function of passives is not agent-demotion (downgrading of the agent), but predicate-promotion. Finally, it will be argued that the linguistic analyses of the group of missionaries mentioned in this paper deserve a distinct place in the history of linguistic analyses of Philippine-type languages.

1. Terminology As the title already suggests, the key terms in this paper are Subject, Topic, Focus, Passive and Perspective, in comparing the ways these notions are used by a group of missionaries working in the Philippine area with the usage of these terms by the adherents of Functional Grammar (FG) and others who use a comparable pragmatic approach to explain linguistic phenomena. In this paper the term Subject is used in the meaning of ‗the constituent to which the predicate refers‘.1 The term ‗perspective‘ is not used explicitly by our missionaries. However, as I will show in this paper, they did use the notion of perspective, in the sense of movement, foregrounding and backgrounding, to explain syntactic choices



1

I want to thank the attendants of the presentation of this paper at the ICHOLS conference 2008 in Potsdam and the anonymous referees for their comments, which I have gratefully used to describe things more accurately. Also, I want to thank Cornelia Stroh for text formatting and for pointing out some mistakes I made in references. Finally, I want to express my sincere gratitude to Lawrence Reid for correcting my English here and there and for identifying some inaccuracies in the pre-final version of this paper. Cf. Ridruejo (2007: 241). See also Seuren (1998: 121ff.).

88

Pierre Winkler

of a speaker. In FG we can recognise the same approach: the term ‗perspective‘ is used now and then to explain syntactic choices, but is not formally defined.2 The terms Topic and Focus play a central role in FG as two distinct notions. Topic is defined as ―the constituent about which the predicate predicates something‖, Focus as ―the constituent that gives the most salient information‖.3 However, the usage of both terms in recent literature, by functionalists as well as non-functionalists, is confusing and quite often, Topic and Focus are used on the pragmatic level indiscriminately.4 Furthermore, other scholars, mainly the linguists belonging to the ‗classical period‘5, have adopted the term Focus in their analyses of Philippine-type languages.6 However, 2 3

4

5

6

See section 3 in this article. Dik (1981: 31). The definitions given here are in content and meaning identical to those given in Dik (1997), but in this later edition they are a bit more ‗loosely‘ defined, for instance as ―characterizing the things we talk about‖ (Topic) and ―characterizing the most important or salient parts of what we say about a topical thing‖ (Focus). In Functional Discourse Grammar (see the end of next section) these basic notions are defined by contrast: Topic has an organisational function by distinguishing itself from the rest of an utterance, which is called Comment, Focus has an organisational function as opposed to the notion of Background. So the basic function of Topic, according to Hengeveld & Mackenzie (2008), is relating the communicated content to existing information, while Focus signals the speaker‘s ―strategic selection‖ of new information (Hengeveld & Mackenzie 2008: 89ff.). ―Of course, there is a notorious lack of consensus among linguists as to how the term Topic should be defined‖ (Kroeger 1993b: 57f.). Ridruejo (2007: 234f.) makes the same distinction as Dik between Topic and Focus, but adds that other scholars use Focus instead of Topic. This confusion in terminology is also noted by Barlaan (1999: 3). Schachter (1976) describes Topic as ―the centre of attention‖, which is actually a description of Focus. Brainard (1994), discussing voice as a means of signalling pragmatic perspective, in Karao refers to the focus-system of Philippine languages, however without making clear whether she means syntactic or pragmatic prominence (see Liao 2004: 257, discussed in the next paragraph), and uses ‗topic‘ and ‗topicality‘ without defining these terms. Kroeger (1993a) defines Topic and Focus not altogether correctly, at least from a pragmatic point of view, in saying that Topic is ―salient old information‖ and Focus is ―crucial new information‖; ‗pragmatic salience‘ is typically an attribute of Focus. Akamine (1996) is another example of confusion in terminology. He gives examples of the diverse functions of the Topic constituent in Manuk Mangkaw Sinama, which he considers to be a ―Philippine-type language‖ (to which Himmelmann 2002 objects). When giving these examples Akamine says that he prefers to use the term Subject, giving the following argument: ―the term Subject is employed in lieu of Schachter‘s Topic, because Topic to me is another syntactic process of putting a NP in a prominent position, i.e. in the beginning of the sentence. However, I have stayed with the term Focus as used by Schachter & Otanes (1972) to refer to the feature of a verbal predicate that determines the semantic relationship between a predicate verb and its subject.‖ (Akamine 1996: 495) Liao (2004) distinguishes three periods in the history of linguistic analysis of Philippine-type languages: the ‗traditional period‘, the ‗classical period‘ and the ‗modern period‘. Liao‘s ‗traditional period‘ includes the analyses of the missionaries, but unfortunately these were not mentioned by her. As I will argue, the works of the missionaries deserve to be separated from the traditional period, forming a distinct period of analysis of Philippine-type languages The languages the Spanish missionaries described are Kapampangan (Pampango), Tagalog, Pangasinan, Bisayan, Sambal, Bikol, Ilokano and Chamorro. As far as I have been able to gather from li-

Subject, Topic, Passive and Perspective

89

as Liao points out7, all of them used Focus in a non-pragmatic sense. The term Focus, in their analyses, refers to the syntactically most prominent NP. Liao rightly concludes that ‗this unique use of the terms ―focus‖ and ―topic‖ to refer to constructions and/or elements that are not pragmatically salient creates unnecessary terminological confusion in typological literature‘.8 The missionaries did not make a distinction between the pragmatic functions Topic and Focus. They used the term Nominative in the meaning of both Topic and Focus, in a pragmatic sense.9 Following the missionaries, Topic and Focus are used in this paper as synonyms and in a pragmatic sense, both referring to the constituent that a speaker wants to place in the ‗centre of attention‘. However, considering the present confusion existing in the usage of these terms, or maybe better said, the ―crippling lack of formal precision‖10, and to avoid getting involved in a discussion that, first of all, shouldn‘t be there,11 and, secondly, is beyond the scope of this paper, I will use the term Pragmatic Topic/Focus (PTF) as a translation of the missionaries‘ term ‗nominativo‘. PTF refers to the constituent that is the subject of talk and as such forms the reference point or perspective from which the action is viewed. PTF does not refer to what is generally called ‗topicality‘ or ‗discourse topic‘. In other words, PTF has nothing to do with findings based on quantitative analyses.12

7 8

9 10 11

12

terature available to me, all of these languages are considered Philippine languages, except for, perhaps, Chamorro (cf. Adelaar & Himmelmann 2005). However, since there seems to be no consensus yet about whether Chamorro is a Philippine language or not, I have chosen to use the term Philippine-type languages. Philippine-type languages have a ‗focus system‘: ―The so-called focussystem is characterized by the use of various verbal affixes to indicate the thematic role of the NP bearing the nominative case in a sentence. It is found in virtually all Philippine languages, in many of the Formosan languages, in the languages spoken in Sabah, northern Sarawak, and northern Sulawesi, in Malagasy, in Palauan and in Chamorro.‖ (Liao 2004: 88; concerning the similarity between Philippine and other Austronesian languages, also see Reid & Liao 2004). Donahue (2002) speaks of a ―Philippine-style voice system‖, referring to a wider group of Austronesian languages, ―characterised by a case marking system that monitors the degree of pragmatic salience of an argument rather than its syntactic role‖. Our missionaries held the same pragmatic view. Liao (2004: 157). Barlaan (1999: 1) comes to the same conclusion. Liao (2004: 158). Also see Barlaan (1999: 6ff.) for a description of the grammatical, semantic and pragmatic views with regard to the focus system in Philippine languages. Winkler (2007) and sections 2 and 5 of this article. Seuren (1998: 51). The lack of consensus in the definition of the terms Topic and Focus doesn‘t, of course, contribute to making progress in this field of linguistic research. Moreover, it has led to very understandable critique of ‗unscientificness‘ of functionalist approaches. Cf. Seuren (1998: 50ff.). As far as we can gather from the texts of the missionaries, they didn‘t use quantitative analyses. Quite often, nowadays quantitative methods are used to find out which constituent in a text is Topic (see for instance Givón 1994, Brainard 1994, Cooreman 1987, Kroeger 1993b). However, as Bolkestein (1985) clearly points out, the frequency of a Topic in a text does not at all exclude the occurrence of other constituents as Topic or Focus in the very same text. In other words, if one wants to say anything about the possible influence of pragmatic Topic or Focus on syntactic choices, only

90

Pierre Winkler

Passive is generally described in terms of subject/object or agent/ patient assignment and the differences in this respect with active sentences, on a syntactic or a pragmatic level.13 However, as I will argue in this paper, the missionaries in the Philippine area, discarding the syntactic level, used the term Passive in a fully pragmatic meaning, describing the differences between active and passive sentences as different choices in perspective over a state of affairs (SoA). Perspective itself, is a term not yet defined properly. I will discuss this later on. In this paper the term Perspective will be used as referring to the ‗picture‘ of a SoA, involving such notions as backgrounding, foregrounding and movement.

2. Introduction In a previous paper I have shown that there are, to say the least, strong indications that missionaries working in the Philippine area developed a functional or pragmatic method in explaining linguistic phenomena, a method very similar to the modern method that goes by the name of Functional Grammar (FG).14 In this paper the method of this group of missionaries will be referred to as ―Philippine Missionary Grammar‖ (PMG). One reason for this is that using PMG next to FG makes one of the objects of this paper more clear, namely comparing both methods. Another reason is that these missionaries indeed can be called a group or, as I have called them in a previous paper, a ‗school of missionary linguistics‘. They were a school, not in the sense that they worked together and simultaneously in developing new linguistic insights, but in the sense that they to some degree knew each other‘s works and in the sense that they all, in describing their own linguistic analyses, followed the method and the linguistic terminology as written down by Father Blancas.15 This school of missionaries working in the Philippine area explained rules of language ―in terms of their functionality with respect to the ways they are used‖, to quote Dik,16 one of the founders of Functional Grammar theory. In fact, as can be seen in

13 14 15

16

quantitative analyses of ‗topic persistence‘ or frequency of a topic in a given text, are always inadequate. Furthermore, Bolkestein also provides very clear and convincing arguments that it is not the Topic constituent that accounts for syntactic choices (e.g. the choice between active and passive), but the pragmatic Focus constituent. See for instance Dik (1997: 247ff.), Brainard (1994). Winkler (2007). See Ridruejo (2004) for evidence of the exchange of knowledge within this group of missionary linguists. See also Zimmerman (2004) who gives a list of reasons to distinguish missionary linguistics in general from other linguistic research activities. Among other factors in this list, he mentions the fact that the missionaries did their own field research, instead of using data collected earlier and the fact that their research and writings were not meant for some community of linguists in general, but only for the missionaries themselves. See note 6 for the specific Philippine-type languages the missionaries described. Dik (1981: 18).

Subject, Topic, Passive and Perspective

91

fragment (1) below, one of these missionaries defined the basic concept of Topic in the same way as Dik does in his Functional Grammar about 400 years later: ―A topic presents the entity about which the predicate predicates something in the given setting.‖17 Regarding the terms the missionaries used in explaining language phenomena from a functional sentence perspective, I have come to the following conclusions:18 A They used the term nominative to refer to PTF (without, as I have mentioned earlier, systematically making a distinction between Topic and Focus). B They used other Latin case names (Accusative, Genitive, etc.) predominantly to indicate a semantic role (Agent, Patient, Beneficiary, etc.) and not for describing morphosyntactic aspects. The following quotations make clear that the Latin case names are used functionally rather than formally:19 (1)

La cosa que padeçe y acerca de que se ha de exercitar lo que el verbo dize, se pone en nominativo y la que haze en genitivo: ysulat ni Pedro yto: esto sea escrito de Pedro.20 ‗The thing that undergoes and about which the predicate predicates something, is PTF [nominativo] and the agent becomes oblique [genitivo]: ysulat ni Pedro yto: this is written by Peter.‘

(2)

De la pasiva de y se ha de usar cuando se significa alguna cosa que va de la persona que hace (la cual se pone en genitivo) a otra parte, y ha de estar en nominativo…21 ‗The passive with ‗y‘ must be used when one wants to express that a thing goes from the agent (which becomes the oblique agent [genitivo]) to another party, which has to become PTF [nominativo].‘

The following fragments (3) and (4) clearly show that the missionaries used case names not in a morphosyntactic sense – one of the main mistaken assumptions on the basis of which their linguistic analyses have unjustly been neglected and even explicitly been depreciated22 – but predominantly in a semantic sense and that they used the case name Nominative in the pragmatic sense of PTF: 17 18

19 20 21 22

Dik (1981: 13). I am very grateful that my esteemed colleague in this field, Wolf Dietrich, agrees with my conclusion that the Latin case names indeed are used in a semantic-pragmatic meaning and says that the argumentation I have used in Winkler (2007) for the translation of Latin cases as such is ―absolutely convincing‖ (Dietrich 2008: 439–445). For a more elaborate discussion of these and other examples, cf. Winkler (2007). Blancas (1997 [1610]: 47). Coronel (1621: 10v). For some quotes from linguists in which they express this depreciation, cf. Winkler (2007: 330– 331). Also see Seidenadel‘s criticism, specifically with respect to the missionaries use of the term Passive, described in Reid‘s contribution to this volume.

92

Pierre Winkler

(3)

Ytem es de saver que el nominativo puesto al principio de la oración sirve para otros casos, V.G. Say cavallo nan sulat si Pedro, acerca del caballo escribe Pedro.23 ‗One should also know that the constituent which has PTF status [nominativo], placed at the beginning of a sentence, can have different semantic functions, for instance Say cavallo nan sulat si Pedro, About the horse Peter writes.‘

(4)

Todo lo que es lugar o como lugar, poniendose en nominativo, pide la passiva de an.24 ‗Everything that is place or like place and is PTF [nominativo], needs the passive with ‗an‘.‘

As for example (3) above, the concordance with modern findings is again striking. The PTF constituent indeed, as López already concludes in 1627, can have different semantic functions, as can be seen in the following examples of Tagalog taken from Liao (2004) (Focus and the ‗focus-affixes‘ are boldfaced):25 (5)

Actor Focus sumùsúlat siyà ng líham ‗He is writing a letter/letters‘

(6)

Goal Focus sinúlat niya ang líham ‗He wrote the letter’

(7)

Locative Focus pinutúlan niya ang káhoy ‗He cut a piece off the wood’

23 24 25

López (1628: 13r). Blancas (1997 [1610]: 51). Focus-affixes are also called ‗case marking affixes‘ or ‗agreement marking affixes‘. Other terms are also in use. For further details, see Liao (2004: 154ff.). Liao takes these examples from Bloomfield (1917). Later in her work she emphasizes that Bloomfield, as others after him, use Focus here as indicating ‗syntactic prominence‘, so not in a pragmatic sense (see also note 4). However, other scholars, like Akamine (1996), Donahue (2002), Cooreman (1987), Givón (1995) and Barlaan (1999) argue, just as the missionaries did, that this Focus system is pragmatically motivated. Barlaan (1999: 28ff.) gives several examples in Isnag of pragmatically focused NP‘s as agents, goals, instruments, beneficiaries and locations. Zobel (2005: 639) also says that pragmatic factors are involved in voice selection and that if there are ‗highly topical arguments, such as an instrument, beneficiary or goal‘, the syntactic rule that the actor voice is used when the patient is indefinite is overruled by this pragmatic factor of ‗high topicality‘. Zobel seems to refer with ‗highly topical‘ to what is more properly called ‗pragmatically most salient‘ or Focus in the definition of Dik (1981). If so, Zobel‘s conclusion is very comparable with mine in Winkler (1985), that pragmatic motivation for the choice for a passive instead of an active construction (and vice versa) can (and frequently does) overrule syntactic rules and even rules concerning the maintaining of a discourse topic. See also note 12 for the reference to Bolkestein (1985) who corroborates this conclusion by evidence from Latin.

Subject, Topic, Passive and Perspective

(8)

Instrumental Focus ipinútol niya ang gúlok ‗He cut with the bolo’

(9)

Benefactive Focus ibilí mo ako nitó baríl ‗(You) buy this gun for me’

93

It seems obvious to assume, as in fact I have done until now, that these missionaries retained, within their pragmatic definition of Nominative, the syntactic notion of Subject as well. For instance because the constituent that is PTF very often also is Subject, and because the term Subject was quite well established in those days, as being the constituent to which the predicate refers.26 However, as Ridruejo has shown,27 they used the term Subject (suppositum, supuesto) only with a specific kind of sentence, while with all other sentence constructions they didn‘t use this term. They used the term Subject only with non-agentive sentences – ‗existential‘ or equative or identifying sentences, but didn‘t use it when explaining agentive sentences, i.e. sentences which ―contain a typically animate instigator of the action identified by the verb‖.28 In other words, in agentive sentences they explained linguistic phenomena on a pragmatic level and a semantic level, but systematically left out the syntactic level. Therefore, we can now draw a third conclusion: C While the notion of Subject was already well established in those days, the missionaries working in the Philippine area systematically discarded this syntactic level of analysis in explaining agentive sentence structures. These three conclusions give rise to the following questions: 1 Why did they retain the ‗old‘ Latin case name terminology and didn‘t introduce at least one new term for the pragmatic level of Topic/Focus instead of the case name Nominative? 2 Why didn‘t they use the syntactic level of Subject in analysing agentive sentences? There are, however, two other issues also relevant in the context of this paper. Firstly, the missionaries‘ use of the term Passive for all sentences in which a constituent other than the actor is Subject. This use of the passive has been severely criticised by some scholars.29 However, the question of whether Philippine languages truly have a passive or not is still the subject of ongoing debate today.30 Secondly, examples (1) to (4) seem to suggest that the missionaries used changes in perspective to explain linguistic phe26 27 28 29 30

Cf. Ridruejo (2007: 241). See also Seuren (1998: 121ff.). Ridruejo (2007). Primus (2009: 263). See section 4 of this article. See Reid‘s contribution in this volume.

94

Pierre Winkler

nomena, a method also used in FG, as I will show later on. Therefore, it is relevant as well to take a closer look at the role that perspective plays in FG and in PMG. This leads to two additional questions to be dealt with in this paper: 3 Why did the missionaries choose to label all sentences with a PTF other than actor as passive? 4 To what extent did the missionaries use the notion of perspective to explain linguistic phenomena? Recently, a new book has been published, written by Kees Hengeveld and Lachlan Mackenzie, in which a new theory is presented: Functional Discourse Grammar (FDG). The fundamental starting point is that ―Functional Discourse Grammar is so called because it adheres to the principle of linguistic functionalism and takes the Discourse Act as its basic unit of analysis‖.31 In the preface of their book, Hengeveld & Mackenzie emphasise that FDG diverges from FG in so many ways that it should not be regarded as a successor to FG, but as a theory in its own right. Where it is relevant to the subject of this paper, for instance when this new theory of FDG explains or defines terms and notions adopted from FG substantially differently, I will refer specifically to this book of Hengeveld and Mackenzie.

2.1. Structure of this paper In this paper I will try to find answers to these questions. Since I have argued that Spanish missionaries working in the Philippines used a method of explaining language phenomena that is comparable to the method used in Functional Grammar, I will take a closer look at the notions of Topic, Focus and Perspective and how they are related to Subject in FG. This will be discussed in section 3. In section 4, I discuss the examples of passive sentences given in section 3 and further consider the role that perspective plays in order to make clear that the most important function of the passive is not demotion of the agent or promotion of the patient, but promotion of the predicate, making the action contained in the verb the central point of perspective. In section 5, I compare the fundamentals of the theory of FG with the theory developed by the missionaries, viz. PMG. Also, I will briefly deal with the question of whether PMG may have been inspired by earlier sources, such as Nebrija (1996 [1488]). In section 6, I deal with the use of the term Passive as a function of perspective in PMG. In section 7, I discuss why the term Subject is systematically left out of the analyses of agentive sentences in PMG. Finally (in section 8), I give an overview of my conclusions. 31

Hengeveld & Mackenzie (2008: 25).

Subject, Topic, Passive and Perspective

95

3. Subject, Topic, Focus and Perspective in FG The fundamental starting point of FG (and FDG) is that a speaker constructs an utterance in such a way that the speaker effects some sort of change in the addressee. These changes may refer to adding new information, but also to a change in attitude or other mental aspects. To achieve this change, the speaker gives pragmatic information in a certain way, taking into account the information he thinks the addressee already has that is related to the subject of talk. Pragmatic information may consist of factual knowledge, but also includes all kinds of social and psychological aspects such as opinions, perceptions, etc.32 Furthermore, linguistic expressions are seen as characterised by functional relations at three different levels:33 1 Pragmatic functions (Topic, Focus) 2 Semantic functions (Agent, Goal, Recipient, etc.) 3 Syntactic functions (Subject and Object) The numbers here indicate a hierarchy. In analysing a speaker‘s utterance one should first look at the pragmatic information (the type of change a speaker wants to realise in the addressee), and then one should analyse how this desired pragmatic information is further effectuated in the semantic and syntactic choices the speaker makes. Dik describes this hierarchy in the following manner: …in the functional paradigm the relation between the different components of linguistic organization is viewed in such a way that pragmatics is seen as the allencompassing framework within which semantics and syntax must be studied. Semantics is regarded as instrumental with respect to pragmatics, and syntax as instrumental with respect to semantics. […..] syntax is there for people to be able to form complex expressions for conveying complex meanings, and such meanings are there for people to be able to communicate in subtle and differentiated ways.34 Topic is defined as ―the constituent about which the predicate predicates something‖, Focus as ―the constituent that gives the most salient information‖.35 Topic and Focus 32 33 34

35

Cf. Dik (1997: 10ff.). Cf. Dik (1997: 26ff.). Dik (1997: 8). Hengeveld & Mackenzie (2008: 1) describe this hierarchy very briefly: ―FDG starts with the speaker‘s intention and then works down to articulation.‖ Hengeveld and Mackenzie, however, have decided to rename the first two levels mentioned here: pragmatic level has become interpersonal level in FDG, semantic level has been renamed representational level. The syntactic level has been broadened a little and is now called the morphosyntactic level. Dik (1981: 31). The definitions given here are in content and meaning identical to those given in Dik (1997), but in this later edition they are a bit more ‗loosely‘ defined, for instance as ‗characterizing the

96

Pierre Winkler

may coincide, but this is not always the case. Compare the following syntactically identical sentences with the same Topic, but with a different Focus: (10)

John (Top, Foc) gave the book to Bill

(11)

John (Top) gave the book to Bill (Foc)

Sentence (10) gives answer to the question: ‗who gave the book to Bill?‘, whereas sentence (11) gives answer to the question: ‗to whom was the book given by John?‘. Now, it is easy to understand that semantics and pragmatics are characterised by functional relationships and thereby influence a speaker‘s linguistic choices with respect to these two levels. However, this is less obvious at the level of the syntactic functions Subject and Object. Dik is actually quite confusing – and mistaken, as we will see – in his arguments for retaining this level in FG. I will go into this in more detail, because it is also very relevant with regard to the linguistic analyses of our missionaries. The first reason Dik (1997: 26f.) gives for retaining this level is: …to avoid unnecessary departure from traditional terminology: the concepts of Subject and Object are used to capture such oppositions as that between active and passive constructions, and these oppositions are usually described in terms of syntactic or grammatical relations. However, a more important reason according to Dik (1997: 254ff.) is that ―Subject and Object determine an additional level of SoA [State of Affairs] organization, which languages may possess…‖ The question, however, still remains to what extent and in what way this level of Subject and Object has a functional meaning. According to Dik, Subject and Object have to be redefined in order to play a functional role and this role is: ―…defining different perspective over the State of Affairs designated by the predication. For that reason, ‗perspectival functions‘ might be a better term to cover their essential nature.‖36 As an example of this ‗perspectival function‘ Dik (1997: 252).gives the following two sentences: (12)

The police removed the demonstrators from the platform

(13)

The demonstrators were removed from the platform by the police

36

things we talk about‘ (Topic) and ‗characterizing the most important or salient parts of what we say about a topical thing‘ (Focus). In FDG these basic notions are ‗defined‘ by contrast: Topic has an organisational function by distinguishing itself from the rest of an utterance, which is called ‗Comment‘; Focus has an organisational function as opposed to the notion of Background. So the basic function of Topic, according to Hengeveld and Mackenzie, is relating the communicated content to existing information, while Focus signals the speaker‘s ―strategic selection‖ of new information (Hengeveld & Mackenzie 2008: 89ff.). Likewise, Barlaan (1999: 67ff.) describes Topic as the constituent that is either old or presupposed and Focus as the constituent that is new and/or asserted. Dik (1997: 27). In FDG the same view can be found. See Hengeveld & Mackenzie (2008: 35f.).

Subject, Topic, Passive and Perspective

97

According to Dik, the difference in perspective can be compared to different photographs taken of the same event. In (12), for instance, the picture represents the event from the viewpoint of the police, whereas (13) tells a somewhat different story, since the event here is represented from the viewpoint of the demonstrators. However, as Dik says, a ―full theory of perspective‖ still has to be developed.37 The reasons for choosing between different perspectives concerned with active sentences and their passive alternatives can be, according to Dik (1997: 252f.):     

Empathy or identification. In the given examples the speaker places himself in the role of the agent (12) or places himself in the role of the demonstrators (13). Preference for making the Given Topic subject rather than making a New Topic subject. The agent is unknown, considered by the Speaker to be unimportant, or the speaker consciously wishes to leave the agent unspecified. Ungrammaticality of the active sentence. Politeness conventions.

As one can easily see in this list of possible reasons for making the Object in the active sentence Subject in the passive alternative, there are quite a lot of terms used in this list – ‗empathy‘, ‗identification‘, ‗considered unimportant‘, ‗consciously leave the agent unspecified‘, ‗politeness‘ – which suggest a strong pragmatic motivation. Without explicitly agreeing that these terms are highly pragmatic, Dik (1997: 255) seems to admit that this indeed creates a problem. He tries to solve it as follows: More complicated is the question about the relationship between Subj/Obj function and pragmatic functions, specifically Topic. We define the Subj as specifying the vantage point from which the SoA is presented in the predication, and the (Given) Topic as that entity about which the clause predicates something in the given setting. Subj has to do with the presentation of the SoA, Topic with the contextual embedding of the information transmitted by the clause. However, why is the question ‗specifically‘ complicated regarding Topic? It is, in fact, equally complicated regarding Focus. If any term, by its meaning, is instinctively associated with the term perspective, it is Focus. As an illustration, see the following examples (14) and (15) that Dik gives to illustrate the meaning of and difference between Topic and Focus. In both examples it is the Focus constituent that has the ‗perspectival function‘, not the Subject. In (14) the event is presented from the viewpoint of the receiver, in (15) the event is presented from the viewpoint of the giver. (14)

This book (SUBJ-TOP) was given to ME (FOC) by John

(15)

This book (SUBJ-TOP) was given to me BY JOHN (FOC)

37

Dik (1997: 254). Also see note 39 for reference to some other scholars who use the notion of perspective in their analyses without giving a proper definition of the term.

98

Pierre Winkler

Now, these examples indicate that Dik, by not giving up the traditional syntactic notions of Subject and Object and, instead, redefining them as he did, has caused himself severe problems. By retaining this syntactic level he actually attributes pragmatic meaning to a level which he has distinguished himself as non-pragmatic and by doing so he blurs his own hierarchic distinction between pragmatic, semantic and syntactic levels. Therefore, in a truly functional grammar, it makes much more sense to consider the syntactic level of Subject and Object irrelevant. It may be influenced by, but does not contribute to the pragmatic presentation of a SoA. The simple examples (14) and (15) from Dik himself, make clear that Subject and Object define syntactic relations and that functions of perspective are not the essential nature of the concepts of Subject and Object. As these examples clearly show, the description of an event can be given from different perspectives, the viewpoint of the Subject, the Object or any other constituent, within one and the same syntactic construction using exactly the same words, and exactly the same Subject. The viewpoint in FDG concerning the meaning of Subject/Object doesn‘t seem to be consistent. FDG recognises that the morphosyntactic level does not add or subtract any pragmatic or semantic meaning. Hengeveld & Mackenzie, however, still attribute a ‗perspectival function‘ to Subject and, furthermore, say that they consider the syntactic pair of Subject/Object ―relevant‖ for functional linguistic analysis, when the speaker‘s intentions are neutral.38 Surprisingly, they do not explain what they exactly mean with the term ‗relevant‘ in this context. Even more surprising is that instead of redefining Dik‘s definition of Subject/Object, they do not properly define this syntactic pair, even though they frequently use these terms. And maybe even more surprising is the fact that Hengeveld & Mackenzie give no explanation, on any level of analysis, of a speaker‘s possible motivation for choosing between an active or a passive construction, nor give any treatise on the term Perspective, a term which indeed, as Dik says, still needs a full definition.39 38

39

For the linking of Subject to Perspective, see note 39. Hengeveld & Mackenzie (2008: 316ff.) leave the possibility open that sometimes the speaker‘s intentions are ‗neutral‘, in the sense that these intentions do not urge the speaker to make a functional choice between a set of possible syntactic constructions. Pragmatic neutrality, however, is a problematic concept. One may object, for instance, that if a speaker‘s intentions are neutral, this as well is a pragmatic choice, or that if syntactic constructions do have pragmatic meaning, a really neutral standpoint is impossible to express. In my own research on the pragmatics of passive constructions (Winkler 1985), I haven‘t found any instance of true neutrality. The acceptance of neutrality also seems to contradict the fundamental basis of FG, as mentioned above, that a speaker constructs an utterance in such a way that the speaker effects some sort of change in the addressee. Hengeveld & Mackenzie (2008: 35) mention the term Perspective as a function of Subject, saying that Subject enables a speaker to present a SoA from the vantage-point or viewpoint of an Actor or an Undergoer, just as Dik (1997). Perspective in the sense of foregrounding and backgrounding of constituents is noted by Klamer (2005) in Kambera. However, she doesn‘t mention the term perspective explicitly and as for foregrounding and backgrounding, she only says that they have to do with the degree of emphasis that is put on a constituent. Mithun (1994: 255) uses the term ‗highlighting‘: ―speakers can often select the participant whose involvement they wish to highlight‖. (The examples given by her concerning this highlighting are relevant in the discussion of ‗predi-

Subject, Topic, Passive and Perspective

99

4. Predicate promotion as the main pragmatic function of passives Quite often, two main functions of the passive are distinguished: patient-promotion and agent-demotion (downgrading of the agent). In the list Dik gives of pragmatic functions (see previous section) one can also recognise these main functions, for what he calls ‗empathy or identification‘ clearly is, in the case of example (13), a pragmatic effect of patient-promotion, and the agent being considered irrelevant or better not mentioned can be seen as pragmatic motivations for agent-demotion. Many others also have distinguished these two functions of the passive, or focus mainly on the function of agentdemotion.40 However, one can also find, though less frequently, descriptions of predi-

40

cate-promotion‘ later in this paper; see note 48). Givón (1994: 8) also mentions, even boldfaced, ―the variant clause-types that code changes in the pragmatic perspective‖, however, again without giving a definition of perspective. Likewise, Brainard (1994: 374) argues that ―voice may be defined as a means of signalling pragmatic perspective‖, without giving a definition of the term. No definition is given either in Brainard & van der Molen (2005), though again the term is used frequently. Ogawa (2009: 782) simply describes perspective as a process used for ―foregrounding of the topic vs. of the comment‘. See also Cornelis (1997) for a good overview of different aspects and notions associated with perspective by functional linguists. The notion of movement, which, as I will show, is a prominent concept in PMG, can also be found in Himmelmann (2005: 366). Concerning the function of perspective, Dik gives examples such as the one about the demonstrators given above. Dik compares it with the perspective in a photograph. However, photographs, paintings etc. have perspective in a different sense than sentences and longer text units. ‗A picture tells a thousand words‘ is a popular expression, but the reverse can be said to be equally valid (‗a word tells a thousand pictures‘), which simply means that pictures and words are essentially incomparable. Both pictures and words each use an entirely different system for building and presenting information, and, as a result, present states of affairs differently. Therefore, the term ‗perspective‘ should be defined differently for both systems. Cf. Shibatani (2009: 326f.) who argues that in passive sentences where the actor is still overtly encoded or understood from the context, the pragmatic motivation is downgrading the discourse relevance of the participant in question. Shibatani (1985) also argues that agent-demotion (―agent defocussing‖) is the main function of the passive. Givón (2001: 125) holds a similar view: ―the prototypical passive voice is used primarily for A-suppression or de-topicalization. The fact that a nonagent argument – most commonly the patient – is then topicalized is but a default consequence of agent suppression‖. Coulter (2005) mentions patient-promotion and agent-demotion as distinct functions of a passive: ‗First, the passive can be used when it is advantageous for the patient of the action to be the grammatical subject… Secondly, a speaker can use the passive to avoid naming the agent, because it is so obvious as to be unnecessary, or because it is unknown, or even to obfuscate the responsibility for an action.‘ Marín Arrese (1997: 203), following Givón (1994) also distinguishes these two functions: ―voice is a fundamentally pragmatic notion, for it allows for the construal of the same semantically transitive event from two discourse pragmatic perspectives.‖ Malchukov (2006) seems to link both processes of agent-demotion and patient-promotion (without using these terms explicitly): ―Perhaps, more usually, non-volitionality licenses the use of an ―experiencer-anticausative construction‖. As for patient-promotion and agent-demotion in relation to impersonal passives, see Brown &

100

Pierre Winkler

cate-promotion as a function of the passive.41 This function of the passive may well be considered not only as indeed a third main function of the passive, but in fact the most frequently occurring motivation for choosing a passive construction. I will go into this a bit deeper, because it is very relevant for a proper evaluation of the linguistic approach used in PMG, and perhaps as well for a proper judgment concerning the existence or non-existence of true passives in Philippine-type languages.42 After having illustrated this phenomenon of predicate-promotion by giving a number of examples, I will come back to this at the end of this section. I define predicate-promotion as focusing on the predicate, as a consequence of which the predication, i.e. the verb and the arguments it needs (like subject and object) plus possible other lexical satellites that together present a SoA – is shown from the perspective of the action itself.43 When there is predicate-promotion, the predicate is the nucleus from which, by adding constituents, the picture the speaker wants to give is built up. So, for instance, in the sentence ‗Peter writes‘ the Focus can be on the predicate ‗writes‘, just as in ‗Peter is writing a letter carefully‘, where, from the focused perspective of the action of writing other constituents are added to create the picture of the action taking place. Levinson (1987: 270ff.). Their theory that the impersonal passive is used for politeness reasons is shown to be incorrect by examples (15) and (16) in this article. Also see note 51. 41 For predicate focus in general, see Van Valin (2009), Wickboldt (2000), Winkler (1985). Klamer (2005: 721ff.) argues that the same pragmatic effect, which she calls ‗emphasis on the predicate‘ can be found in Kambera: ―The accusative S appears when the predicate is emphasized, and S is subsequently backgrounded.‖ Lambrecht (1994) and Van Valin & LaPolla (1997) distinguish three pragmatically motivated focus types, ‗sentence focus‘, ‗narrow focus‘ and ‗predicate focus‘. The latter means that the focus lies on the constituents that function as a comment on the Topic of the sentence. ‗Sentence focus‘ means that all the information given in the sentence is regarded as equally new, and narrow focus means that there is focus on only one specific constituent. Dery (2007) argues that the term ‗predicate focus‘, as defined by Lambrecht (1994) is not applicable in Tagalog and should be replaced by ‗multiple focus‘. This discussion is somewhat beyond the scope of my paper. In the distinction of three focus types by Lambrecht (1994) one can find some overlap with the distinctions used in this paper between agent-demotion and patient-promotion (both considered as ‗narrow focus‘). My definition of predicate-promotion, however, is essentially different from Lambrecht‘s predicate focus (see the next paragraph). Himmelmann (2005: 367) notes that pragmatic factors for the choice among different voices are ‗still poorly understood‘ and the overview of Kazenin (2001) of diverse views on the function of the passive shows that treatises on this function of predicate-promotion or predicate focus, in whatever way defined, are still not very common, though, as he rightly says: ‗voice is a verbal category, so it would not be natural for it to have demotion or promotion of an argument, rather than marking of some characteristic of the event or the event in general, as its core meaning‘ (Kazenin 2001: 907). 42 See Reid and Liao (2004) and Liao (2004). They are both discussed further on in this section. 43 My definition of predicate-promotion is essentially different from the definition of ‗predicatefocus‘ by Lambrecht (1994, see note 41). The main difference is that in Lambrecht‘s definition, the Topic constituent is the nucleus and in my definition the predicate is the nucleus from which the picture is built up. My definition fits in with Kazenin‘s view (see note 41) that it would be natural for voice itself to have a marking of some characteristic of the event in general as its core meaning.

Subject, Topic, Passive and Perspective

101

Predicate-promotion by means of a passive construction can be used to emphasise the reality, the factual truth or the logicality of the action itself. Because of this pragmatic effect, the passive construction is quite often chosen to add a tone of formality or rational neutrality to the message, or to add vividness or dramatic effect to the picture of the event itself. Predicate-promotion may also be used to form a so-called impersonal passive, in which case it can be used as an imperative. In the vast majority of cases predicate-promotion is the main reason for a speaker to choose the passive construction.44 So, in (13), the example given by Dik, Focus can indeed be on either the agent or the patient, depending on the context and the effect the speaker wants to realise, but it can also be on the predicate. Focus on the predicate means that the speaker wishes to present the SoA mainly from the perspective of what is actually happening without making either agent or patient too prominent. Compare this with Dutch sentences like: (14)

Dit wordt nooit gedaan. (This is never being done.)

(15)

Er wordt nu geslapen! (There be slept now!)

(16)

Ik wil geen geklets meer! Er wordt vanaf nu alleen nog maar geluisterd. Is dat duidelijk?! (I don‘t want any talking anymore! From now on, there be only listened. Is that clear?!)

Example (15) and (16) illustrate as well the aforementioned use of the impersonal passive as an imperative construction. In all three examples agent and patient are completely left out of the picture, in order to focus maximally on the predicate. Another illustration is a fragment of Virgil‘s Aeneid, in which he describes the hasty preparations on the shore by Aeneas and his men in order to leave as quickly as possible, seen through the eyes of Dido: (17)

44

45

quis tibi tum, Dido, cernenti talia sensus, quosve dabas gemitus, cum litora fervere late prospiceres arce ex summa, totumque videres misceri ante oculos tantis clamoribus aequor!45

Winkler (1985), in which, based on extensive research in mainly Latin and Dutch (written and spoken) texts, a rough estimate is made that in about 70% of cases the passive is chosen to promote the predicate. Vergil, Aeneid IV, 409–411.

102

Pierre Winkler

(What did you feel, Dido, when you saw such things, how did you sigh, when from the top of the fortress you saw before your eyes the beach aglow far and wide, and the whole main being filled (lit.: ‗being mixed‘) with loud cries!) Virgil is using the passive to let his readers concentrate on the picture (as a whole) of Aeneas‘ hasty flight, the picture of people rushing to make the ship ready, of loud noises and people shouting commands at each other, without focussing on any of the participants in particular. In other words, there is Focus on the passive ‗misceri‘: the perspective from which the information is presented is the actual ‗misceri‘ of things and people and emotions.46 These pragmatic functions of the passive and especially the main function of predicate promotion, may well shed a different light on the passives in Philippine-type languages. Reid & Liao (2004) argue that one of the conditions for labelling a sentence as passive is that the agent is downgraded.47 However, the main pragmatic reason for choosing a passive, at least in languages like Dutch and Latin is not to downgrade the agent, but to upgrade the predicate. As a result of which the agent may be downgraded, but in the majority of cases this downgrading is not the primary motivation to choose a passive.48 Also, the argument of Liao (2004: 149) that ―passives are far more frequent than would be expected if they were true passives‖, does not necessarily imply that we need to reject the practice of calling all the non-actor voices passives.49 Reasons for a 46

47

48

49

Mithun (1994: 256) gives an interesting example of a sentence in which this pragmatic function of emphasis on the predicate may be present for very much the same pragmatic reasons (vividness, dramatic effect) as in Virgil‘s picture of the hasty preparations of Aeneas and his men. In comparing ―George is cooking the turkey‖ and ―George is cooking‖, she says that the first sentence ―highlights the immediate involvement of the turkey, while in the second the involvement of the turkey is bypassed in order to highlight that of George, scurrying around the kitchen.‖ However, she concludes that a speaker‘s choice between these two sentences only has to do with the participant he sees as most directly involved in the action, so in this case George or the turkey. As I have tried to make clear by giving this text fragment of Virgil (and examples 13 to 15), a speaker‘s motivation for choosing a sentence construction is not restricted to highlighting the object (the turkey) or the subject (George), but can also lie in his wish to highlight the action itself (‗scurrying around the kitchen‘). They consider the term passive only justified in the case of ‗stative verbs‘, because ―they are intransitive, there exists a clear derivational relationship between them and transitive verbs, and their actors are typically not expressed.‖ (Reid & Liao 2004: 462) Liao (2004: 150) argues that in Philippine-type languages agents are typically obligatory in these so-called passive constructions, while cross-linguistically passives typically lack an overt agent. Kroeger (1993b: 50) also notes this difference, recognising it as ―a seemingly unique fact about the Philippine voice system‖. However, passives in languages like Dutch and Latin allow for an explicit mentioning of the agent as well, and the lacking of an overt agent is not as typically present as Liao suggests. Furthermore, if downgrading of the agent is involved, then most of the time this downgrading is not the primary motivation for the passive but a direct consequence of the upgrading of the predicate. Cf. Reid & Liao (2004). See also Reid‘s contribution to this volume. For an overview of those scholars who, as the missionaries, do consider them as passives, see Liao (2004).

Subject, Topic, Passive and Perspective

103

higher frequency of passives and even for the lacking of related active counterparts50 may well be found, for instance, in cultural differences: different ways of looking at things, different rules concerning how to say things – for instance in a formal, nonformal or polite way, differences in the ways people are expected to interact, differences in what is seen as a neat or eloquent way of saying things, and so on. It needs further investigation if, and to what extent, these kinds of differences can account for the higher frequency of passives in Philippine type languages, or as Kroeger formulates it, for ―the pragmatic preference for patients‖ in subject selection.51

5. Perspective and cases in PMG The pragmatic and perspective reasons for choosing a passive construction given in the previous two sections are very relevant for a proper evaluation of the linguistic approach used in PMG, and not only concerning the passive. When using the term perspective in several passages in his book Dik often associates the term with ―foregrounding‖, ―backgrounding‖, ―direction‖, ―movement‖ and, of course, Focus.52 All these general notions of backgrounding and so forth associated with perspective53 appear to play an important role also in the pragmatic linguistic analyses of the missionaries. As I have said above, in PMG the term Nominative is used pragmatically, in the meaning of PTF, and Nominative and the other cases are used in the semantic meaning of Agent, Patient, Beneficiary, Oblique agent, Locative and Instrumental. However, looking more closely at the texts written by the missionaries, there seems reason enough to add another dimension of analysis to these pragmatic and semantic roles attributed to case names, namely the perspective dimension. This not only becomes clear from the perspective function of the nominative as PTF. In fact, if this were the only evidence of the use of perspective in PMG, it would be a rather meagre proof, for nominative and Topic and Focus all are easily, in fact even quite naturally, associated with the notion of perspective, as being the ‗thing we talk about‘, ‗the centre of attention‘, ‗the most salient information‘. All these and other definitions or descriptions one can give of these terms 50

51

52 53

Liao (2004: 152). The higher frequency of passives was already explicitly noted by Totanes in 1865, quoted by Kess (1979). The quote is taken from Kroeger (1993b: 55). Liao (2004: 151ff.) rejects politeness as a possible factor for the higher frequency because the passives also occur in imperative sentences and these sentences do not by any means express politeness. However, passive imperatives also are used in Dutch and also not for politeness reasons at all; while in other contexts, passive sentences in Dutch are used for reasons of politeness. Furthermore, politeness is a pragmatic effect related to Agentdemotion, not to predicate-promotion, while passive imperatives are related to predicate-promotion (cf. Winkler 1985). Cf. Dik (1997), see under Perspective. For an overview and discussion of these and other notions associated with the term Perspective, see notes 39 and 41.

104

Pierre Winkler

immediately suggest some sort of perspective. In other words, the fact that the missionaries used the term nominative in the way they did, can by itself hardly be considered to be a proof that they used perspective as a functional or pragmatic concept in their linguistic analyses. However, more substantial proof of the thesis that PMG uses perspective as a linguistic concept can be found in the way the missionaries explain the use of other case names. For instance, in example (2) given above, the missionary speaks of a thing that ‗goes from the agent (which becomes the oblique agent, expressed by the genitive) to another party, which has to become PTF (expressed by the nominative). In examples (3) and (4) the linguistic analysis is based on the attribution of a locative function, which in itself has a semantic but also a perspective function (as being the place where, the place from, or the thing or situation about). Other fragments as well indicate strongly that the missionaries explicitly used perspective functions to explain linguistic phenomena. To give some examples: (18)

La particula en … significa ser hecha la cosa ad intra como atryendolo assi, a difirencia de y, que significa ser hecha la cosa ad extra, como echandolo hazia fuera…., a diferencia de an, que significa lugar…54 ‗The particle en means that the thing is done inwardly directed, as attracting it to itself, different from the particle y, which signifies that the thing is done outwardly directed, as throwing it away…, different from the particle an, which signifies place…‘

(19)

De esta segunda pasiva de y se usar para cosas que dicen echar hacia fuera la cosa que va de la persona que hace a otra parte; que estas acciones son las que llamamos adextra55 ‗This second passive is used for other things which express throwing away the thing that goes from the person who acts to another party; these actions we call outwardly directed.‘

(20)

…digo que no se significa por esta composicion en sino al término a quo de la accion…porque el término ad quem …56 ‗I say that this construction with en distinguishes itself by the notion from where the action…, because the notion to whom… ‗

The examples (1) through (4) and (18) through (20) strongly suggest that the missionaries attributed a perspective function to the passive. I will come back to this in the next section. In examples (18) through (20), this notion of perspective is expressed in terms of foregrounding and backgrounding, distance and movement. Examples (1) through (4), however, indicate that they used perspective as an even more fundamental starting 54 55 56

Anónimo (1601: 96). San Agustin (1795[1647]: 134). López (1628: 165).

Subject, Topic, Passive and Perspective

105

point in their analyses, by using the case names in their original senses of perspective and from there on attributing specific semantic and pragmatic meanings to case names. So, the missionaries may have followed the following process in attributing semantic and pragmatic functions to the case names: Nominative Perspective meaning: ‗Centre of attention‘, ‗Subject of talk‘, ‗Starting point of a sentence‘, etc. Hence its meaning in PMG: PTF. Accusative Perspective meaning: ‗Movement‘, ‗Direction‘, ‗Space or Time‘, ‗Causal‘, ‗Instrumental‘. Hence in PMG: Goal, Patient, Instrumental, Temporal.57 Genitive Perspective meaning: ‗belonging to‘, ‗by whom the action is undertaken‘. Hence in PMG: Oblique agent (backgrounded agent). Dative Perspective meaning: ‗to whom it is given‘. The same function (beneficiary) in PMG Locative Perspective meaning: ‗place where‘, ‗from where‘, ‗about what‘. The same functions in PMG. The given perspective meaning of the genitive as ‗by whom the action is undertaken‘, the meaning the missionaries used to attribute the function of oblique agent, can be found in Latin as well, for instance: (21)

Adventus hostium populum terrebat. ‗The coming of the enemies terrified the people.‘58

Latin examples of the perspective meaning of the accusative the missionaries used to attribute the goal function to this case can easily be found and don‘t need further explanation; it is simply the accusative of the Object in a sentence. Examples the missionaries may have used to attribute the movement and the temporal function to this case can easily be found as well in Latin, for instance: 57

58

Thus far, I haven‘t found examples of temporal constructions in the texts of the missionaries working in the Philippines. However, from this overview of perspective functions, it may be expected that if examples of temporal constituents are found, they will be attributed the accusative case. Hengeveld & Mackenzie give other examples from Turkish of this function of the genitive as denoting the agent (2008: 359ff.). However, unlike from the missionaries, they attribute the Subject function to this use of the genitive, while in PMG the term Subject is not considered applicable (see section 7).

106 (22) (23)

Pierre Winkler

Romam ‗to Rome‘ totam hiemem ‗during the whole winter season‘

The attribution in PMG of the instrumental function to the accusative may seem a little bit more puzzling. In Latin, one can find examples of the instrumental meaning of the accusative, in the sense of ‗in the capacity of‘ or ‗with the task of‘, in sentences like: (24)

senilem aetatem onerare tantos dolores ‗to burden his old age with such grief‘

(25)

Ciceronem consulem creaverunt ‗they made Cicero consul‘

Moreover, the original meaning of the accusative in Latin is one of causality and instrumentality. Ernout and Thomas explain in their work on Latin syntax, that the Latin term ‗accusativus‘ is actually a wrong translation by Roman grammarians of the Greek αἰτιατικὴ πτῶσις , which actually means ‗cause‘ or ‗effective case‘. Compare this with the adjective αἰτιατός, which means ‗produced by a cause‘. As Ernout and Thomas show, Priscian was aware of this causal/instrumental origin of the term accusative and so, instead of using the term ‗accusativus‘, he uses the term ‗causativus casus‘.59 So, Priscian – or one of the later grammarians who followed the descriptions of Priscian,60 may well have been the source for the missionaries to attribute the instrumental function to the accusative. Nebrija, however, certainly has not been a primary source altogether with respect to the perspective function of cases, as we can gather from the list below of Nebrija‘s description of why cases have their respective Latin names.61 Most of the differences in usage of the case names compared with the way they are used in PMG speak clearly for themselves. However, the ablative and the ‗effective case‘ Nebrija takes up in his list, need special attention. Let me first give the list of descriptions as given by Nebrija: Nominativus Quia per eum nominamur. ‗Because by this [i.e. case] we are named.‘ 59 60

61

Cf. Ernout & Thomas (1953: 17ff.). The traditional Latin school grammars of Donatus (~350 A.D.) and Priscian (~500 A.D) ―dominated Western linguistic thought for more than a millennium‖ (Bossong 2007: 136). Nebrija (1996[1488]: 105ff.). Esparza Torres (2007) argues that all the missionary linguists of the 16th and 17th century accepted the authority of Nebrija‘s Latin grammar. However, his research only covered texts from missionaries working in Mexico. Calvo Pérez (2000: 126) rightly says that Nebrija‘s grammar inspired all the grammatical studies of the ‗new world‘ (the Americas). Indeed, his inspiration or authority may have been substantial in the linguistic descriptions of languages of the ‗new world‘, but is less tangible in the analyses of linguistic works of Philippine languages discussed in this paper.

Subject, Topic, Passive and Perspective

107

Genitivus Quia per eum nominamur cuius sumus filii. ‗Because by this we are named of whom we are the son.‘ Dativus Quia in eo casu ponimus eum cui aliquid damus. ‗Because in this case we place him to whom we give something.‘ Accusativus Quia in eo casu ponimus eum quem accusamus. ‗Because in this case we place him who we accuse.‘ Ablativus Quia in eo casu ponimus eum a quo aliquid auferemus. ‗Because in this case we place him of whom we take something away.‘ Effectivus Quia in eo casu ponimus instrumentum quo aliquid efficitur. ‗Because in this case we place the instrument by which something is effected.‘ Contrary to Nebrija, the missionaries did not use the ablative case. This, in fact, does corroborate my thesis that they went back to the original perspective meanings of the case names, since the ablative is a syncretistic case of three functions: ‗point of departure‘, combined with instrumental and locative.62 A case with such diverse functions probably was considered inadequate to explain clearly and consistently the perspectives which in PMG form the basis of the analysis of various constructions of sentences. Nebrija has added a seventh case63, which he calls the effectivus, which is actually identical to the instrumental case I discussed before. Apparently, Nebrija himself also realised that the syncretism of three functions in one ablative case would be a bit confusing (in the list given above, he actually attributes only one of the functions to the ablative). However, Nebrija didn‘t follow Priscian in restoring the original instrumental/ causal value of the accusative. So, in this respect as well the missonaries did not follow Nebrija‘s model.

6. Passive as a perspective function in PMG The missionaries labeled all the sentences in which the actor is not PTF as passives. In his contribution to this volume, Lawrence Reid describes (and endorses) Seidenadel‘s criticism in his grammar of Bontok Igorot with respect to the labeling of these sentences as passives. 62 63

Cf. Palmer (1961). In the list I give of Nebrija‘s cases the attentive reader may have counted six cases. I have left out the vocative case because it is not a subject of discussion in this paper.

108

Pierre Winkler

Seidenadel and Reid may be right in rejecting as passives all ―non-actor voice‖ constructions in Philippine languages.64 However, this does not justify Seidenadel‘s criticism of the missionaries‘ use of this term. The examples given in the introduction of this paper and the description in the previous section of the way the missionaries used case names, all make abundantly clear that they didn‘t use the term Passive in a syntactic sense. Just as with the case names, the missionaries moved the term Passive to the pragmatic and perspective level. The choice between passive and active, in their view, is a conscious pragmatically motivated choice of a speaker in order to present a state of affairs from a different perspective. This view of the missionaries is completely in concordance with the pragmatic view of the passive as described in section 4. It even goes a step further. By defining case names on a perspective level and moving the term passive also to this level, they appear to have abandoned the syntactic level almost entirely in their linguistic analyses. I will go into this with respect to their use of the term Subject in the next section.

7. Subject in PMG As it appears, the missionaries didn‘t need the syntactic level of Subject/Object, being well aware that, contrary to Dik‘s thesis, perspective is not the essential nature of this syntactic pair but the essential nature of case names. However, this does not fully explain why they used the term Subject with non-agentive sentences but didn‘t use it with agentive sentences. A possible reason, given by Ridruejo (2007), may be that in agentive sentences there are almost always other constituents involved, for which they used case names such as genitive and accusative, and in concordance with this they also used the term nominative, instead of Subject. In non-agentive sentences they could simply use the term Subject for the Topic of the sentence. Another possible reason may be that non-agentive sentences are very static, so they may have thought that using terms to which they had given a predominantly pragmatic, semantic and perspective function would be inappropriate and even confusing. However, the more plausible reason may well be that they were aware of the fact that Subject isn‘t a valid concept in ergative languages.65 Schachter, for instance, argues quite convincingly that Subject is a composite of Topic and Actor, and in fact, the missionaries may well have fully agreed with this view. In full conformity with Schachter 64

65

However, see the discussion in section 4 of this article. Reid‘s principal argument that the minimal condition for a passive is a downgraded agent, does not seem to be valid. Cf. (for instance) Schachter (1976), Starosta (2001). Dik, however, doesn‘t agree with their view and says that Subject and Object are also valid concepts in ergative languages, but on the condition that these terms are redefined as perspective functions. Earlier in this paper I have already argued why he is mistaken in this respect.

Subject, Topic, Passive and Perspective

109

they also ‗decomposed‘ the notion of Subject into the two distinct functions of Topic – or better said, also in the case of Schachter, PTF,66 and Actor. Compare also Falk:67 Schachter (1976) argues that the familiar notion of Subject is an amalgam of two distinct functions. A similar suggestion has been made on the basis of ergative languages (e.g. Dixon 1979; 1994). One function, which Schachter calls Topic and Dixon calls Pivot, is (roughly) a discourse function. The other, which Schachter calls actor and Dixon Subject, is linked to semantics and argument structure. In nominative-accusative languages, the same nominal has both functions. Or compare Starosta (2001): Languages need some way to distinguish Actor from Patient so that the hearer knows who is performing the action and who or what is being affected. The mechanism that accomplishes this is the case form Nominative. Nominative is like a show window that highlights either Patient or Actor… Compare this with a final example taken from Sanvitores‘ description of Chamorro:68 (26)

Passiva fit per in loco um activae collocandum, construiturque cum nominativo rei et genitivo personae agentis, dativoque personae cui, quae etiam nonnunquam ponitur in nominativo… ‗The passive voice is made by placing ‗in‘ instead of ‗um‘ and is constructed with a Goal as PTF [nominativo], with the Actor as Oblique Agent [genitivo] and with a Beneficiary [dativo]. The Beneficiary can also be made PTF …‘

In PMG we recognize a view on the Philippine languages that is remarkably similar to the modern view of for instance Schachter and Starosta. The missionaries did indeed make a distinction between Actor and Topic, they defined Topic as a pragmatic discourse function (PTF) and neatly placed Actor on the semantic level (together with other semantic functions), and they regarded nominative as being a ‗highlighting show window‘.69 Fur66

67 68

69

See note 4. Schachter (1976) describes Topic as ‗the center of attention‘, which is actually a description of Focus. Falk (1999). See also Falk (2006) and Hirano (2006). Sanvitores (1954[1668]: f30r). For a discussion of this fragment see Winkler (2007). Sanvitores also describes the active in a fully functional way (Sanvitores 1954[1668]: f29v): ―Construitur … quasi Latine, persona agens in nominativo, persona cui datur in dativo, res quae datur in accusativo absque praepositione.)‖ (The active voice is constructed as in Latin, with the Agent as PTF, a Beneficiary and a Goal-Object, without preposition). Sanvitores‘ description here of the um-infix as placing Focus on the Agent is confirmed in a research by Cooreman (1987: 42). For further discussion of this fragment, see Winkler (2007). The rejection of the term Focus, as being misleading or inappropriate, by Himmelmann (2005, 2002) and Hirano (2005) is based on the use of the term Focus in the ‗classical period‘ (for this term see note 5), where it refers to syntactic prominence, in which case it would be better to replace it with Subject. Focus, however, in the sense of pragmatic salience, is a universally applicable con-

110

Pierre Winkler

thermore, in conformity with Dik, they argued that nominative, in its definition of PTF, can have all kinds of semantic functions, not only Actor and Patient.70

8. Conclusions 1. In PMG case names are used in their semantic and perspective meanings. 2. Perspective in PMG is used as a pragmatic linguistic concept, reflecting the way a speaker wants to present a SoA in terms of foregrounding, backgrounding, movement, Topic, Focus, direction, contrast, etc. 3. Different from the missionary linguists in the Americas, Philippine missionaries did not follow Nebrija‘s ‗Introductiones Latinae‘ as a reference model regarding the use of case names. They developed their own method of explaining linguistic phenomena and defined cases in a different way from the case descriptions found in Nebrija. With respect to the instrumental function they attributed to the accusative, they may have been inspired by Priscian‘s definition of the accusative. 4. In PMG the syntactic notion Subject is considered inapplicable in agentive sentences. 5. In FG the level of Subject (and Object) should be considered inapplicable or irrelevant as well, in the sense that it does not contribute to the pragmatic presentation of a SoA. FDG lacks formal precision in this matter, on the one hand referring to the perspective function of Subject and on the other hand arguing that the syntactic level neither adds nor subtracts pragmatic meaning. 6. The definition of Subject and Object in FG as having a perspective function is mistaken. 7. In the still ongoing discussion about whether Subject or Topic is a more applicable term in the linguistic analyses of Philippine languages, the missionaries appear to have prefigured the view of modern linguists like Schachter in preferring to use the term Topic. 8. The missionaries used the term Passive not in the sense of a syntactic structure different from the active. They used the term Passive to describe pragmatic variations in Perspective.

70

cept, so also in Philippine-type languages. Starosta‘s notion of ‗highlighting‘ is very comparable with the pragmatic term Focus. The concept of movement as an important linguistic factor – expressed by the missionaries in examples (1)–(4) and (18)–(20) – is also recognized in Localist Case Grammar. This model assumes that every clause expresses a locative relation (physically or metaphorically), that there are three core semantic roles (Agent, Theme and Loc) and that other semantic roles, like Beneficiary and Instrument, are nonnuclear Themes or Locs. This way of working with the concept of movement goes a step further than the way the missionaries used movement to explain linguistic phenomena. For attempts to use Localist Case Grammar in Philippine linguistics, see Brainard (2006).

Subject, Topic, Passive and Perspective

111

9. The PMG method of moving the term Passive (and Active) to the pragmatic level of linguistic analysis, and taking into account that the most frequently occurring pragmatic motivation for using a passive construction in languages like Dutch and Latin is not patient-promotion nor agent-demotion, but promotion of the predicate, may well lead to different conclusions in the current debate concerning whether or not Philippine-type languages have passives. 10. As the missionaries already understood, Perspective is a powerful pragmatic function. Perspective deserves to be more accurately described and more extensively applied in modern functional or pragmatic linguistic theories like FG and FDG. 11. To the three periods of linguistic analysis of Philippine-type languages that Liao (2004) distinguishes, a fourth period should be added. This period, actually the first one chronologically, during which the missionaries made the first analyses of Philippine-type languages, may well be called the ‗pioneer period‘. For in all kinds of ways pioneers is what they were. Confronted with languages that were not known before and that didn‘t fit the classical syntax and grammar system they knew, they maintained classical terms but redefined them and developed an at that time new method of analysis, which we now call a functional or pragmatic method, and by doing this they were actually four hundred years ahead of their time.

Bibliography A. Primary sources Anónimo (1601): Arte de la lengua sambala y espanola. Compuesto por un religioso de la orde de Nuestro P.S. Agustín Ministro que had sido por muchos años en ella; para ayuda de los Nuevos Operarios que vinieren a la Administración. Dedicado al gran Apostol de España Santiago. Manuscrito de 268 páginas. Archivo de los PP. Agustinos Filipinos de Valladolid. Blancas, Farncisco de S. Joseph (1997 [1610]): Arte y reglas de la lengua tagala. En el Partido de Bataan, por Thomas Pinpin Tagalo, Tercera Edition, impreso en la imprenta nueva de don José María Dayot, por Tomás Oliva, Madrid, 1832; Edición facsímil de la edición princeps con Estudio de Antonio Quilis, Madrid, Ediciones de Cultura Hispánica, 1997 (BICRES II, 49 y 170) Coronel, Francisco (1621): Arte y reglas de lengua pampanga. Compuesto por el Padre Predicador del Orden de N.P.S. Agustín y Prior Ministro del Convento de Macabele, manuscrito. [A lápiz] Acabado de copiar el 1677, Biblioteca de PP. Agustinos (Valladolid). 45 páginas de un volumen facticio con textos en tagalo (BICRES II, 331). López, Francisco (1628): Arte de la lengua ilocana. Manila. En el Colegio i Universidad de S. Thomas de Aquino, por Thomas Pipin, 1617 (BICRES II, 436). Nebrija, Elio Antonio de (1996 [1488]): Introductiones latinas, contrapuesto el romance al latín (c.1488). Miguel Ángel Esparza and Vicente Calvo (eds.). Münster: Nodus Publikationen San Agustín, Andres de (1795 [1647]): Arte del idioma bicol para la enseñanza de este idioma en la provincia de Camarines. Manila 1617. Dispuesto y ordenado por Nuestro hermano Fr. Andrés de San Agustín … segunda vez reimpreso … en el Convento de Nuestra Señora de Loreto del Pueblo de Sampaloc: poe el Hermano Pedro Arguelles de Concepción, 1795.

112

Pierre Winkler

Sanvitores, Diego Luis de (1954 [1668]): Lingua Mariana. Micro-Bibliotheca Anthropos, Vol. 14. 46pp. duplices. Posieux prope Friburgum Helvetiorum, 1954.

B. Secondary sources Adelaar, Alexander (2005): The Austronesian languages of Asia and Madagascar: A historical perspective, in: Adelaar, Alexander & Himmelmann, Nikolaus P. (eds.), 1–42. Adelaar, Alexander & Himmelmann, Nikolaus P. (eds.) (2005): The Austronesian languages of Asia and Madagascar. London and New York: Routledge. Akamine, Jun (1996): Sinama passive constructions. Diliman: University of the Philippines. (avalaible online via http://sealang.net/sala/archives/pdf4/jun1996sinama.pdf) Barlaan, Rodolfo Rosario (1999): Aspects of focus in Isnag. Manila : Linguistic Society of the Philippines Barrett, Ward (1975): Mission in the Marianas: An account of Father Diego Luis de Sanvitores and his companions 1669–1670. Minneapolis: University of Minnesota Press. Bloomfield, Leonard (1917): Tagalog texts with grammatical analysis. Urbana, Ill.: University of Illinois. Bloomfield, Leonard (1933): Language. Chicago: Holt Reinhart and Winston Inc. (reprint 1984, Chicago: The University of Chicago Press). Bolkestein, A. Machtelt (1985): Cohesiveness and syntactic variation: Quantitative vs. qualitative grammar, in: Bolkestein, A. Machtelt; de Groot, Casper & Mackenzie, J. Lachlan (eds.), Syntax and pragmatics in functional grammar. (Functional Grammar Series 1). Dordrecht: Foris, 1–14. Bossong, George (2007): The influence of missionary descriptions of Far Eastern languages on western linguistic thought: The case of Cristofori Borri, S.J. and Tommaso Campanella, in: Zwartjes, Otto; James, Gregory & Ridruejo, Emilio (eds.), Missionary linguistics III. Amsterdam/Philadelphia: John Benjamins, 123–143. Brainard, Sherri (1994): Voice and ergativity in Karao, in: Givón, Talmy (ed.), Voice and inversion. Amsterdam/Philadelphia: John Benjamins, 365–402. Brainard, Sherri (1997): Ergativity and grammatical relations in Karao, in: Givón, Talmy (ed.), Grammatical relations. Amsterdam/Philadelphia: John Benjamins, 85–154. Brainard, Sherri (ed.) (2006): Localist case grammar & Philippine verbs. Manila : Linguistic Society of the Philippines. Brainard, Sherri & van der Molen, Ena (2005): Word order inverse in Obo Manobo, in: Liao, Hsiuchuan & Rubino, Carl R. Galvez (eds.), Current issues in Philippine linguistics and anthropology, Parangal kay Lawrence A. Reid. Manila: The Linguistic Society of the Philippines and SIL Philippines, 364–418. Brown, Penelope & Levinson, Stephen C. (1987): Politeness – some universals in language usage. Cambridge: Cambridge University Press. Burrus, E. J. (1954): Sanvitores‘ grammar and catechism in the Mariana language, in: Anthropos 49, 934–939 (microfilm). Calvo Pérez, Julio (2000): Las gramáticas del Siglo de Oro quechua: Orginalidad y diversidad, in: Zwartjes, Otto (ed.), Las gramáticas misioneras de tradición hispánica (siglos XVI y XVII). Amsterdam & Atlanta: Rodopi, 125–204. Cooreman, Ann M. (1987): Transitivity and discourse continuity in Chamorro narratives. Berlin; New York [etc.]: Mouton de Gruyter. Cornelis, Louise (1997): Passive and perspective. Amsterdam; Atlanta: Rodopi Coulter, H. George (2005): Expressions of agency in Ancient Greek. Cambridge: Cambridge University Press.

Subject, Topic, Passive and Perspective

113

Dery, Jeruen (2007): Pragmatic focus and word order in Tagalog, in: Language and linguistics 8(1), 373–402. Donahue, Mark (2002): Voice in Tukang Besi and the Austronesian voice system, in: Wouk, Fay & Ross, Malcolm (eds.), The history and typology of Western Austronesian voice systems. Canberra: Pacific Linguistics 518, 81–99. Dietrich, Wolf (2008): Missionary Linguistics III — Lingüística misionera III: Morphology and Syntax. Selected Papers from the Third and Fourth International Conferences on Missionary Linguistics, Hong Kong/Macau, 12–15 March 2005, Valladolid, 8–11 March 2006, in: Historiographia Linguistica 35(3), 439–445. Dik, Simon C. (1981): Functional grammar. Dordrecht: Foris Publications. Dik, Simon C. (1997): The theory of functional grammar. Edited by Kees Hengeveld. New York: Mouton de Gruyter. Dixon, Robert M. W. (1994): Ergativity. Cambridge: Cambridge University Press. Ernout, Alfred & Thomas, François (1953): Syntaxe Latine. Paris: Klincksieck. Esparza Torres, Miguel Ángel (2007): Nebrija y los modelos de los missioneros lingüistas, in: Zwartjes, Otto; James, Gregory & Ridruejo, Emilio (eds.), Missionary linguistics III. Amsterdam/Philadelphia: John Benjamins, 3–40. Falk, Yehuda N. (1999): Philippine subjects in a monostratal framework. Paper presented at the 6th Annual Conference of the Austronesian Formal Linguistics Association, 16–18 April 1999, University of Toronto. Falk, Yehuda N. (2006): Subjects and universal grammar: An explanatory theory. Cambridge: Cambridge University Press. Givón, Talmy (1994): The pragmatics of de-transitive voice: Functional and typological aspects of inversion, in: Givón, Talmy (ed.), Voice and inversion. Amsterdam/Philadelphia: John Benjamins, 3–46. Givón, Talmy (1995): Functionalism and grammar. Philadelphia: John Benjamins. Givón, Talmy (2001): Syntax: A functional-typological introduction. Volume II. Amsterdam/Philadelphia: John Benjamins. Hengeveld, Kees & Mackenzie, Lachlan (2008): Functional discourse grammar: A typologicallybased theory of language structure. Oxford: Oxford University Press. Himmelman, Nikolaus P. (2002): Voice in Western Austronesian: An update, in: Wouk, Fay & Ross, Malcolm (eds.), The history and typology of western Austronesian voice systems. Canberra: Pacific Linguistics, 7–16. Himmelmann, Nikolaus P. (2005): Tagalog, in: Adelaar, Alexander & Himmelmann, Nikolaus P. (eds.), 350–376. Hirano, Takanori (2006): Subject and topic in Tagalog. Paper presented at the 10th International Conference on Austronesian Languages (available via: http://homepage.ntu.edu.tw/~gilntu/data/workshop%20on%20Austronesian/2%20hirano.pdf) Kazenin, Konstantin I. (2001): The passive voice, in: Haspelmath Martin (ed.), Language typology and universals: An international handbook. Volume 2. Berlin: Walter de Gruyter, 899–916. Kess, Joseph (1979): A note on the Tagalog passive in the Totanes manuscript, in: Nguyen Dang Liem (ed.), South-east Asian Linguistic Studies. volume 4. Canberra: Research School of Pacific Studies, 233–239. Klamer, Marian (2005): Kambera, in: Adelaar, Alexander & Himmelmann, Nikolaus P. (eds.), 709– 734. Kroeger, Paul R. (1993a): Another look at subjecthood in Tagalog, in: Philippine Journal of Linguistics 24(2), 1–16.

114

Pierre Winkler

Kroeger, Paul R. (1993b): Phrase structure and grammatical relations in Tagalog. California: CSLI Publications. Lambrecht, Knud (1994): Information structure and sentence form. Cambridge, New York, Cambridge University Press. Liao, Hsiu-chuan (2004): Transitivity and ergativity in Formosan and Philippine languages. Ph.D. Dissertation. University of Hawai‗i. Malchukov, Andrej (2006): Transitivity parameters and transitivity alternations: Constraining covariation, in: Kulikov, Leonid; Malchukov, Andrej & de Swart, Peter (eds.), Case, valency and transitivity. Amsterdam/Philadelphia: John Benjamins, 329–357. Marín Arrese, Juan I. (1997): Cognitive and discourse-pragmatic factors in passivisation, in: Atlantis 19(1) (June 1997 [issued February 1999]), 203–218. Mithun, Marianne (1994): The implications of ergativity for a Philippine voice system, in: Fox, Barbara & Hopper, Paul J. (eds.), Voice, form and function. Amsterdam/Philadelphia: John Benjamins, 247–277. Ogawa, Aiko (2009): Case in a topic-prominent language: Pragmatic and syntactic functions of case in Japanese, in: Malchukov, Andrej & Spencer, Andrew (eds.), Handbook of case. Oxford: Oxford University Press, 779–788. Payne, Thomas E. (1994): The pragmatics of voice in a Philippine Language: Actor-focus and goalfocus in Cebuano narrative, in: Givón, Talmy (ed.), Voice and inversion. Amsterdam/Philadelphia: John Benjamins, 317–364. Palmer, Leonard R. (1961): The Latin language. London: Faber and Faber Ltd. Primus, Beatrice (2009): Case, grammatical relations and semantic roles, in: Malchukov, Andrej & Spencer, Andrew (eds.), Handbook of case. Oxford: Oxford University Press, 261–275. Reid, Lawrence A. & Liao, Hsiu-chuan (2004): A brief syntactic typology of Philippine languages, in: Language and Linguistics 5(2), 433–490. Ridruejo, Emilio (2004): La Tradicion gramatical sobre el Pampango, in: Zwartjes, Otto & Hovdhaugen, Even (eds.), Missionary linguistics. Amsterdam/Philadephia: John Benjamins. 179–200. Ridruejo, Emilio (2007): Sujeto y tópico nominal en las gramáticas Filipinas, in: Zwartjes, Otto; James, Gregory & Ridruejo, Emilio (eds.), Missionary linguistics III. Amsterdam/Philadelphia: John Benjamins, 233–250. Schachter, Paul (1976): The subject in Philippine languages: Topic, actor, actor-topic, or none of the above, in: Li, Charles (ed.), Subject and topic. New York: Academic Press, 493–518. Schmalstieg, William N. (2004): The common origin of the *-o stem dative, accusative and instrumental cases, in: Baltistica 39(1), 5–11. Seuren, Pieter (1998): Western Linguistics: An historical introduction. Oxford: Blackwell Publishers Ltd. Seidenadel, Carl Wilhelm (1909): The first grammar of the language spoken by the Bontoc Igorot with a vocabulary and texts: Mythology, folklore, historical episodes, songs. Chicago, London: Open Court Publishing Company; Kegan Paul, Trench, Trubner & Co., Ltd. Shibatani, Masayoshi (1985): Passives and related constructions: A prototype analysis, in: Language 61, 821–848. Shibatani, Masayoshi (2009):Case and voice, in: Malchukov, Andrej & Spencer, Andrew (eds.), Handbook of case. Oxford: Oxford University Press, 322–338. Starosta, Stanley (2001): Off the subject. Paper presented to the Philological Society. Cambridge: Cambridge University. Van Valin, Robert D. (2009) Case in role and reference grammar, in: Malchukov, Andrej & Spencer, Andrew (eds.), Handbook of case. Oxford: Oxford University Press, 59–71. Van Valin, Robert D. & LaPolla, Randy J. (1997): Syntax: Structure, meaning and function. Cambridge and New York. Cambridge University Press.

Subject, Topic, Passive and Perspective

115

Wickboldt, Jun (2000): Some effects of manner adverbials on meaning, in: Tenny, Carol & Pustejovsky, James (eds.), Events as grammatical objects. California: CSLI Publications, 359–374. Winkler, Pierre (1985): Pragmatische en stilistische functies van het passief, in: Tijdschrift voor Talen Tekstwetenschap 5(1), 65–91. Winkler, Pierre (2007): Functional grammar in Austronesian linguistics, in: Zwartjes, Otto; James, Gregory & Ridruejo, Emilio (eds.), Missionary linguistics III. Amsterdam/Philadephia: John Benjamins, 329–344. Zimmermann, Klaus (2004): La construcción del objeto de la historiografía, in: Zwartjes, Otto & Hovdhaugen, Even (eds.), Missionary linguistics. Amsterdam/Philadephia: John Benjamins, 7–32. Zobel, Erik (2005): Buol, in: Adelaar, Alexander & Himmelmann, Nikolaus P.(eds.), 625–648.

IWAR WERLEN (BERNE/SWITZERLAND)

Wilhelm von Humboldt and the role of the verb in Tagalog

Abstract This paper examines the analysis of the Tagalog verb by Wilhelm von Humboldt considering the background of what was known to European linguists of Humboldt‘s time. Comments on Tagalog by authors such as Franz Carl Alter, Simon Peter Pallas, Don Lorenzo Hervás y Panduro and Johann Christoph Adelung are looked at and a short description of the Arte de lengua Tagala of Sebastian de Totanes is given. All authors use the framework of the Greco-Latin tradition. It is shown that Humboldt moves from a universally conceived verb as it is found in Indo-European languages to a more relativistic view of a nominally based verb structure in Tagalog.

1. Introduction In the history of linguistics, Wilhelm von Humboldt (1767–1835) is often seen as the founder of general linguistics (see Arens 1969: 203ff.). His main contribution seems to be a theoretical one, the Introduction to what is known as the Kawi work with the German title ―Ueber die Verschiedenheit des menschlichen Sprachbaues und ihren Einfluß auf die geistige Entwicklung des Menschengeschlechtes‖.1 Many of his readers, however, were not aware of the fact that the introduction was part of the first of three volumes published posthumously by J. C. E(duard) Buschmann and that it was followed by the results of Humboldt‘s empirical studies of grammars and texts of the Austronesian languages. Humboldt had collected a great number of grammars, dictionaries and other linguistic materials on all languages available and he made notes about many of them. These sources of his empirical work were not very well known until the publication of Mueller-Vollmer (1993) and of Schwarz (1993). The former published and commented 1

This text has been edited in four somewhat differing versions: the first one was published after Humboldt‘s death with a foreword by his brother Alexander and it was sent to Humboldt‘s correspondents and friends all over the world; the second and third were published in the two editions of the first volume of the Kawi work (Humboldt 1836) – the one published by the Academy, the other one by Dümmler; the fourth edition is the text published by A. Leitzmann (1907) in the edition of Humboldt‘s works by the Academy; see De Cesare (ed.) (1998) for the last edition of the text (known to us). Humboldt‘s text was translated into English by Peter Heath (1988).

118

Iwar Werlen

on Humboldt‘s unpublished linguistic materials following his death; the latter compiled a list of the books given by Humboldt to the then Royal Library of Berlin (today the Library of the Humboldt University). Interested in Tagalog2 at the time I first read the two books, I only found one small volume on Tagalog listed in Mueller-Vollmer (1993) and a few grammars and dictionaries in Schwarz (1993). After having read Humboldt‘s own text on the Tagalog verb in volume II of the Kawi work, I became aware that his sources were essentially Spanish written grammars by Spanish friars living in the Philippines and I was interested in what European linguists of Humboldt‘s time outside the Spanish friars‘ circles knew about Tagalog.

2. The knowledge of Tagalog by European linguists during Humboldt‘s time The only work on Tagalog listed in Mueller-Vollmer (1993) is a small booklet published by Friedrich Carl Alter (1749–1804) with the title Ueber die tagalische Sprache [On the Tagalog Language], published 1803.3 According to the Allgemeine Deutsche Biographie, Alter was a well-known linguist (ADB 1 1875: 363–364). He had been a Jesuit but after the abolition of this religious order by Pope Clemens XIV in 1773, he became the curator of the library of the University of Vienna where he taught diplomatics. In the foreword of his booklet Alter says that he was visited in 1799 by a Spanish educator called Fabera. Fabera showed him a hand-written Dictionarium HispanoTagalog um. Alter continues by saying that at the time he knew nothing about the Tagalog language. But then he found some words listed in Simon Peter Pallas‘ Linguarum totius orbis vocabularia comparativa (1786–1789). Pallas‘ work lists translations of Russian words into 200 other languages; number 187 is Tagalog (po tagalanski). It is not clear who or what was the source of Pallas‘ Tagalog words as many translations are missing. Alter tried to fill in the gaps in the hand-written dictionary he had received from Fabera. In some cases, he did not read the words correctly. Pallas no. 41 ―Genu. Knie‖ is given by Alter as To Eor; the word must have been tohor (modern Tagalog tuhod ‗knee‘); and on no. 85 ―Nix. Schnee― he comments ―Perhaps there is no snow in the region where Tagalog is spoken (my translation)‖ (28). For no. 109 ―Vallis. Tal‖ he provides a circumscription which he did not understand: ―Lupang pantay nanatotonghan. Nangbondoc‖ (Alter 1803: 32) or in modern orthographic style:

2

3

I would like to thank to an anonymous reviewer for his/her very valuable comments on some topics of Philippine linguistics related to the text of this article which lead to substantial revisions of the text. Alter (1803) is listed in Ward (1971: 67f.), the most comprehensive bibliography of Philippine Linguistics.

The role of the verb in Tagalog

(1)

119

Lupang land-LK ng

pantay na natotonghan flat LK INCOMPL-Stative-direction-LOC (situated) bundok GEN mountain ‗flat piece of land situated at a mountain‘

Alter, it seems, had no idea where Tagalog was spoken and he had no access to other sources than Pallas, the Fabera dictionary and some of the works of Don Lorenzo Hervás y Panduro (1735–1809). Hervás y Panduro had been a Jesuit like Alter; he collected a lot of hand-written grammars and texts from his fellow Jesuits who, after the abolition of the order, had to leave the regions where they were working as missionaries. In various books, but especially in the Saggio pratico delle lingue con prolegomeni, e una raccolta di orazioni Dominicali in più di trecento lingue, e dialetti, con cui si dimostra l‘infusione del primo idioma dell‘uman genere, e la confusione delle lingue in esso poi succeduta, e si additano la diramazione, e dispersione delle nazioni con molti risultati utili alla storia (1787) [practical outline of languages with introductory remarks, and a collection of the Lord‘s prayer in more than three hundred languages, and dialects, with which is shown the outpouring of the first language of the human kind, and the subsequent confusion of the languages in it, with the addition of the ramification, and dispersion of the nations with many results suitable for history] he published some of the collected material.4 4

It is impossible for us to cite all the places that Tagalog or other Philippine languages (esp. Visayan) are mentioned in the various books of Hervás y Panduro. His main work was the Idea dell'Universo, che contiene la Storia della vita dell'uomo, elementi cosmografici, viaggio estatico al mondo planetario, e Storia della terra, e delle lingue [Idea of the Universe, which contains the History of the life of man, cosmographic elements, the exstatic travel to the planetarian world, and History of the earth, and of the languages] (Cesena, Biasini 1778–1787), containing 21 volumes. Five of these were especially interesting for us: Catalogo delle lingue conosciute e notizia della loro affinità, e diversità [Catalogue of the known languages and notice of their relationship and diversity] (1784); Trattato dell‘Origine, formazione, meccanismo ed armonia dell‘Idiomi [Treatise of the origin, formation, mechanism and harmony of the languages] (1785); Aritmetica delle nazioni e divisione del tempo fra l‘orientali [The number systems of the nationas and the division of time in the east] (1786); Vocabolario poligloto con prolegomeni sopra piu di CL lingue [Polyglot vocabulary with introductory remarks on more than sixty languages] (1786); Saggio pratico delle lingue con prolegomeni, e una raccolta di orazioni Dominicali in più di trecento lingue, e dialetti, con cui si dimostra l‘infusione del primo idioma dell‘uman genere, e la confusione delle lingue in esso poi succeduta, e si additano la diramazione, e dispersione delle nazioni con molti risultati utili alla storia [Practical outline of languages with introductory remarks, and a collection of the Lord‘s prayer in more than three hundred languages, and dialects, with which is shown the outpouring of the first language of the human kind, and the subsequent confusion of the languages in it, with the addition of the ramification, and dispersion of the nations with many results suitable for history] (1787). The Catalogo was published also in a Spanish version with the title: Catálogo de las lenguas de las naciones conocidas, y numeracion, division, y clases de estas segun la diversidad de sus idiomas y dialectos [Catalogue of the languages of the known nations, and counting, divisions,

120

Iwar Werlen

Concerning Tagalog, he published inter alia two versions of the Lord‘s Prayer; the first one was part of the Doctrina Christiana (1593), the oldest printed book in the Philippines (Wolf 1947), the second one was written in contemporary Tagalog. In various parts of his works, one can find Tagalog words and texts, e.g. the text of the Ave Maria from the already mentioned Doctrina Christiana compared to two other versions which were also published in Alter‘s small booklet. Hervás‘ grammatical remarks to his edition of the Lord‘s Prayer in contemporary Tagalog in the Saggio (Hervás 1787: 128) show that he had a far better understanding of Tagalog than Alter. Although he uses the grammatical categories of the Spanish grammarians (e.g. gender, article, tense – Tagalog has no gender at all, what he calls ―article‖ is better seen as a case marker (ang) and most grammarians agree that there is no tense system, but a system of aspects), he is able to identify various word formations correctly and he gives a more or less correct translation. In his text, there are some minor mistakes such as camin instead of cami in line 12 or Et instead of At in line 16 (Hervás 1787: 128) and there are some mistakes also in the text of the Doctrina Christiana (1593) he printed; but most of them are not very severe with the exception of casanan instead of casalanan ‗sin‘ and nagcasasa instead of nagcasasala ‗(we) sinned‘ (both in line 6). We may say that Hervás had a relatively good understanding of the structure of the Tagalog text. He himself writes in this context: La sintassi dell‘idioma Tagalo comparisce assai nota coll‘esposta traduzione letterale, che ho fatta cogli ottimi documenti, che per formare la gramatica, ed il dizionario Tagalo, e Bisayo mi hanno dato i Signori D. Antonio Tornos, e D. Bernardo de la Fuente : i quali documenti uniti ad altri della lingua Malaya, che ho raccolti, mi hanno dato lume per azzardami a fare la traduzione letterale dell‘orazioni, che in altri dialetti Malayi metterò in appresso [The syntax of the Tagalog language seems to be pretty known with the printed literal translation I made with (the help of) the very good documents given to me by Messrs. D. Antonio Tornos, and D. Bernardo de la Fuente in order to form the grammar and lexicon of Tagalog and Bisaya: these documents together with others of the Malay language collected by me gave me the help to dare to do a literal translation of the prayers which I will publish later on in other Malay dialects]5 (Hervás 1787: 128). Another well-known linguist of the time, Johann Christoph Adelung (1732–1806), published two Tagalog versions of the Lord‘s Prayer in Mithridates (1806). The texts are both taken from Hervás y Panduro‘s Saggio. As I have already said, Hervás published

5

and classes of them according to their idioms and dialects] (Madrid, Administración del Real Arbitrio de Beneficencia, 1800–1805). Two of the sources of Hervás y Panduro were missionaries having lived on the Philippines. He seems to have been one of the first authors to have identified the Malayan languages as a genetically related family. I would like to thank my colleague Bruno Moretti who helped me to translate the Italian text.

121

The role of the verb in Tagalog

the text of the Doctrina with some minor mistakes but without any translation. Adelung does not correct these mistakes but adds some more. He consequently uses upper case for elements he thinks to be nouns (as in German). Three of his mistakes were corrected later by Johann Severin Vater in the fourth volume of Mithridates (Adelung 1817: 262): Ngala was corrected to ngalan ‗name‘. Then Adelung‘s word Pahiahari must be read as pagcahari ‗kingdom‘ (according to the Doctrina); this mistake was due to a misprint in the Saggio-text. Vater reads the word as pagchahari, but there is no in Tagalog, thus it must be pagcahari. The third mistake is nang camin Cacanin instead of nang amin cacanin ‗of our (cooked) rice‘. Since Hervás did not give a comment on the text taken from the Doctrina Christiana, Adelung tries to translate it himself. Despite the fact that the Lord‘s Prayer is a well known text, he does not succeed in translating the text. In the first instance, he seems to interpret the form mo (second person singular personal pronoun in the non-topic form) as an auxiliary expressing the passive voice (he is translating it consistently with German sein or werden). Take for example the following Tagalog citation with Adelung‘s translation (superscripted in the original, 609): (2)

Bigian-mo cami ‗Gegeben werde uns

ngai-on nang jetzt der

camin Cacanin ... unser Reis‘

In today‘s orthography, the sentence reads: Bigyan mo kami ngayon ng aming kakanin. Bigyan is an imperative focusing on the recipient (kami ‗we, exclusive‘); the agent is mo ‗you (sg.), non-topic form‘. Adelung chooses a passive translation; this leads him to think that mo is an auxiliary parallel to German werden. In the second half of the text, he has some difficulties in understanding the constructions used. So he stops translating the following sentence after the third word: (3)

At ‗Und

pacavalin-mo vergeben werde

ang-amin Casalanan uns ...‘

He does not recognize that the Tagalog text omits the pronoun (in the Latin text, we read: ―et dimitte nobis debita nostra‖; Tagalog only says ―and forgive our sin‖). Adelung seemingly did not identify the root of kasalanan: sala ‗error, mistake‘ and he was therefore not able to translate the sentence. Adelung also commented on Hervás‘ comments on the younger text. Hervás analysed the form sungmasalangit in the following way: ―Sungmasalangit is composed of the particle sa and ungm which together become or form a verb, in which the particle sa is reduplicated, because it is used in the present tense, and so it means to be at the place which is indicated by langit ‗heaven‘ (my translation)‖ (Saggio 128). Adelung seems to misunderstand Hervas‘ comment, because he interprets sungma as the verb to be and sa langit (correctly) ‗in heaven‘. It is well known that there is no copula-verb expressing ‗to be‘ in Tagalog. The construction type s-um-alangit is not mentioned in Schachter & Otanes (1972: 360), but they have a similar construction sa akin > suma-akin ‗be mine‘.

122

Iwar Werlen

The verbs sumalangit ‗be in heaven‘ and sumalupa ‗be on earth‘ are still used in contemporary Tagalog. Summarizing, one can say that Adelung and Alter – both well known linguists at the end of the 18th century – had only restricted knowledge of Tagalog, while Hervás had access to more information. However, none of them was really aware of the difference of this language to the Indo-European languages they were used to analyzing.

3. The Tagalog verb in the grammar of Sebastian de Totanes (1687–1748) The main source for Humboldt‘s chapter on the Tagalog verb is the Grammar of Sebastian de Totanes (Ward 1971: 403 lists four editions: 1745, 1796, 1850, 1865 all printed in Manila). Humboldt had access to the first (1745) and the second (1796) edition of this grammar. Today both are to be found in the library of the Humboldt University in Berlin. As is well known, the first grammars of Tagalog were written by Spanish missionaries (see Wolf 1947, Phelan 1955, Regalado Trota 1998 (CD)). There were four religious orders in the Philippines (Augustinians, Dominicans, Franciscans and Jesuits). There was a lot of competition between the orders on the one hand and the local clergy on the other, whereby the orders were in competition with each other as well. The Spanish crown tried to settle these rivalries. They decided that every order was allowed to evangelise in the region of Manila; other parts of the islands were divided between the orders. This regulation had the effect that members of every order published grammars and other books in Tagalog while other languages were less described. According to Phelan (1955: 158), each order founded its own printing press at the end of the 16th and the beginning of the 17th century. The books printed were essentially grammars (arte), dictionaries (vocabulario), catechisms, confessionaries (confessionario) and books of the type of the Doctrina Christiana. Phelan says: ―The purpose of the handbooks was to give the friars a sufficient grasp of the particular language to enable them to preach in that tongue and to administer the sacraments.‖ (Phelan 1955: 159). The missionaries pushed Tagalog as the ―general‖ language of the Philippines. Totanes, for example, says at the very beginning of the foreword of his Arte: If any language of the many which are spoken on these islands may be called general, it is Tagalog, because it is spoken in the major district and because this district is the center of all (the districts) in which innumerable people from all parts, islands and provinces come together like at a court, which is Manila. And because a person who expresses himself well in Tagalog, may go through the whole kingdom being sure that he will find in each part somebody who will understand him (my translation) (CD, Prologo). [citations of the arte follow the

The role of the verb in Tagalog

123

third edition from 1850, easily available on a CD of the Colección Clásicos Tavera, Serie IX, vol. 9 ―Obras Clásicas de Lenguas Indigenas de Filipinas‖.] According to Regalado Trota (1998), Totanes was born in 1687 in the town Totanes in the Spanish province of Toledo. He became a member of the Franciscan order. In 1717 he went to the Philippines where he lived in Sampaloc and in Santa Ana de Sapa, both places are situated near Manila. Later on he went to Liliw and to Pagsanhan (both in Laguna province) where he wrote his grammar. The four editions of the grammar (1745, 1795, 1850 and 1865) show essentially the same text but in different formats; the second and third editions show some mistakes due to the new typesetting; unfortunately I was not able to find the fourth edition. In the prologue, Totanes describes the grammar and vocabulary writing tradition of his own order; he insists on his use of the same dialect of Tagalog as his predecessors. Totanes‘ grammar is only part of the book; it is bound together with a Manual Tagalog a collection of texts for administrating the sacraments. For the sake of illustration I am going to cite the whole title: Arte de la lengva tagala, y manval tagalog, para la administracion de los Santos Sacramentos, qve de Orden de svs svperiores compvso Fray Sebastian de Totanes, Hijo de la Apostolica, y Seraphica Provincia de S. Gregorio Magno, de Religiosos Descalzos de la Regular, y mas estrecha Observancia de Nuestro Seraphico Padre S. Francisco de las Islas Philipinas, Para alivio de los religiosos de la misma Sta. Provincia, que de nuevo se dedican a aprender este Idioma, y son Principiantes en la Administracion Espiritual de las Almas. Reimpreso en la Imprenta del vso de dicha Santa Provinicia, sita en el Convento de N~ra. Señora de Loreto en el Pueblo de Sãpaloc Extra-muros de la Ciudad de Manila. Año de 1745 [Grammar of the Tagalog language, and Tagalog Manual, for the administration of the Holy Sacraments which on order of his superiors was composed by Father Sebastian de Totanes, son of the Apostolic and Serpahic Province of St. George the Great of the discalced friars of the regular, but very severe observance of our Seraphic Father St. Francis of the Philippine Islands for helping the friars of the same holy Province who are starting to learn this idiom and who are beginners in the spiritual administration of souls. Printed in the printing press used by the mentioned Holy Province, located in the Convent of Our Lady from Loreto in the village of Sampaloc, Extra-Muros of the City of Manila in the year 1745]. In all editions consulted, the grammar itself consists of four sections. Section one gives some introductory remarks followed by the presentation of proper nouns, common nouns, pronouns of various types and adjectives including comparison. Section two comprises verbs and participles. Section three treats nominal and verbal derivations and compositions. Section four is dedicated to adverbs, interjections, conjunctions and numerals. This section is followed by a treatise on ―Ligaziones‖ or linkers or ligatures (as

124

Iwar Werlen

they are known today.) Linkers do not exist in the Indo-European tradition. Therefore, Totanes does not include them in the frame of a traditional ars grammatica, which is the model of his grammar. He closely follows this model even when it is clear that it is fully inappropriate. As an example, I cite his paradigm of the noun Tavo ‗man‘ (Totanes 1850: 8; modern Tagalog tao ‗human being‘): Nominativo Genetivo Dativo Acusativo Vocativo Ablativo

Ang tàvo sa, l. nang tàvo sa tàvo sa, l. nang tàvo ay, l. oy tàvo sa, l. nang tàvo

El hombre Del Hombre A, l. para el hombre A, l. contra el hombre O la hombre De, en, con, por, sìn el hombre

Table 1: The case forms of the noun tàvo in Sebastian de Totanes (1850: 8) The Latin case system is the conceptual mold Totanes uses here. Even for Spanish, the system is inappropriate and for Tagalog it does not work at all. While ang is seen as presenting the nominative, nang and sa are more or less interchangeable for genitive, ablative and accusative. The vocative is expressed by an interjection just as in Spanish. In his text, Totanes compares the three particles with the articles in Spanish. This is largely incorrect, since the three particles do not have the function of referring to definite or indefinite referents. They may better be seen as case markers (but see Reid 2002 and the literature cited there for a discussion and critique of the categorization of ang as a case marker). But let us now have a look at his presentation of the verb. The chapter on the verb and the participle is the longest of the four sections, it counts 62 pages (roughly half of the whole grammar) in the consulted edition. In chapter one of this section he begins with the verb ‗to be‘. Since there is no such verb in Tagalog, he lists a series of constructions which translate the Spanish auxiliaries ser, estar and haber/tener. Due to lack of space, we will not go into the details of this first chapter. More important for us is the second chapter, where Totanes gives ―general rules‖ for the verb in Tagalog. We will briefly describe these rules. The first rule introduces the term ―raiz‖ for the simple root or stem, e.g. laro ‗play‘. To such a root, a ‗particle‘ is added which is called ―propria particula‖ – in the case of laro it is the prefix mag-. This ‗proper particle‘ decides the ―species‖ the verb belongs to. In the second rule, Totanes introduces the category of mood. He distinguishes four moods: indicative, optative, subjunctive and imperative. The indicative and imperative, he says, are in no way marked. For the optative, the particle nauá ‗ojala, utinam‘ is added to the verb and for the subjunctive, one of the conjunctions cun, nang, nion (or noon) and sa are used to introduce a subordinate clause. Both moods, optative and subjunctive, formally differ in no way from the indicative. The third and fourth rules are dedicated to the tense-system. Totanes says that there are only six tenses in the active and passive verbs. And he continues to say that there is

125

The role of the verb in Tagalog

no number distinction nor person distinction in the verb. In the following table, Totanes‘ tense system is shown: Presente Pret. imperf. Pret. perfecto Pret. plusq. perf. Futuro imperf. Futuro perfecto Imperativo

Naghahánap acó Naghahánap ca cahápun Naghánap siyá Nacapaghánap camí Maghahánap cayo Macahánap silá Maghánap cayó

Yo busco Tu buscabas ayer El buscó Nosotros habíamos buscado Vosotros buscareis Ellos habrán buscado Buscad vosotros

Table 2: Tenses in Tagalog according to Totanes (after his remarks p. 40 for the verb maghanap ‗to search‘) It is interesting to see that here Totanes seems to give the impression that Tagalog has a personal, verbal system. In other words he gives nearly all personal pronouns (with the exception of the 1st person plural inclusive) although he stated in his text that there are no personal distinctions in the Tagalog verb. A closer look at this temporal system shows that the present and preterite do not differ formally; it is only the addition of cahapun ‗yesterday‘ that distinguishes the latter from the former. The perfect preterite shows the form called completed in modern grammars. The ―futuro imperfecto‖ is the same as the contemplated form in modern grammars and the imperative is the same too. More interesting are the two forms for the pluperfect and the ―futuro perfecto‖. Totanes introduces the prefix maca(-pag) which in modern grammars is not seen as expressing tense but ability or mere chance. (Humboldt sees this problem when discussing Totanes‘ explications). The fifth rule introduces what is called the three passives with suffix -in, prefix i- and suffix -an. It is not easy for Totanes to explain that there are three passives given the fact that in the Indo-European languages normally only one passive is attested. He therefore says in rule 6: ―One of the major difficulties in this language is to know how to use these passives according to their meaning (my translation).‖ In order to help the learner of the language, Totanes gives some mnemonics such as: Tempus , & Causa, instrumentum, & quasi Y passivum petunt, semperque anteponitur illis. In rule seven, Totanes discusses when the active form and when one of the passives are to be used. Due to lack of space, we will not discuss these rules in detail. Rule eight concerns the participles. Totanes mentions that the forms are not different from the verbs, but that they receive the case forms indicated by ang, sa and nang, e.g. ang sungmusulat ‗the one who is writing‘. I omit the following rules dealing with the construction of verbs and rule twelve where Totanes gives the total paradigms of tenses, moods, actives and passives in the style of a Latin grammar.

126

Iwar Werlen

The following chapters of section two are dedicated to the ―species‖ of verbs. For Totanes, a species consists of one active affix and one or more passive ones. I refrain from repeating the 10 major and the 7 minor species. There is at least one species which, according to my knowledge, is no longer productive in today‘s Tagalog, e.g. species 13 with the two prefixes manhi- and panhi-. If one looks at the examples given, the prefix should perhaps be analysed as manhing- with an assimilating velar nasal similar to the velar nasal in the prefix mang-. As far as I can see, modern Tagalog has some derived roots with the prefix hing- such as himatay ‗swooning, fainting‘ from patay ‗dead‘, even though Schachter & Otanes (1972) do not list such a prefix as a productive one. Summarizing, one may say that Totanes deals with Tagalog on the basis of GrecoLatin categories. His treatment of the verb posits moods and tenses inappropriate for Tagalog. In the discussion of the species, he compares them to the Latin prefixes such as re- in religo or inter- in intelligo. In some way this is not incorrect as some of the prefixes have a semantic meaning influencing the meaning of the verb as a whole, however, the prototypical ones (-um-, mag-, mang- etc.) indicate primarily the focused element in the sentence and in this case: the actor.

4. Humboldt‘s interpretation of the verb in Tagalog The three volumes of the so-called ―Kawi‖-work of Wilhelm von Humboldt, are a comparative description of the Polynesian and the Melanesian languages. The work was edited posthumuosly by J. C. E(duard) Buschmann who was Humboldt‘s secretary during the last years of his life. Later on, he was responsible for Humboldt‘s unpublished literary materials; partially he ordered Humboldt‘s manuscripts in a new way for publication in the three volumes (cf. Mueller-Vollmer 1993). In the third volume of the work, he added some of his own writings. In the Kawi-book, the chapter on the Tagalog verb is located in the second volume and in the third section titled as follows Über den Malayischen Sprachstamm [On the Malayan language stem]. The second chapter of this section is a ―Treatment of the singular languages of the stem‖. Part of this treatment is a chapter ―On the grammatical construction of the Malayan languages‖. A chapter on the noun and pronoun is followed by a second chapter on the verb and especially the Tagalog verb (Humboldt 1838: 347– 96). Tagalog is mentioned in other places too, but nowhere as prominently as in this chapter on the verb.6

6

This chapter is commented on by Buchholz (1986: 224–261); she follows the text very closely but has some difficulties in interpreting the Tagalog examples.

The role of the verb in Tagalog

127

The reason why Humboldt dedicates so many pages to the Tagalog verb is provided by him through the following citation:7 Mit dem Tagalischen Verbum ist die vollständigste und am meisten ausgebildete Verbalform des Malayischen Sprachstamms aufgestellt worden. Es war nothwendig, um gleichsam das Höchste zu zeigen, was der grammatische Organismus in diesem Sprachstamm vermag, und es war möglich, da schon systematische Bearbeitungen des Tagalischen Sprachbaues vorliegen. (Humboldt 1838: 396) [With the Tagalog verb, the most comprehensive and the most developed verbal form of the Malayan language family has been presented. This was necessary in order to show so to speak the highest possible level that the grammatical organism can achieve in this language family, and this was possible because there are already systematic descriptions of the Tagalog language construction]. Humboldt gives two types of reasons. For one he is convinced that Tagalog is the key to the analysis of verbal forms in Malayo-Polynesian languages. At the same time, it was the only one which had been treated in a systematic way. This second reason is somewhat striking because in his theoretical writings, Humboldt was convinced that a language could only be analyzed properly if one were using continuous texts (in German: verbundene Rede) and not only grammars and vocabularies. But here, he uses only these systematic descriptions and not the texts available to him (e.g. the texts of the Manual Tagalog). In his presentation, Humboldt follows Totanes‘ description of content and his examples, although he is primarily concerned with passive constructions and species. Mood, tense and number are only treated briefly and superficially. In this context, Humboldt criticizes Totanes as being too dependent on the Spanish grammatical tradition. Humboldt‘s most central insight is that the verb in Tagalog in a certain sense has no verbal character. There are some essential aspects missing, which are necessary for a verb. In the following citation Humboldt gives his definition of a verb: Denn das actuale Sein, welches in der grammatischen Vorstellung das Verbum charakterisiert, läßt sich nicht leicht an sich ausdrücken, sondern verkündigt sich nur dadurch, daß es ein Sein auf eine/ bestimmte Weise, in einer bestimmten Zeit und Person ist, und daß der Ausdruck dieser Beschaffenheit unzertrennlich in das Grundwort verwebt ist, zum sichren Zeichen, daß dasselbe nur mit ihnen und gleichsam in sie versetzt gedacht werden soll. Wo daher das Verbum in Sprachen diese ächte grammatische Form besitzt, da läßt sich sein Begriff gar nicht ohne jene Beschaffenheit anders, als in dem, sich schon dem Substantiv nähernden In7

Citations from Humboldt‘s text are given first in German, followed by an English translation (by I. W.) The translation tries to be as close to the German text as possible. Humboldt uses generally complex constructions which often have intricate anaphoric relations not always easy to translate into English.

128

Iwar Werlen

finitiv, auffassen. Seine Natur ist gerade diese Beweglichkeit, liegt in der Unmöglichkeit, anders, als in einen einzelnen Fall, fixirt zu werden. (Humboldt 1838: 79f.) [The actual being which is characterizing the verb in the grammatical conception may not be expressed by itself alone, but is only announced by the fact that it is a being in a certain way, at a certain time and in a certain person, and that the expression of this condition is inseparably interwoven with the basic word as being a sure sign of that it (i.e. the basic word) can only be thought with them and so to speak be placed within them. Wherever therefore the verb in languages has this authentic form, the concept cannot be thought without those aspects except in the case of the infinitive which is itself near the substantive. Its [i.e. the verb‘s] nature is exactly this flexibility, which is based on the impossibility to be fixed in another way than in one singular case]. For Humboldt, an authentic grammatical verbal form is only given, if mood, tense and person are intimately interwoven with the basic word. This conception is taken from classical Indo-European languages, which express tense, mode, person and number directly through and in the verb and – as he stresses – in an inseparable way (which means, I think, that the expression of TAM and person, number are given in a fusional way). In this definition, we see Humboldt presenting a universal definition, generalizing the verb as it is found in languages such as Ancient Greek or Sanskrit. In a next step, Humboldt says that Tagalog verbal forms are ―denominatives‖ (Humboldt 1838: 317). The basic simple roots of all derivations in Tagalog are conceived of as being nouns. Verbs are derived from such nominal bases. Humboldt continues: und der Verbalbegriff liegt immer in den begleitenden Partikeln und der Wortveränderung, selbst da, wo die Bedeutung, wie in gehen, werfen, schon an sich verbal ist. (Humboldt 1838: 348) [and the verbal concept is always in the accompanying particles and the mutation of the word, even there, where the meaning is verbal by itself, such as in to go, to throw]. It seems that Humboldt does not only argue for a formal definition of a verb, but also for a semantic one, if he says that meanings such as those belonging to ―to go‖ and ―to throw‖ are inherently verbal. But even in case of such verbal meaning he thinks that the Tagalog verb is determined by derivational means and not by semantic ones. This opinion is stressed by him for the whole Malayo-Polynesian family: Es liegt daher jedem Verbum ein vollkommen ausgebildetes, primitives Wort zum Grunde, welches entweder wirklich ein Nomen von irgend einer NominalKategorie ist, oder die unbestimmte Anwendbarkeit zum Nomen und Verbum in sich trägt, aber nur dann, auch nur materieller Weise, schon ursprünglich ein wahres Verbum heißen kann, wann sein Begriff ganz und ausschließlich verbal ist. Es ist daher eine vielleicht zu weit gehende, aber gewiß keine ganz unrichtige Ansicht, im Mal. Sprachstamm das Nomen als die Grundlage des Verbums anzuse-

The role of the verb in Tagalog

129

hen, da jedes Nomen zum Verbum werden kann, und das Grundwort auch des, seinem Begriff nach ursprünglichen Verbums so unbestimmter grammatischer Natur ist. (Humboldt 1838: 82) [The basis of each verb is a perfectly formed, primitive word which is either really a noun of some nominal category or which in an indeterminate way may become a noun or a verb, but only then can this be called an authentic verb, also in material respect, if its concept is completely and exclusively verbal. Therefore, the opinion that the noun is the basis of the verb in the Mal. language family may be too far leading but is certainly not untrue. This is the case because any noun can become a verb, and a basic word with an original verbal conceptual basis has an undetermined grammatical nature]. This citation is meant for the whole language family but it is immediately motivated by the discussion of the verb in Tagalog. Later on in the text, Humboldt seems to change his opinion. After the analysis of the example ―p-in-a-pa-o-óo mo siya wird-befehligtja-zu-sagen von-dir er?‖ ‗is-ordered-to-say-yes of-you he‘ (Humboldt 1838: 384) which contains a derivation of the particle óo ‗yes‘, he says: Ableitung, und mithin die Verwandlung in Verbalausdruck kann aber von jedem Wort in der ganzen Sprache geschehen, und es giebt hiervon, der grammatischen Form nach, durchaus keine Ausnahme. (Humboldt 1838: 348). [Derivation, and consequently the transformation into a verbal expression can be made from every word in the whole language and there is absolutely no exception concerning the grammatical form]. Here, the hypothesis of the nominal nature of stems is weakened – every root word of the language may be the base of a verbal derivation.8 Nowhere in his text does Humboldt give a reason as to why he thinks that the roots are nominal and how he wants to distinguish nouns from other categories. (In parentheses only: the background of the whole discussion of the nominal or verbal nature of the roots seems to be motivated by Herder‘s idea presented in Treatise on the origin of language that the first words of the language were verbs). Examining the derivation of verbs more closely, Humboldt thinks that in Tagalog something is going wrong. He says: Anstatt die Verbindung der Person mit dem Verbum, ohne welche die wahre Symbolisierung seines Begriffs unmöglich ist, gehörig mit ihm zu verschmelzen, und wieder mit dieser auch die Zeitbedingungen in Verbindung zu setzen, bildet sie die Seiten des Verbums aus, die ihm, auch wenn es unpersönlich und gleich8

I would like to underline that the hypothesis is Humboldt‘s. There is an ongoing debate concerning the nature of Tagalog stems. From a comparative linguistic point of view, the stems seem to be nouns (see e.g. Starosta et al. 1982 and Ross 1995). From a synchronic point of view, we prefer Himmelmann‘s term of concept expressions (Begriffsausdrücke) used in his dissertation (Himmelmann 1987: 37) which are neither verbs nor nouns.

130

Iwar Werlen

sam in der Flüchtigkeit seiner Natur fest geheftet erscheint, zukommen können. Hiervon ist der Numerus ein auffallendes Beispiel. Statt diesen zugleich mit der Person in das Verbum zu verflechten, versetzt sie denselben zugleich oft in den Grundbegriff selbst, und vermischt also wenigsten die Begriffe, in dem sie ein Präfix in gewissen Fällen dem Verbum nur dann zugiebt, wenn dasselbe die Mehrheit der Handlungen anzeigt. In der verbundnen Rede kann nun daraus freilich keine Undeutlichkeit entstehen, da das Pronomen in der Mehrzahl verschiedene Formen hat; aber in das Verbum wird der Begriff dadurch um nichts mehr verflochten. (Humboldt 1838: 317) [Instead of fusing the person with the verb as it should be and without which the true symbolization of its concept is impossible, and instead of fusing this with the temporal conditions, [the language] is forming those aspects of the verb which may be attributed to it even if it is impersonal and so to say fixed in the volatility of its nature. A remarkable example of this is number. Instead of tying it to the person and the verb, [the language] is mixing it into the basic concept itself, and is mixing at least the concepts, when it fuses in certain cases the prefix with the verb only if it [i.e. the prefix] indicates the pluralitiy of activities. In continuous speech there is no ambiguity because the plural of the pronouns is clearly marked, but the concept is not fused with the verb at all]. As is well known, number in the sense of the Indo-European languages is not present in Tagalog but there is a series of constructions which can express plurality (see Kolmer 1998). In verbs, for example, the prefix magsi- may indicate that there are several actors implicated and it seems that Humboldt meant such prefixes in the text cited. In the context of such verbs, pronouns may be used in the plural (pronouns are the only words in Tagalog which have distinct plural forms), but in the verbal form itself neither person nor number are encoded. Since for Humboldt, person and number are essential in encoding the verb, while other things such as ability to do something or intensifying and other semantic modifications are only accidental, he thinks that Tagalog is encoding only accidental aspects of the verb and not the essential ones. In a next step, Humboldt follows Totanes in saying that verbal forms may be used as nominal ones by putting an article before it (he thinks, like Totanes, that ang, ng and sa are articles, I myself regard them as case markers, but see Reid (2002) for another interpretation of ang). As we already saw, Totanes calls such forms participles, but Humboldt correctly states that they are formally identical with verbal forms. His conclusion: Die angeblichen Verbalformen der Sprache sind also an und für sich gegen die Nominal- und Verbalnatur gleichgültig, und es hängt erst von der Behandlung in der Construction ab, ob man sie als Participia oder als Formen des flectirten Verbums betrachtet. Setzt man ihnen ein Pron. nach, so sind sie Verba, setzt man ihnen einen Artikel vor, so sind sie Participia oder richtiger Verbal-Adjectiva. (Humboldt 1838: 349) [The alleged verbal forms of the language are therefore

The role of the verb in Tagalog

131

actually indifferent towards the nominal or verbal nature and it depends on the construction whether they are seen as participles or as forms of the inflected verb. If they are followed by a pronoun, they are verbs, if an article is put before them, they are participles or more adequately said verbal-adjectives]. Humboldt‘s argumentation is straightforward. Totanes gives the correct facts, but he is not able to detach himself from the Greco-Latin model. Humboldt, on the other hand, sees that the same form can function as a predicate but also as an argument. What Humboldt does not mention in his texts is the fact that the same ambiguity is also given for normal nominal elements: nominal elements may also be used as predicates in nominal sentences such as: (4)

duktor ang lalaki doctor TOP man ‗The man is a doctor‘

(5)

lalaki ang duktor man TOP doctor ‗The doctor is a man‘

After having shown that verbal forms are in some way indifferent for a verbal or a nominal interpretation he summarizes: Die Bezeichnung der Verbalformen trifft also die transitive oder intransitive, active und passive Natur, die Modi, Tempora, und andre, theils die Verbalbeschaffenheit überhaupt (z. B. Frequentativa, Inchoativa), theils die einzelne materielle Bedeutung angehende Modificationen. (Humboldt 1838: 349) [The symbolization of verbal forms includes therefore the transitive or intransitive, active or passive nature, moods, tenses and others, partly the general character of the verb (e.g. frequentatives, inchoatives), partly the modification of the singular material meaning]. With this statement he summarizes his findings in Totanes. However, later on, when discussing mood in Tagalog, he no longer accepts mood as a verbal category in Tagalog. So he says, some fifty pages later: Der Begriff des Modus gehört sichtbar der Sprache selbst gar nicht an. Er wird am Verbum, worunter ich immer das Wort in der Con-// jugationsform versteh, nie bezeichnet, sondern ist, wo er keine Tempusbezeichnung mit sich führt, daran, wo er sie hat, an begleitenden Conjunctionen kenntlich. (Humboldt 1838: 390f.) [The concept of mood observably does not belong to the language at all. It is never encoded in the verb (by verb I always mean the word in its inflected form), but it is, if not present in the tense form, only indicated in accompanying conjunctions].

132

Iwar Werlen

This citation is perhaps not easily understandable. Humboldt means to say that there is no formal expression of mood in the verb. Mood is recognizable in verbal forms without tense (e.g. the infinitive) exactly through the absence of tense (the infinitive may be used as imperative). In tensed verbal forms, mood is indicated by accompanying conjunctions (as we already saw in the chapter on Totanes). Totanes‘ subjunctive and optative are therefore seen as mere constructs of a grammarian on the base of Greco-Latin grammar and not as linguistic facts in Tagalog. After this deconstruction of the category of mood, Humboldt discusses extensively what he calls the two main differences in verbal expressions in Tagalog. He says: Der eine [Verbalausdruck] betrachtet die Handlung als eine Thätigkeit des Subjects, und als sich auf ein Object beziehend, und stellt daher das Verbum in // natürlicher Folge mit seinem Subject und Object zusammen; ang bálang ai s-ung-m-ira nang palai, die Heuschrecke (Bindepart.) hat-zerstört den Reiß. Der zweite Verbalausdruck betrachtet die Handlung bloß als etwas Geschehenes, und kleidet also das Verbum in eine Nominalform mit ausgelassenem Verbum sein ; sisirà-in co ang bibìg mo, wörtl. zerstören-werden mein der Mund dein, d.h. dein Mund wird von mir zerstört werden, oder wird mein Zerstören sein. Dieser Ausdruck stellt auch die Handlung in den verschiedenen Beziehungen dar, in denen sie bald auf den Handelnden, bald auf das Gewirkte, bald auf Nebenumstände (Werkzeug, Ort) genommen werden kann. Er hat daher mehrfache Formen, der erste nur Eine. Denn in diesem wird die mit Thätigkeit vollbrachte Handlung als die Hauptsache angesehen, welche so nur Eine Richtung erlaubt. Bei jenem ist noch ungewiß, auf welche der verschiedenen vorkommenden Beziehungen der Redende seine Aufmerksamkeit richten will. (Humboldt 1838: 349f.) [The one [verbal expression] considers action as an activity of the subject, and as one related to an object, and it arranges consequently the verb in natural order with its subject and object ang bálang ai su-ng-m-ira nang palai, the grasshopper (linker) has-destroyed the rice. The second verbal expression considers action only as something that happened and expresses therefore the verb in a nominal form with an absent verb to be; sisirà-in co ang bibig mo, verbally: destroy-become mine the mouth of-you, i.e. your mouth will be destroyed by me, or will be my destroying. This expression also presents the action in various respects; it can be seen either with respect to the actor, or to the result, or to the accessories (instrument, location). It has therefore many forms, the first one is only one of them. Thus in this case the action done in the activity is seen as the principal thing which allows only one direction. In the second form, however, it is yet uncertain on which of the various possible respects the speaker will focus]. Humboldt takes up Totanes‘ distinction between an active and three passives. He considers the active form as the one presenting an activity of the subject related to an object (which means, that he considers especially semantically transitive verbs). We do not

133

The role of the verb in Tagalog

clearly understand why he speaks of a ―natural order‖. The example he cites can be used without any problem with the predicate as the first word of the sentence: (6)

sumira ng AF-COMPL-destroy NT ‗the grasshopper destroyed rice‘

palay rice

ang balang TOP grasshopper

The so called ‗inverted order‘ with the linker ay is optional and can be chosen independently from the verbal form. It seems, therefore, that Humboldt does not mean the order subject – verb – object, but the fact that the subject is marked by ang, and the object by ng. We may thus hypothesize that Humboldt thinks in terms of the IndoEuropean pattern of the nominative (encoded by ang) and accusative (encoded by ng) and that he considered the active encoding as the ―natural‖ one.9 The so-called passive verbal form is analyzed by Humboldt in a substantially different way than by Totanes. He assumes that the action is no longer seen as an activity but as something that has already happened. He therefore thinks that the verbal form with in is nominal. He analyses the construction as a nominal sentence without a copula. His formulation ―with an absent verb to be‖ is ambiguous, but Humboldt knew that there is no copula in Tagalog. Therefore it seems that he really thought that the so-called passive construction was a nominal sentence. This is also shown in a second example he analyses at another place: bigy-an mo aco n-ang tubig, … gegeben-werde von-dir ich des Wassers, aber eig. und der wahren Construction nach: Gebungsort dein (sei) ich des Wassers, oder ich sei der Ort, wo du das Wasser hingiebst (Humboldt 1838: 351) [bigy-an mo aco n-ang tubig, ... given-become of-you I of-the-water, but essentially and according to the real construction: giving-location of-you (be) I of-the-water, or I be the place where you are giving the water]. Humboldt shows, that -in and -an may also be used in nominal derivations (e.g. kanin ‗cooked rice‘ and bakuran ‗backyard‘, to name only two). And he often repeats: Die mit den drei Partikeln verbundenen Grundwörter sind an und für sich, und vermöge dieser Verbindung, gar keine Passiva, sondern Nomina (Humboldt 1838: 352) [The basic words combined by the three particles are strictly seen and are by virtue of this combination not passives, but nouns]. But there is at least one problem with this analysis: Humboldt seems to assume that mo in his example is a genitive put after the nominal verbal form. mo may be used in this way: 9

In a comment to our hypothesis, L. A. Reid argued that Humboldt considered SVO as the ―natural order‖ and would have treated VSO as an inverted order, had he known about it. Humboldt knew the VSO for ‗active‘ constructions as other examples in his text show. We prefer therefore to maintain our hypothesis by arguing that for Humboldt, the prototypical S (subject) was a nominative used in an ‗active‘ construction, and the prototypical O (object) an accusative in the same frame.

134 (7)

Iwar Werlen

ang

bahay mo house GEN-2.SG ‗your house‘ TOP

But in Humboldt‘s example, mo is placed at the structural second place, the place where most clitic particles of Tagalog are expressed. If Humboldt had used a non-clitic element, the place of this element would be no longer directly after bigyan: (8)

Binigyan ako ng tubig DF-COMPL-give TOP-1.SG NT water ‗I have been given water by Fernando‘

ni Fernando NT Fernando

In this construction, ni Fernando cannot be interpreted as a genitive to binigyan. Therefore, at least one implicit argument for the nominal nature of the passive constructions seems to be not valid.10 Consequently, Humboldt interprets the three passives in the sense of a pragmatic theory of focus when he says that ―der Redende seine Aufmerksamkeit‖ [auf etwas richten wolle]: ‗the speaker is focusing on‘ something. We assume that here he accepts Totanes‘ expression of intending something. Schachter & Otanes use the terms focus and topic and we use them too, but they can easily be understood in a wrong way. A short history of the interpretation of this system can be seen in Matsuda French (1987/88) and in Quakenbush (2005: esp. 9); we will not go on discussing it at this stage. It is well known that Tagalog speakers prefer the non-actor focused forms, however, why and how is often unclear. But the fact that in constructions with object focus, the object can be absent does not support a pragmatic theory of ―focusing‖. See the following example: (9)

basahin mo OF-IMP-read NT-2.SG ‗read (it)!‘

Here we find no explicit mention of the focused element because it is contextually determined and because normally pronouns are only used for human beings (for a discussion of such constructions see Himmelmann 1999: 244ff.). One could argue that the non-linguistic context is focused on the object to be read, but we think that the focus of the verbal form is directed on things which are accepted as given in the situation and which therefore pragmatically can be left out. This assumption could also explain the 10

In a comment by one of the reviewers it is argued that Tagalog word order is not as free as is implicated in the example: if ako is replaced by a full NP too, the resulting word order should be binigyan ni Fernando ng tubig si Ronaldo which results in a genitive reading of ni Fernando. This leads to the question of the freedom of word order in Tagalog sentences. The reviewer cites Himmelmann (1991: 49f.) who showed that Tagalog sentences have strong preferences for the sequence of the predicate followed by the ng-phrase (Genitive) and the full ang-phrase, which is seen by Himmelmann as a confirmation of what he calls the equational hypothesis.

The role of the verb in Tagalog

135

fact that focused elements are often translated with a definite article. It is the essential function of a definite article to characterize an element as given in the situation. We assume therefore that the focus-topic-system has something to do with given and new information. But it is also clear that affixes affect the semantics of verbs. This is even more the case with other affixes, called ―species‖ by Totanes. Humboldt‘s analysis of the active and passive verbal forms leads him to a critical remark: Die Tag. Grammatiker nennen den ersten dieser beiden Verbalausdrücke den activen, den zweiten den passiven. Jenes ist vollkommen richtig, dieses aber nur insofern, als wirklich die Construction immer eine passive, wenigstens intransitive ist. Allein übrigens umfaßt der zweite Tag. Verbalausdruck auf der einen Seite mehr, als unser einfaches Passivum, und auf der andren wird das Passivum auch durch eigne, aus der Verbalform activer Construction abgeleitete Partikeln bezeichnet, so daß die Benennung zugleich zu eng und zu weit ist. (Humboldt 1838: 350) [Tagalog grammarians call the first of the two verbal expressions the active one, the second the passive one. The first one is completely correct, the second one however only with respect to the construction being a passive one, or at least an intransitive one. But the second Tagalog verbal expression contains on the one hand more than our simple passive, and it comprises on the other hand more through particular particles derived from the verbal form in the active construction, so that the denomination is at the same time too narrow and too wide]. What Humboldt calls ―the simple passive‖ is the passive of transitive verbs in the IndoEuropean tradition. He sees the passive as an instance of intransitivity. It is clear that Tagalog knows other intransitive constructions such as weather verbs which are built by -um-: umulan ‗to rain‘ or b-um-agyo ‗to typhoon‘ etc. His own interpretation of these passives is formulated in the following citation: In Wahrheit ist die Sache offenbar die, daß die Sprache da, wo der Gedanke ein Verbum erfordert, an dessen Stelle eine Nominalform anwenden will, und daß ihr diese Richtung des Ausdrucks so geläufig geworden ist, daß sie dieselbe dem wahren Verbalausdruck vorzieht. Sie geht daher bei diesem angeblichen Passivum gar nicht hauptsächlich von der Darstellung des Thuns und des Leidens aus. Die Form des Ausdrucks wird nicht durch die Natur des Auszudrückenden nothwendig gemacht, sondern liegt weit mehr, wenn nicht ganz, in der Eigenthümlichkeit der Sprache. Sie benutzt dieselbe aber, um dadurch zugleich einen anderen Zweck und, wie man nicht läugnen kann, einen eignen Vorzug zu erreichen. Es soll nämlich durch diese zwei verschiedenen Darstellungen des Verbalbegriffs deutlich gemacht werden, worauf der Redende die Wichtigkeit seiner Aussage legt. Thut er dies in die Handlung, so braucht er die active Construction. Liegt // ihm aber der Nachdruck in dem Subject oder Object, oder

136

Iwar Werlen

den Nebenumständen, so wendet er die passive an, und wählt die seiner Absicht entsprechende Form. (Humboldt 1838: 350f.) [In truth the situation is manifestly so that language instead of expressing thought through a verb wants to use a nominal form and that this way of expression is so common that it [i.e. the language] prefers these nominal forms to the true verbal expressions. In the case of this alleged passive, language does not try to represent activity or passivity. The form of expression is not made necessary by the nature of what should be expressed, but lies for the most part or perhaps totally in the peculiarity of the language. But it [i.e. the language] uses this peculiarity at the same time to achieve another advantage and this cannot be denied. Through these two presentations of the verbal concept it should be made clear which aspect of the statement is focused on by the speaker. If he is focusing on the action, he uses the active construction. If he is focusing on the subject or the object or the circumstances, he uses the passive one and chooses the form appropriate to his intention]. In this text, Humboldt is in some way ambivalent. The explanation that language chooses a nominal form instead of a verbal form ―which is required by thought‖ shows that his semantic concept of the verb is incompatible with the technique of Tagalog. So he assumes that this language prefers something which should not be the case. In the second part of the text, he goes on to interpret this technique positively: he repeats his theory of focusing and he evaluates this as an advantage of the language. At this point, Humboldt seems to turn his universalistic view of the verb sketched out at the beginning of the chapter on the Tagalog verb into a relativistic one by interpreting the technique of Tagalog as something particularly valuable. We know now the theoretical basis for Humboldt‘s reading of the Tagalog verb. Summing up what he says: verbs according to his interpretation are active verbal forms which are derived with a simple actor-focusing affix from a root word. The so-called passive forms on the other hand, are seen by him as nominal derivations. The verbal character of these active forms is only recognizable in a syntactic context, i.e. connected to an article, such a verbal form is interpreted as a nominal one. In the remaining part of the text, Humboldt tries to summarize Totanes‘ species. He critizices Totanes for not being really systematic (an impression one cannot avoid if one reads Totanes‘ chapters on this topic) and he ends with the following statement: Die Sache ist offenbar die: die Sprache verbindet jedes als Verbum gebrauchte Wort mit einem oder einer bestimmten Anzahl wechselnder Präfixe (die infigierende Conjug. hier mit eingerechnet), sie giebt vermittelst derselben dem Verbalbegriff gelegentlich alle im vorigen Paragraphen erwähnte Modificationen, und es liegt in ihr hierbei die doppelte Eigenthümlichkeit, daß sie für sehr verschiedene Kategorien (untrennbare Präpos., transitive und Causalverba) eine und die nämliche Bezeichnungsart durch die Conjugationsformen hat, und daß sie Dinge durch die Verbalform ausdrückt, die andere Sprachen besser be-

The role of the verb in Tagalog

137

sonderer Erläuterung durch adverbialische Zusätze überlassen … (Humboldt 1838: 367) [The facts are evidently the following: language combines every word used as a verb with one or a certain number of changing prefixes (the infixes included). Through these it [the language] modifies the verbal concept in all the respects mentioned in the preceding paragraph, and this leads to the double particularity that it [the language] expresses very different categories (inseparable preposition, transitive and causal verbs) in one and the same way in all inflected forms, and that it [the language] expresses things in verbal forms which are expressed better in other languages by adverbial constructions]. In this passage we can see a sort of backsliding into an Indo-European view of the Tagalog verb. Even if Humboldt accepts that the way Tagalog is using its verbs can be seen positively, he evaluates the use of the prefixes for categories he interprets as very different, negatively: other languages do express these categories adverbially and this way of doing it is ―better‖ for him than the way Tagalog does it. It is a little surprising that he does not accept the Tagalog way as another and more grammaticalized way of expressing content which Indo-European languages cannot do in the same grammatical way. If one looks at Humboldt‘s many theoretical writings one can see that he starts with the idea that fusional languages such as Sanskrit and Ancient Greek are the best types of languages. Later on he accepts the idea that other types of languages such as Chinese are also capable of expressing complex and culturally valuable thoughts. His shift from a universalistic view to a relativistic one is mirrored in his interpretation of the Tagalog verb, but in the end, the basic idea reappears that Indo-European languages do their job better than Tagalog.

5. Conclusion Looking at Humboldt‘s interpretation of the verb in Tagalog given the background of the knowledge available at his time, we can state that he is practically the only one who was able to refrain from the influence of Greco-Latin grammar. He does not however do it in every case and not always consistently. One can find contradictions in his texts. In the beginning of a chapter, he follows Totanes more or less without any criticism; but in the end he interprets the verb in a totally different way, no longer accepting mood as a category, destroying the idea of three passives and positing the theory of focussing. These aspects are new, but his starting point is a universalistic semantic definition of the verb determined by Indo-European languages. It seems that Humboldt used his IndoEuropean conception of language as a starting point which allowed him to move beyond this conception and to put another, more relativistic one in its place.

138

Iwar Werlen

Abbreviations AF COMPL DF GEN IMP INCOMPL

actor-focus completed (aspect) direction-focus genitive imperative incompleted (aspect)

LK LOC NT OF SG TOP

linker locative non-topic object-focus singular topic

Bibliography Adelung, Friedrich von (1976 [1815]): Catherinens der Grossen Verdienste um die Vergleichende Sprachenkunde. Nachdruck der Ausgabe von 1815 mit einer Einleitung und einem biobibliographischen Register von Harald Haarmann. Hamburg: Buske. Adelung, Johann Christoph (1806): Mithridates oder allgemeine Sprachkunde mit dem Vater Unser als Sprachprobe in bey nahe fünfhundert Sprachen und Mundarten. Erster Theil. Berlin: Vossische Buchhandlung. Adelung, Johann Christoph (1817): Mithridates oder allgemeine Sprachenkunde mit dem Vater Unser als Sprachprobe in bey nahe fünf hundert Sprachen und Mundarten. Mit wichtigen Beiträgen zweyer großer Sprachforscher fortgesetzt von Dr. Johann Severin Vater. 4. Theil. Berlin, Voss 1817 [Nachdruck Hildesheim, New York, Olms 1970]. Alter, Franz Carl (1803): Ueber die taglische Sprache von F‘ C‘ A‘. Doctor der Philosophie. Custos der K. K. Universitaets Bibliothek. Wien: Gedruckt bei Johann Thomas Edl. v. Trattnern, k. k. Hofbuchdruckern und Buchhändlern. Arens, Hans (21969): Sprachwissenschaft. Der Gang ihrer Entwicklung von der Antike bis zur Gegenwart. Zweite, durchges. u. stark erw. Auflage. Freiburg, München: Alber (Orbis Acade–micus, Bd. I/6). Buchholz, Ulrike (1986): Das Kawi-Werk Wilhelm von Humboldts. Untersuchungen zur empirischen Sprachbeschreibung und vergleichenden Grammatikographie. Münster: Institut für Allgemeine Sprachwissenschaft der Westfälischen Wilhelms-Universität (Studium Sprachwissenschaft, Beiheft 4). Doctrina, Christiana (1947): The first book printed in the Philippines, Manila, 1593. A Facsimile of the Copy in the Lessing J. Rosenwald Collection. Library of Congress. Washington. With an Introductory Essay by Edwin Wolf II. Hervás, Don Lorenzo (1784): Catalogo delle lingue conosciute e notizia della loro affinità, e diversità. Cesena: Biasini. Hervás, Don Lorenzo (1785): Origine formazione, meccanismo, ed armonia degl‘idiomi. Cesena: Biasini. Hervás, Don Lorenzo (1786): Aritmetica delle nazioni e divisione del tempo fra l‘orientali. Cesena: Biasini. Hervás, Lorenzo (1787): Saggio pratico delle lingue con prolegomeni, e una raccolta di orazioni Dominicali in più di trecento lingue, e dialetti, con cui si dimostra l‘infusione del primo idioma dell‘uman genere, e la confusione delle lingue in esso poi succeduta, e si additano la diramazione, e dispersione delle nazioni con molti risultati utili alla storia. Opera dell‘abbate Don Lorenzo Hervàs Socio della Reale Accademia delle Scienze, ed Antichità di Dubliono, e dell‘Etrusca di Cortona. Cesena: Biasini. Hervás, Don Lorenzo (1989[1800–18005]): Catálogo de las lenguas de las naciones conocidas y numeración, división y clases de éstas según la diversidad de sus idiomas y dialectos. Vol. I

The role of the verb in Tagalog

139

(1800), Vol. II (1801), Vol. III (1802), Vol. IV (1804), Vol. V (1804), Vol. VI (1805). Madrid: Ranz. Nachdruck Madrid. Himmelmann Jr., Nikolaus (1987): Morphosyntax und Morphologie – Die Ausrichtungs-affixe im Tagalog. München: Fink. Himmelmann, Nikolaus P. (1991): The Philippine challenge to Universal Grammar. (Arbeitspapier Nr. 15; Neue Folge). Köln: Institut für Sprachwissenschaft, Universität zu Köln. Himmelmann, Nikolaus P. (1999): The lack of zero anaphora and incipient person marking in Tagalog, in: Oceanic Linguistics 38(2), 231–269. Himmelmann, Nikolaus P. (2007): Lexical categories and voice in Tagalog, in: Musgrave, Simon & Austin, Peter (eds.), Voice and grammatical relations in Austronesian languages. Stanford: CSLI, 247–293. Humboldt, Wilhelm von (1836): Über die Kawi-Sprache auf der Insel Java, nebst einer Einleitung über die Verschiedenheit des menschlichen Sprachbaues und ihren Einfluß auf die geistige Entwickelung des Menschengeschlechtes. Erster Band. Berlin, Königliche Akademie der Wissenschaften (Abhandlungen der Könglichen Akademie der Wissenschaften zu Berlin. Aus dem Jahre 1832. Zweiter Theil. Berlin: Königliche Akademie der Wissenschaften. In Commission bei F. Dümmler. Humboldt, Wilhelm von (1838): Über die Kawi-Sprache auf der Insel Java, nebst einer Einleitung über die Verschiedenheit des menschlichen Sprachbaues und ihren Einfluß auf die geistige Entwickelung des Menschengeschlechtes. Zweiter Band. (Fortsetzung der Kawi-Sprache; Malayischer Sprachstamm im Allgemeinen und dessen westlicher Zweig.). Berlin: Königliche Akademie der Wissenschaften. In Commission bei F. Dümmler. Humboldt, Wilhelm von (1839): Über die Kawi-Sprache auf der Insel Java, nebst einer Einleitung über die Verschiedenheit des menschlichen Sprachbaues und ihren Einfluß auf die geistige Entwickelung des Menschengeschlechtes. Dritter Band. (Südsee-Sprachen, als östlicher Zwei ges Malayischen Sprachstammes.) Berlin: Königliche Akademie der Wissenschaften. In Commission bei F. Dümmler. Humboldt, Wilhelm von (1988): On Language: The diversity of human language structure and its influence on the mental development of mankind. Transl. by P. Heath; with an Introduction by H. Aarsleff. Cambridge: Cambridge University Press. Humboldt, Wilhelm von (1998 [1836]): Über die Verschiedenheit des menschlichen Sprachbaus und ihren Einfluß auf die geistige Entwicklung des Menschengeschlechtes. Hg. v. Donatella de Cesare. Paderborn et al.: Schöningh (UTB 2019). Kolmer, Agnes (1998): Pluralität im Tagalog. (Arbeitspapier Nr. 31, Neue Folge). Köln: Institut für Sprachwissenschaft. Matsuda French, Kathleen (1987–88): The focus system in Philippine languages: An historical overview, in: Philippine Journal of Linguistics 18(2); 19(2), 1–27. Mueller-Vollmer, Kurt (1993): Wilhelm von Humboldt‘s Sprachwissenschaft. Ein kommentiertes Verzeichnis des sprachwissenschaftlichen Nachlasses. Mit einer Einleitung und zwei Anhängen. Paderborn et al.: Schöningh. Pallas, Peter Simon (1977 [1787]): Linguarum totius orbis vocabularia comparativa; augustissimae cura collecta. Petropoli: Typis Iohannis Caroli Schnoor. Nachdruck. Hrsg. u. mit e. Vorw. verseh. von Harald Haarmann. Hamburg: Buske (2 Bde). Phelan, John Leddy (1955): Philippine linguistics and Spanish missionaries, 1565–1700, in: MidAmerica. An Historical Quarterly 37 (New Series, Vol 26), 153–170. Quakenbush, J. Stephen (2005): Philippine linguistics in an SIL perspective: Trends and prospects, in: Liao, Hsiu-chuan & Rubino, Carl R. Galvez (eds.), Current issues in Philippine linguistics and anthropology. Parangal kay Lawrence A. Reid. Manila, Philippines, The Linguistic Society of the Philippines and SIL Philippines, 3–27.

140

Iwar Werlen

Regalado Trota, José (1998): Obras Clásicas de Lenguas Indígenas de Filipinas. Madrid: Fundación Histórica Tavera; Madrid: Digibis (Colección Clásicos Tavera. Serie IX. Fuentes lingüísticas indígenas. Vol. 9). [CD containing grammars and dictionaries]. Reid, Lawrence A. (2002): Determiners, nouns, or what? Problems in the analysis of some commonly occurring forms in Philippine languages, in: Oceanic Linguistics 41(2), 295–309. Reid, Lawrence A. (2006): On reconstructing the morphosyntax of Proto-Northern Luzon, in: Philippine Journal of Linguistics 37(1),1–64. Reid, Lawrence A. & Liao, Hsiu-chuan (2004): A brief syntactic typology of Philippine languages, in: Language and Linguistics 5(2), 433–490. Ross, Malcolm D. (1995): Reconstructing Proto-Austronesian verbal morphology: Evidence from Taiwan, in: Jen-kuei Li, Paul; Tsang, Cheng-hwa; Huang, Ying-kuei; Ho, Dah-an & Tseng, Chiuyu (eds.), Austronesian Studies relating to Taiwan. Symposium Series of the Institute of History and Philology, Academia Sinica, Number 3. Taipei: Academia Sinica, 727–791. Schachter, Paul & Otanes, Fe T. (1972): Tagalog reference grammar. Berkeley, Los Angeles, London: University of California Press. Schwarz, Christa (ed.) (1993): Ex Libris a Gulielmo L. B. de Humboldt Legatis. Das Legat Wilhelm von Humboldt‘s an die Königliche Bibliothek in Berlin. Verzeichnet und kommentiert von Chr. Schwarz mit einem Geleitwort von Kurt Mueller-Vollmer. Paderborn et al.: Schöningh. Starosta, Stanley; Pawley, Andrew & Reid, Lawrence (1982): The evolution of focus in Austronesian, in: Halim, Amran; Carrington, Lois & Wurm S. A. (eds.), Papers from the Third International Conference on Austronesian Linguistics, Vol. 2: Tracking the travellers. Pacific Linguistics C-75. Canberra: Australian National University, 145–170. Totanes, Sebastian de (1745): Arte de la lengva tagala, y manval tagalog ... Sampaloc: Imprenta de la Provincia. [The first and the second edition of 1850 were consulted at the library of the Humboldt University, Berlin; the third edition is to be found on the CD ed. by José 1998]. Ward, Jack H. (1971): A bibliography of Philippine linguistics and minor languages. With annotations and indices based on works in the Library of Cornell University. Ithaca, N. Y.: Cornell University (Data Paper Nr. 83, Southeast Asia Program, Department of Asian Studies): Wolf, Edwin (1947): Doctrina Christiana, in: Doctrina Christiana. The first book printed in the Philippines, Manila, 1593, 1–50.

LAWRENCE A. REID (HONOLULU/USA)

Seidenadel‘s grammar of Bontoc Igorot: one hundred years on

Abstract In 1909, Carl Wilhelm Seidenadel published a grammar, vocabulary and texts of a Philippine language that he called Bontoc Igorot. Although there are serious problems with his description of the phonology, his treatment of the syntax of the language was in some ways quite perspicacious. He was probably the first linguist to reject the traditional (at that time) treatment of the various transitive constructions of Philippine languages as passives, and offered an analysis that considered them to be active constructions. This chapter will review the work in the light of what is now known about the language, in particular his descriptions of the phonology and syntax, and relate his views to current issues in the typological characterization of the languages of the Philippines.

1. Introduction In the early 1900‘s, several groups of men and women (more than ninety all told), were taken from their homes in various villages of Bontoc, presently the municipal capital of Mountain Province, to be displayed at the St. Louis World‘s Exposition in Missouri during the summer and autumn of 1906. In the autumn of 1906 the first group was joined by another contingent of about thirty men and women who had been scattered at different sites in the United States. During this period, and also in 1907, Carl Wilhelm Seidenadel (henceforth SDL), at one time, according to Conant (1911) a member of the faculty of the University of Chicago in the Departments of Greek and Latin, met regularly with various members of the group in order to learn their language and describe it. Many of his language assistants had only a rudimentary knowledge of English, and SDL was forced therefore to gather his data monolingually. It was published in 1909, and is available on line, as The First Grammar of the Language Spoken by the Bontoc Igorot with Vocabulary and Texts. 592 pp. (Chicago: Open Court Publishing Co.) Given the general view at that time of these loin clad, spear-carrying ‗natives‘ who were put on display, somewhat like animals in a zoo, SDL‘s description of them is refreshing. He speaks of them as ―these most sympathetic people, men of astonishing intelligence, inborn independence and frankness, strong principles of honesty, kind disposition, a vivid desire for learning, and blessed with the divine gift of healthy hu-

142

Lawrence A. Reid

mor‖. He was less kindly disposed to those who had previously published studies of the language, as noted in Chamberlain‘s (1910) review of the work, …he dismisses the Bontoc words in Sawyer‘s The Inhabitants of the Philippines (London, 1900) with the remark (p. 277), ‗Sawyer‘s list is harmlessly incorrect‘; of Schadenberg‘s vocabulary (Z. f. Ethno., 1889) he says that it ‗is teeming with blunt errors‘; and the section on language in Jenks‘s The Bontoc Igórot (Manila, 1905) gets praise for the vocabulary, but the grammatical notes are considered ‗superficial‘ (p. xii). Given the extensive errors in his own transcriptions and translations, he could have been a little more generous. SDL‘s language assistants came from a number of dialectally distinct areas. He lists Gonogon, Alab, Samoki, and Tocucan, various towns along the main road going into and out of Bontoc town, as well as Bontoc town itself, each community speaking a distinct dialect. He also lists Sabangan, Basao, and Sagada, towns which were, at that time, part of Bontoc subprovince, but which today are subsumed as municipalities in Mountain Province. Their language is closely related to Bontok but is typically referred to as Kankanay or Kankanaey. He also mentions a town called Tagkong which I cannot identify. The home towns of some of his assistants are identified (e.g., Anauwasal from Tocucan, Agpauwan from Alab, and Matyu from Bontoc), however the spelling of the names of some of his assistants (Abakid, Bugti, Bumegda, Domingo, and Langagan) suggests that they came from one or another of the Kankanaey towns, since most of the towns of Bontoc [fʊntʊk] municipality do not allow pre-vocalic voiced obstruents (see sec. 2.2 for discussion). Considering the wide variety of dialects that SDL drew data from, and the fact that he apparently never visited the Philippines, his description is in some respects groundbreaking. This paper will review the work in the light of what is now known about the language, in particular his descriptions of the phonology and syntax, and relate his views to current issues in the typological characterization of the languages of the Philippines. SDL was well aware of the considerable differences which existed between Bontok and the Indo-European languages with which he was familiar, but decided to retain for the convenience of students the customary order of the chapters in [grammars of Indo-Germanic languages], if he would treat first the article, then the noun, pronoun, adjective, etc., just as if the Bontoc language would distinguish the same grammatical categories as the Indo-Germanic Languages (xiii), but was careful to point out that the sooner [the student] can free himself completely from clinging to his former notions of the structure of a language and adapt himself to new categories of lin-

Seidenadel‘s grammar of Bontoc Igorot

143

guistic elements, the earlier he will succeed in entering into the spirit of this admirable idiom (ibid.). Throughout the grammar, SDL is concerned with describing how Bontok expresses concepts familiar to English speakers, such as ―subject‖ and ―object‖, but notes that he uses these terms ―in our, but not the Igorot‘s conception!‖ (27).

2. Phonology SDL was supremely confident of his ability to recognize and represent the sounds that he heard, and of his own erudition, frequently citing translation equivalents from Greek, Latin, Russian, French, Spanish, Scotch and German. In his words, As all examples are recorded exactly as they were obtained from the Igorot, and as the men pronounced the same word in the same construction often with changed sounds and accents, it happens that some inconsistency prevails in orthography, accents and quantity. This is due to the natives‘ elocution, but not to the Author (xiv). Unfortunately, every page of data is replete with misrepresentations, despite his claim that Each word and phrase has been repeatedly verified by various single individuals, by small and larger groups of men and women, young and old, at different times and occasions, often employed unexpectedly in conversation, and special care was taken not to tire a man, as there is danger lest tired men answer so as to please the inquirer (xiv). The generosity of his Bontok assistants, carefully noted by SDL, cannot be doubted. The following two sections will examine his treatment of the vowels and consonants, respectively.

2.1. Vowels There are four phonemic vowels in Bontok, /i, u, ɨ, a/ (see Table 1) (Reid 1963, 2005). SDL established an alphabet of 13 vowels and 10 diphthongs for Bontok, based on the non-phonemic traditions of the time. Stress is contrastive, and is marked primarily by length on the vowels of open, penultimate syllables. A short open, penultimate syllable usually signals stress on the ultimate syllable. SDL correctly notes that some words differed only in vowel length, but incorrectly claims that vowel length and word accent can be located on different vowels in the same word (xiv). While also noting that a long vowel is sometimes only slightly longer than its short counterpart, he nevertheless in-

144

Lawrence A. Reid

correctly represents it in a large number of forms, e.g., fāto for [fǎˈtu] ‗stone‘ (5), kṓko for [kŭˈku] ‗(finger or toe) nail‘ (390), and al wid for [ʔaliˈwid] ‗friend (the kinship relation between the parents of spouses)‘ (66).1 One of the main problems with SDL‘s transcriptions, however, resulted from his inability to hear the unrounded central vowel which ranges from high to mid positions, and tends to be fronted before velars. In the Bontok broad phonetic examples to follow, the sound will be represented as [ɨ] (or before velars as [ɛ]), although it is frequently identified in other works as schwa [ə]. SDL lists it among the vowels as ö, ―as in G[erman] könig, or F[rench] feu.‖ He also notes that ―final ö is frequently followed by a scarcely audible y‖, the latter combination being equivalent to one of his diphthongs, öy. He notes, ―e and i are constantly interchanged; often ö is pronounced instead‖ (5). He also confused it with a. ―A, which usually has a clear sound, is sometimes obscured, especially in unaccented syllables. In a few words initial a is interchanged with i.‖ One of the examples he gives is ǎpắt vs. ǐpắt ‗four‘. This form in all areas of Bontok and Kankanaey is [ʔɨˈpat]. Another example he cites, however, involves the prepositions is and id vs. as and ad (5). For these forms, the variation is real, but dialectal; some towns use one set, other towns use the other set and no central vowel is involved (Reid 2006). p b m

t d n s l

k g 



i

 a

u

ˈCV (Syllable stress) w y Table 1: Central Bontok phonemic system in 1960 (from Reid 2005: 385)

2.2. Consonants When SDL did his research, Bontok had fourteen consonantal phonemes, however within recent generations, a number of allophonic variants that have English equivalents have taken on phonemic status, primarily because of education in English, and borrowings from Ilokano and other Philippine languages (Reid 1963, 2005) (see Table 2). He lists 17 consonants, ―b, d, f, g, k, l, m, n, p, s, t, w, y, ng, sh, dj, tj‖. In addition, SDL notes that there is a ―Glottal Check‖ in Bontok, stating that its occurrence is ―strictly idiomatic; the words… in which it is employed can only be learned by observation.‖ When he heard it, he represented it with a slash (/), but did not list it as one of the consonants in the language. Generally however he didn‘t hear it (or didn‘t think it should 1

Where Seidenadel‘s definitions are mistaken, or inadequate, additional terms are added in parentheses, see Reid (1976) and the revised, on-line version of the Talking Dictionary of Khinina-ang Bontok: The Language Spoken in Guina-ang, Bontoc, Mountain Province, the Philippines (2009).

Seidenadel‘s grammar of Bontoc Igorot

145

be represented), and sometimes represented it where it doesn‘t occur, for example between a sequence of two identical letters, n/ng for nng ([nŋ]), or geminate consonants (real or perceived), as in in/n kid for [ʔinˈni k d] /ʔinˈnigid/ ‗left-handed‘ (376), papắt/tay for [paˈpattay] ‗(sacrificial) grove‘ (356), or kek/kek for [k ɨtˈkɛk] /gɨtˈkɨk/ ‗I know‘ (373). (For discussion of Bontok geminates, see Aoyama & Reid 2006). p t k  f ts kh b d g i  u m n  e o s () h a l r ˈCV (Syllable stress) w y Table 2: Central Bontok phonemic system in 2004 (from Reid 2005: 391) Apart from glottal stop, there are three voiceless stops in Bontok, /p, t, k/, all unaspirated. None of them alternates in any position with a voiced stop. SDL correctly notes that p and t lack the following ―spiritus asper‖ and that at the end of a syllable they (at least the alveolar and velar stops) are difficult to distinguish. He nevertheless commonly misrepresents them, especially at the end of a word as voiced b and d (because of his German background?), e.g., he represents [ˈʔa nap] ‗look for‘ as anab (with a b), but, [ʔaˈna pɛk] /ʔaˈnapɨk/ ‗I find (look for)‘ as an pek (with p in syllable-initial position). Likewise, [ˈʔi k t] /ˈʔikit/ ‗grandfather (grandparent)‘ is represented as kǐd (12), and [faˈlʊkn t] /baˈluknit/ ‗battle‘ as fǎlǒgnǐd (16). He also sometimes represents voiced consonants as voiceless, even at the end of a word, thus [ˈfo tʊg] /ˈbutug/ ‗pig‘ is represented as f tuk (17), but elsewhere as f tug (14, 334). Although [p] and [t] do not alternate with their voiced counterparts, as claimed by SDL (8), voiced stops [b], [d], and [g] only occur postvocalically in Bontok dialects. Their prevocalic counterparts are respectively [f], [ʦ] and [k ]. [ʦ] is a voiceless, alveolar affricate (SDL‘s tj and ts), and in a few Bontoc towns, such as Mainit and Dalican, a voiced variant occurs, [ʣ] (SDL‘s dj and dz). [k ] is a voiceless, fronted, lightly aspirated velar stop. None of the Kankanaey towns have these variants (Himes 1984–85) (see Figure 1). /b/ /d/  /g/

[b] [d] [g]

/ V __ C, #

/b/ /d/  /g/

[f] [ts] [kh]

/ __ V

Figure 1: Central Bontok voiced stops (1960) (from Reid 2005: 387)

146

Lawrence A. Reid

SDL‘s claim then that f to and b to ‗stone‘ exemplify f and b ―interchange‖, while dj la and d la ‗tongue‘ exemplify dj and d ―interchange‖, is not phonological variation; the first of each pair is restricted to Bontoc towns, the second to Kankanaey towns. SDL did not hear the difference between the two voiceless velar stops, [kh] fronted and aspirated (the prevocalic variant of /g/), and [k] backed and unaspirated (the voiceless stop that contrasts phonemically with /g/). It is only the latter, however, that he frequently represents as g, e.g., mang fag for [maˈŋa fak] /maˈŋabak/ ‗to conquer‘ (317), while he always represents the former as k, e.g., mǎmǎgkǐd for [maˈmagk t] /maˈmaggit/ ‗young girl (young unmarried woman)‘ (16). His example of k and g ―interchange‖, kinw nik and ginw nik for [k nˈwa n k] /kinˈwanik/ ‗I said‘, is neither an example of dialect (or language) variation, but of his inability to distinguish voiceless unaspirated stops from their voiced counterparts; there is no voiced velar stop in the word. The four variants that SDL provides for ‗he makes‘ (5), kapḗna, kap na, kap na, kap na – all of which represent [k abˈʔɨ na] (/gabˈʔɨna/) – illustrate at least three of the problems discussed above, SDL‘s inability to distinguish the high, central vowel from other unrounded vowels, his inconsistency in marking glottal stops, and his tendency to misrepresent unaspirated voiceless and voiced stops. The phoneme /s/ has two conditioned variants in Bontok. Adjacent to the high, front vowel /i/, Guinaang Bontok /s/ has an alveolar articulation point, similar to the pronunciation of /s/ in Tagalog, Ilokano and many other Philippine languages, as well as English, as in Bontok [siˈtso k an] /siˈdugan/ ‗water jar‘. In other environments, /s/ is pronounced as a voiceless post-alveolar apical central fricative [s], i.e., in non-high front vowel environments, the body of the tongue is retracted, with the apex of the tongue pulled back to a post-alveolar position. This sound can also be described as a voiceless alveolar retroflexed grooved fricative [ʂ]. SDL represents this variant of /s/ as sh ―as in shield‖. Although the sound is not palatal, as the English sh representation suggests, it approximates the sound, and even today, some local orthographic representations of Bontok use sh for this sound (Reid 2005: 395). Thus, SDL represents [ˈʔa ŋɨʂ] /ˈʔaŋɨs/ as ngash ‗face (to breathe)‘ (336).

3. Syntax 3.1. Word order Bontok word order is in many respects the same as in most other Philippine languages, with predicates, whether verb, noun or other word class occurring at the beginning of a sentence in unmarked constructions. Topicalized noun phrases occur before the predicate, either linked to it by a monosyllabic ―particle‖, such as Bontok ya or ket, or by a break in intonation. ―Auxiliary‖ verbs such as negatives precede lexical verbs and attract clitics. Copula verbs typically are not found.

147

Seidenadel‘s grammar of Bontoc Igorot

SDL analyzed Bontok as having free word order Substantives, demonstrative and indefinite Pronouns, Numerals, as subjects, either precede or follow the verb. If these subjects precede, the copula ya (for singular and plural and all tenses) is often placed between subject and verb; but never if the subject follows (66). SDL treated personal pronouns differently (see sec. 3.4 below). Large numbers of example sentences are included in the grammar, many of which (despite his claim that all his fieldwork was done monolingually) are clearly elicited via English, with the expected translation effect of copying English word order, as in ex. (1)–(2) (the first line is from SDL; the second is a phonetic representation; the third line provides the same sentence in the orthography of the language as commonly used today; the final line provides an English literal gloss):

(1)

nan

[nan

nan the

(2)

m ma

ya

ʔaˈmamʔa ya

amam-a, old.men

ya

omil-engcha

‗as for the old men, they are resting‘

rest=3PL

masŭyep

‗the child is sleeping‘ (66)

ong nga

ya

nan

ong-onga,

ya

the

ʔʊŋʔʊˈŋa

child

ʔumilˈʔɨŋtsa]

‗the old men are resting‘ (66)

TOPLK

nan

[nan

umiléngtja

ya TOPLK

masɨˈyɨp]

maseyep

‗as for the child, it is sleeping‘

sleep

3.2. Seidenadel‘s problem with ―passives‖ SDL had a completely negative view of all descriptive work previously done on Philippine languages, claiming ―It were best to consider the entire field of Philippine Languages as yet untouched and to begin anew to study [their languages] (but not without personal sympathy with the natives!)‖ (7). This view was based on his recognition, probably unique at the time, that what were claimed as ―the three passives‖ in Philippine languages were, from his point-of-view, not passives at all. The following excerpts from the grammar summarize SDL‘s statements regarding this claim. The term ―las tres pasivas‖ unfortunately invented some centuries ago by Spanish Grammarians for the three active conjugations in other Philippine languages (but not Bontoc Igórot!) must be rejected as erroneous… practically all Philippinists and Copyists of more or less obsolete Philippine Grammars are clinging to the wrong designation… (53) However convenient for minds trained, to some extent, in Latin the Doctrine of the Three Passives has appeared, centuries ago, to its inventor, and however cre-

148

Lawrence A. Reid

dulously his disciples clung to this perverse interpretation of the Active Verbal Noun (Nom. actionis) in Tagalog and in the dialects of several other tribes – in the Bontoc Igorot Language the Verbal Noun is certainly not passive, but active in its character… The fact that the Three Passives Fallacy has been propagated in good faith for two centuries and is still indefatigably copied and republished and taught, show (as also other factors do) how necessary it is to revise and to compare the ―Artes‖ of time-honored ―authorities‖ and the entire material of sacred books, catechisms, confessionals, prayer books, with the living dialects spoken by the natives. The result of such future careful investigations into the people‘s vernacular, the collection of tales and songs in the unbiased dialects of the different tribes ought to be most welcome to Comparative Philologists who seem to rely only on the unreliable material at hand, faute de mieux, material collected by unphilological compilers, with a few admirable exceptions, such as Totanes, [and] Minguella. The unfelicitous term of the Three Passives (which may have sprung from its originator‘s inability to distinguish between the Gerundium and Gerundivum) was employed unscrupulously in many grammars and learned articles and papers on various Philippine dialects; Bontoc Igorot excepted.2 (71–72) SDL then cites (with page references) some twenty-five (mainly Spanish) publications on Philippine languages that he apparently had access to, in which ―the Three Passives and their alleged applications occur‖ (72). These range from the Spanish grammars of Ilokano by Francisco Lopez (1628) and of Tagalog by Sebastian de Totanes (1796) to those of Hiligaynon by Alonso Mentrida (1818), of Ibanag by José Maria Fausto de Cuevas (1854), of Maguindanao by Jacinto Juanmarti (1892), and numerous other works by scholars such as Brandstetter and Kern which appeared toward the end of the 19th and the early 20th centuries. Such early works were also probably available to van der Tuuk, who referred to the three passive constructions of Toba Batak in Sumatra (van der Tuuk 1971 [1864–67]), characterizing them as ―substantives‖, and also to Adriani who similarly claimed in his description of Sangir, the Philippine-type language spoken in Sulawesi (Adriani 1893), The active and passive forms in Sangir are sharply distinguished. The active, as has already often been noted, is actually the only verbal form, the passive is a noun, and the imperative likewise; the distinction between active and passive in the latter is observed only in speech (translated and cited by Blust 2002: 64). In order to put these comments into context, the following is a set of data from the dialect of Bontoc spoken in the Guina-ang community, referred to locally as Khinina-ang. 2

Gerundium: ―In Latin, a verbal noun used in all cases but the nominative.‖ Gerundivum: ―A Latin verbal adjective, with a typical gerund stem form, used as a future passive participle expressing duty, necessity, fitness, etc.‖ (Webster‘s New Twentieth Century Dictionary, unabridged).

149

Seidenadel‘s grammar of Bontoc Igorot

Each of the examples is a simple declarative construction, although with appropriate intonation each could function as an imperative. The actor pronoun in all examples is obligatory, unless recoverable from context.3 SDL calls all such constructions ―active‖. Reid & Liao (2004: 453) refer to them as ―dynamic‖.4 Only constructions patterning like ex. (3)–(4), in which the actor is the grammatical subject, were labeled ―active‖ in early descriptions. Constructions patterning like ex. (5)–(7), in which the actor has a genitive case form and the other noun phrase (the grammatical subject) expresses an undergoer of some sort, were labeled as ―passive‖.5

(3)

omálikad

(4)

omárakas

(5)

ará-em

(6)

(7)

3

4

5

6

ontok si wákas.6 ʔali=ka=ʔad funtuk ʔas ˈwakas come=NOM.2SG=LOC Bontok OBL next.day ‗Come to Bontoc tomorrow.‘

fótog. ʔara=ka=ʔas ˈfutug get=NOM.2SG=OBL pig ‗Get a/some pigs.‘ ʔaˈra-ʔɨn=mu get-DFCT=GEN.2SG ‗Get the pig.‘

charosam

ʦaruˈs-an=mu clean-LFCT=GEN.2SG ‗Clean the house.‘

iyálim

nan

[nan

fótog.

SPEC

ˈfutug]NOM pig

nan

áfong.

[nan SPEC

ˈʔafuŋ]NOM house

nan

fótog.

ʔi-ˈʔali=mu [nan MFCT-come=GEN.2SG SPEC ‗Bring the pig.‘

ˈfutug]NOM pig

This is not only true for Bontok, it is also true for similar constructions in other Philippine languages, such as Cebuano (Payne 1994, Tanangkingsing and Huang 2007). The Appendix contains the same set of data (with corresponding numbers) showing how each sentence would have been analyzed by SDL. Glossing and abbreviations follow the Leipzig Glossing Rules. The first line of each example is given using a local orthography. The second line provides a phonemic transcription, showing verbal affixation and basic forms of verbs and pronominal clitics. It also brackets and labels phrases that are (morphologically) unmarked for case. The third line provides a literal translation for each form and the fourth line is a free translation. Other abbreviations are: DFCT direct affect; LFCT locative affect; MFCT manner affect (sometimes referred to as voice-marking affixes); and TOPLK topic linker. The oblique preposition /as/ has a variant /si/ following words ending in a consonant. Following words ending in a vowel, it is typically reduced to /=s/, as in ex. (4). The full form occurs in deliberate speech, and in sentence initial position.

150

Lawrence A. Reid

Various overviews of the history of terminology in the description of Philippine languages have appeared, including French (1987–1988), Himmelmann (2002), Ross (2002) and Blust (2002). The use of the term ―passives‖ to characterize those constructions in Philippine languages in which the grammatical subject (variously referred to in some works as the topic, trigger, or pivot) expresses some role other than the actor, continued to be used by the American linguists Blake (1904, 1905, 1906b, 1906a, 1917, 1925) and Bloomfield (1917; 1942) in their descriptions of Tagalog, Bisayan and Ilokano, and subsequently by various other linguists, such as Givón (1979), Bell (1976; 1983) in her description of Cebuano, and Wolff (1996: 17) in his reconstruction of Proto-Austronesian ―active and three passive‖ verbal affixes. Linguists belonging to the Summer Institute of Linguistics in the Philippines, beginning around 1955, recognizing the inapplicability of the term passive for these constructions, began using the term ―focus‖ instead, which as noted by Himmelmann (2002: 14) only served to contribute to the confusion in terminology. He warned, nevertheless, that one shouldn‘t be confused by the ―passive‖ terminology, noting that most authors, including Bloomfield, were well aware that the change in voice, that is the change from ―active‖ to ―passive‖ didn‘t affect the transitivity of the overall construction (Himmelmann 2002: 14). But this statement is controversial, as many linguists have claimed that it is only the ―passive‖ constructions that are transitive and that the ―active‖ constructions are all intransitive (see section 4.2). The fact remains however, that the structures in question typically do not meet two of the prototypical minimal conditions for being labeled as passive (Shibatani 1985, Comrie 1988), namely: 1) the defocusing of agents, and 2) markedness, or relatively low text frequency relative to active constructions. SDL was correct in claiming that all such constructions are ―active‖, although referring to them as ―active conjugations‖ incorrectly suggests that there is an inflectional relationship among them. The so-called ―voicemarking affixes‖ which occur on the verbs are not inflectional but derivational,7 in that they cannot freely occur on all verbs, do not freely commute with one another as in a voice-marking system, and are typically maintained in nominalizations and other derivational processes (Reid & Liao 2004: 453). SDL notes that these affixes (―particles‖) transform a root into an active verbal form, and indicate that that action named by the root passes from the agent to an object. They give the active verbal form transitive force (70). SDL was also correct in noting that for each of the active transitive constructions there is a corresponding intransitive construction which can be considered to be passive (71, 7

In the analysis that is followed here, the ―voice-marking‖ affixes are considered to be phonological sequences that indicate that the verb carries an ―affect‖ feature. A DIRECT AFFECT feature indicates that the undergoer is directly or fully affected by the action. A LOCATION AFFECT feature indicates that the undergoer is only surface affected, or is the location of the action. A MANNER AFFECT feature indicates that the undergoer is a concomitant, or is physically transported or otherwise moved through space by the action. Affixes marking a BENEFACTIVE AFFECT feature, distinct from those listed here, is also found in Bontok and is discussed by SDL. He was fully aware of the semantic effects of these affixes and described them in detail.

151

Seidenadel‘s grammar of Bontoc Igorot

93–97). Thus, corresponding to ex. (5)–(7) above, constructions such as ex. (8)–(10) are found in which agents are obligatorily absent. Reid & Liao (2004: 453) refer to them as ―stative‖, although other linguists call them ―passive‖ (Tanangkingsing & Huang 2007).

(8)

na-ára

(9)

nacharosan

nan

na-ˈʔara [nan PRF.STTV-get.DFCT SPEC ‗The pig was gotten.‘

nan

na-ʦaruˈs-an [nan PRF.STTV-clean-LFCT SPEC ‗The house was cleaned.‘

(10) na-iyáli nan na-ʔi-ˈʔali [nan PRF.STTV-MFCT-come SPEC ‗The pig was brought.‘

fótog.

ˈfutug]NOM pig8

áfong.

ˈʔafuŋ]NOM house

fótog.

ˈfutug]NOM pig

3.3. Seidenadel‘s verb classification SDL divides verbs (in ―active declarative main sentences‖) into two major types: a) ―personal verbs‖ and b) ―possessive verbs‖ (52). The former are the equivalent of what in later descriptions would be called ―actor focus‖. The latter are the equivalent of ―nonactor focus‖ verbs. SDL labels the former as ―personal‖ because he treats all pronominal endings, now recognized as nominative clitic pronouns, as inflectional endings on the verb. Without the personal ending (as when a non-pronominal substantive occurs as the grammatical subject) he says the verbs of this category are most similar to our participles or verbal adjectives (51). He says that the term ―intransitive‖ which would be quite appropriate for many verbs of this category (as ex. 3 above) ―would be misleading, as many of them are used also as transitives, though with less transitive force than the verbs of the class b‖ (as ex. 4 above). He correctly notes that such verbs ―have more stress on the verb than on a definite object, the object being indefinite or general or taken in a partitive sense‖ (54). He also notes that such verbs, although transitive have no corresponding passive construction. The transitivity of verbs of this class continues to be argued in the literature, with some claiming that while such verbs carry semantic transitivity, their affixation typically matches clearly intransitive verbs, and the sentences they head are syntactically intransitive antipassive constructions, with downgraded undergoers, case-marked (when expressed) by either oblique (as in Bontok), or genitive or locative (as in Tagalog) prepositions. Of the class b verbs, SDL similarly 8

There is no overt marking for the ―direct affect‖ feature, when the verb carries the perfective infix .

152

Lawrence A. Reid

considers all these forms to be ―nomina actionis‖ (active) verbal nouns, to which the ―possessive suffixes are agglutinated to distinguish the person of the agent‖ (36). He claims that ―Possessive suffixes – equivalent to the possessive genitive of the personal pronouns – are employed in Bontoc Igorot, instead of our possessive pronouns‖ (34). SDL was confronted with the problem of trying to describe the syntax of what are clearly transitive sentences in an ergative system from the viewpoint of someone who was apparently only familiar with accusative languages. This forced him into several problematic analyses, particularly with his description of the pronouns of the language. The following section will provide a brief overview of the way pronouns are currently understood to function in the language as a basis for understanding the analyses that SDL provided.

3.4. Pronouns There are three sets of personal pronouns in Bontok (see Table 3). Two of the sets are enclitic to the verb; one is nominative (or absolutive)9 and functions as the S of an intransitive sentence (as in ex. 11);10 the other is genitive (or ergative) and functions as the A of a transitive sentence or as the possessor of a possessed noun (as in ex. 12). The third set consists of independent forms that are not case marked. They can function without any additional marking as the P of a transitive sentence (as in ex. 13), as a fronted, topicalized S or A (as in ex. 14–15), or as the nominal predicate of a sentence (as in ex. 16). They can also be preceded by a personal oblique-marking preposition to function as dative pronouns (as in ex. 17). 1SG 2SG 3SG 1DL 1PL.EXCL 1PL.INCL 2PL 3PL

Unmarked sak-en sik-a siya cha-ita chakami chataku chakayu cha-icha

Nominative =ak =ka Ø =ta =kami =taku =kayu =cha

Genitive =ku, =k =mu, =m =na =ta =mi =taku =yu =cha

Table 3: Guinaang Bontok personal pronouns 9

10

Although Bontok is an ergative language (an analysis not without controversy), as are many other Philippine languages (Liao 2004), the term nominative is retained as the case of the grammatical subject, for a number of reasons, among which are: a) the syntactic features of an absolutive NP in an ergative language are typically no different from those of a nominative NP in an accusative language, and b) the term nominative has survived from earlier analyses, such as that by SDL, and is commonly used in descriptions of ―Philippine-type languages‖ in the Philippines and elsewhere, regardless of the analyst‘s view as to the case-alignment of the language. Relevant forms in the examples appear in bold font.

153

Seidenadel‘s grammar of Bontoc Igorot

(11) inméyak ad ʔɨy=ak ʔad go=NOM.1SG LOC ‗I went to Bontoc.‘

Fontok.

(12) inneyko ʔɨy=ku go=GEN.1SG ‗I took my pig.‘

nan

fótogko.

(13) inílam ʔila=mu saw.DFCT=GEN.2SG ‗You saw me.‘

sak-en.

(14) sak-en, (ya) [sakʔɨn]TOP ya 1SG TOPLK ‗As for me, I went.‘ (15)  () TOP 1SG

TOPLK

[nan SPEC

funtuk Bontoc

ˈfutug=ku]NOM pig=GEN.1SG11

[sakʔɨn]NOM 1SG

inméyak.

ʔɨy=ak go=NOM.1SG

  =





.

=NOM

took=GEN.1SG SPEC pig=GEN.1SG

‗As for me, I took your pig.‘

(16)    PRED  1SG



 

 =NOM

SPEC PRF.NMLZ-go OBL

SPEC

pig=GEN.2SG

‗I was the one who took your pig.‘ (Lit. ‗The taker of your pig was me.‘)

(17)       =  NOM   MFCT-come=GEN.2SG SPEC pig OBL 1SG ‗Bring the pig to me.‘ The only forms that SDL considered to be true pronouns were the unmarked forms, noting also that shortened forms of the pronouns were attached as personal agreement suffixes on ―personal verbs‖. He considered the unmarked forms to be nominative and was forced to conclude, because the same form was used for the undergoer of a ―possessive verb‖, that ―The form of the personal pronoun for nominative and accusative is alike in Igorot‖ (30). SDL considered the subjects of his ―personal verbs‖ to be in the 11

The ―manner affect‖ feature on the Guina-ang Bontok verb ‗go‘, infix //, is irregular. In other communities this verb carries the regular marking for ―manner affect‖, prefix /ʔi-/, as occurs also on the verb ‗come‘ in ex. (10) and (17).

154

Lawrence A. Reid

nominative case, but not the agents of his ―possessive verbs‖ (65). The pronominal endings of such verbs were considered to be suffixed agreement markers, not pronouns (although he likens them to the ―pronominal subject of our ‗transitive‘ verbs‖ (36). This analysis was supported by the fact that the forms are in fact phonologically attached to the verb, and by his view that transitive verbs are not really verbs at all, but nouns, and therefore the agreement forms were genitive and different from those that appeared on intransitive verbs. SDL‘s analysis suggests that even third person substantives would also be marked on the verb, as, for example, in Chamorro (Topping 1973: 106, Cooreman 1988: 565, Reid 2002: 70), but typically they are not. In order to consider the independent, unmarked pronouns as the grammatical subject of intransitive verbs, SDL needed to analyze the optional topic linker ya in sentences such as ex. 14 and 15 as a copula, giving a possible SV word order. ―If these subjects precede, the copula ya (for singular and plural and all tenses) is often placed between the subject and the verb; but never if the subject follows‖ (15, 66). He recognized however that personal pronouns in sentences with this word order were focused, and therefore ―[t]he personal pronouns, as subjects of verbs, are only used to emphasize the subject.‖ SDL‘s view that unmarked pronouns are used for emphasis fitted neatly into his analysis of the difference between the various transitive sentences, in which ―accusative‖ NPs, when pronouns, were ―emphatic‖. He states, ―Personal pronouns, used only if the subject shall be emphasized, precede the verb‖ (66). Similarly, ―If emphasized, place, cause, instrument, time, the indirect object or dative, etc., can be made the subject … of peculiar verbal forms‖ (27). This is little different from the views of Blake and later linguists who considered that the ―subject‖ or ―focused item‖ functioned to emphasize that NP. While considering the undergoer of a transitive verb to be ―accusatively marked‖, he nevertheless clearly recognized that such NPs don‘t correspond to the accusative NPs of English. In introducing the verbs that head transitive sentences, he said, …by employing special verbal forms the person in whose behalf, for whom an action is performed, or the instrument used in the action, or the place, time, cause, where, when, why the action takes place, took, will take place, can be made the ―subject‖ or ―object‖ as we would say (italics provided) (89). Similarly, in discussing the position of ―[the] ―subject‖ and ―object‖ [of transitive verbs]‖, he notes that ―verbs which we consider customarily [to be] transitive… are of a completely different nature in Bontoc Igorot‖ (25). SDL considered what we refer to today as cleft constructions as constructions which emphasize one of the nominals in a sentence, as illustrated in ex. (16). He says, Thus while some stress is laid upon the elements treated as ―subjects‖ or objects‖, stronger emphasis is expressed by placing the important substantive or pronoun etc. at the beginning of a sentence, followed by nan and the Nomen actionis (92).

Seidenadel‘s grammar of Bontoc Igorot

155

Similarly, ―The ‗Accusative Object‘ as we should say, is strongly emphasized by being placed at the beginning of a sentence, followed by the Nomen actionis with its endings‖ (100).

3.5. Other verbal categories 3.5.1. Voice, tense and mood As noted above, SDL analyzed Bontok as having both active and passive voice, with both personal verbs and possessive verbs having active voice, but only possessive verbs able to appear in passive voice. The active constructions, he says, are much preferred to the passive (53). In this analysis he foreshadows analyses of Philippine languages that have appeared only relatively recently. Verb forms in Bontok have only two tense-aspect forms, one which is basically unmarked, which is used as the infinitive and also for actions that are non-completed, including past, present, on-going, and future actions, and the other marking perfective aspect for actions that are, have been or will be completed at some specified point in time relative to the present. According to SDL, Bontok has three tenses, present, preterite and future. The present is unmarked. The preterite, he says, is marked by ―the particle in-‖ which he then labels an ―augment‖ to distinguish it from the ―prefix in- of some personal verbs‖ (60). This analysis reflects SDL‘s inability to hear, or recognize glottal stop as a consonant, since the so-called ―augment‖ is actually an infix, although with glottal stop initial forms, it may be heard as a prefix. The ―prefix in- of some personal verbs‖ is a true prefix, and the initial glottal stop of a word is always retained after the prefix. Thus Bontok /ʔin-ʔuˈdan/ [ʔinʔʊˈtsan] in-ochan ‗to rain‘ is represented by SDL as in tjǎn (with a prefix and no indication of the glottal stop that follows it, and incorrect stress) (61), while the preterite form of a possessive verb that begins with a glottal stop is shown in the same way. Thus /ʔaˈyag-a-k/ [ʔinaˈya:kʰak] inayákhak ‗I called‘ is represented by SDL as inay kak (78). SDL recognized the ―augment‖ as an infix when ―it is placed between the initial consonant and the following vowel of verbs beginning with a consonant‖ but seemed surprised that an affricate is treated as a single segment in the phonology, giving an example tsinŭblak ‗I smoked‘, with the comment ―[ts] is taken as one consonant!‖ Other ways of marking the perfective form of verbs, such as nin-, na- and nang- that correspond to the unmarked affixes in-, ma- and mang- respectively are correctly identified. SDL claims that future tense is marked by ―the prefix ad- [at-]‖ preceding the forms of the present. The form is actually /at/,12 no voicing is involved, and it is not a prefix but an adverbial form procliticized to the verb, or to other adverbial forms which inter-

12

In Guina-ang Bontok, the equivalent form is /ʔas/.

156

Lawrence A. Reid

vene between it and the verb, as in SDL‘s adts dlo fumǎng nak. for at chadlo fomángonak ―I shall indeed awake‖. Following the outline of grammars of European languages, SDL identified two moods in Bontok… indicative and imperative, although typically the only feature that distinguishes the two is intonation (62).

4. Conclusion SDL‘s grammar of Bontoc is supplemented by a ―Vocabulary‖, actually an English to Bontoc finder list of approximately 2000 English entries, and by about 100 pages of interlinear texts, ―the only Bontoc Igorot texts in existence‖ at that time. (A considerable number of text materials from various dialects are now available, see Reid 1992 and Kikusawa & Reid 2003). SDL‘s texts (with their explanatory notes) are probably the most valuable part of the whole work although flawed by the problems of incorrect transcription discussed in the section above on phonology. Work is in progress to re-transcribe the texts (with corrections of inadequate translations) to increase the value of the work. The major value of the grammatical section of the work is its recognition of the true nature of the so-called ―passive‖ constructions as the main transitive structures in the language, perhaps one of the first of the post-Spanish tradition grammarians to do so, and certainly foreshadowing work that was not to appear for another fifty or so years. The major problems with the work are first its inadequate transcriptions, which completely obscure the patterned phonological alternations that exist in the language, brought about not only by SDL‘s inability to distinguish between voiceless unaspirated and voiced stops and the consonantal significance of glottal stop, but also by his mixing of forms from dialects that have different phonological rules. The other major problem lies in its form of presentation, which follows that of a European grammar, and required that distinctions which exist in such languages be the basis for the description of Bontok, even though SDL was well-aware of the inadequacies of the system for describing the grammar of a language which was so different. SDL‘s view of the nominal nature of the syntax of Bontok, specifically his claim that transitive verbs appear to be nouns in that their actors are expressed by ―genitive‖ forms that are identical to the possessors of nouns in noun phrases was certainly not new then (it had been claimed at least by van der Tuuk and by Adriani (as cited above) in the nineteenth century, and was no doubt a common claim among Spanish grammarians as well, and continued to be a staple of grammarians who have published on Philippine-type languages throughout the twentieth century, for example Naylor (1995, 162), who writes: The syntax of Tagalog predications is based on nominal relations rather than verbal relations and the type of predication is attributive rather than predicational.

Seidenadel‘s grammar of Bontoc Igorot

157

This is because the verbal word, as Lopez (1928; 1941) pointed out, is not a ‗real verb‘ but only a ‗quasi-verb‘; Tagalog ‗verbs‘ are in fact nomina actionis…. (See also Naylor 1979, 1999, 2001, 2005; De Wolf 1988, 1979; Himmelmann 1991, etc.)13 It has long been recognized that the syntax of languages as spoken today is an amalgam of the ebb and flow of individual changes in the history of their phonology and morphology, often characterized as ―grammaticalization‖, which introduce irregularities and ultimately result in systematic restructuring. These are also the processes which have brought about the current systems in so-called Philippine-type languages, of which Bontok is an example. In a paper published some twenty-five years ago on the development of the ―focus system‖ in Austronesian languages, Starosta et al. (1982) claimed that the verb forms of today‘s transitive constructions were in fact nominal derivations in Proto-Austronesian, but subsequently developed as verbs when occurring as predicates. Others (Wolff 1973, 1979, 1996; Ross 1995) have claimed that already by Proto-Austronesian the forms were functioning as verbs in the language. Finally, it should be noted that in the absence of any other substantial grammar of the language (but see Reid 1965, 1970), the editors of the online World Atlas of Language Structures (http://wals.info/index) have mined SDL for facts about the language, referencing it for seventeen typological features. Among those listed which need revision (simply because SDL was mistaken, often basing his claims on English translation equivalents, and not only because there are questions about what constitutes a ―subject‖ or an ―object‖ in an ergative language), are Features 81 and 82, which list it among languages which have no dominant order for either subject, object and verb, or for subject and verb. Bontok is clearly a language that has initial verbs in unmarked constructions and agents (―subjects‖) typically precede undergoers (―objects‖), making it a ―VSO‖ language.

Appendix Guina-ang Bontok examples corresponding to the same numbers in the body of this paper as they would have been analyzed by SDL. (3) -= F   INTR-come-2SG=PLOC Bontok PTIME next.day ‗Come to Bontoc tomorrow.‘ (4) --=  TR-get-2SG=PACC pig ‗Get a/some pigs.‘

13

I wish to thank Hsiu-chuan Liao for providing some of the references in this section.

158

Lawrence A. Reid

(5) --  get-TR-2SG [ART ‗Get the pig.‘



(6) --  clean-TR-2SG [ART ‗Clean the house.‘



pig]ACC

house]ACC

(7) --   TR-bring-2SG [ART pig] ACC ‗Bring the pig.‘ (8) -   PRET.PASS-get [ART pig] NOM ‗The pig was gotten.‘ (9) --   PRET.PASS-clean-TR [ART house]NOM ‗The house was cleaned.‘ (10) --  PRET.PASS-TR-come [ART ‗The pig was brought.‘ (11) --- PRET-INTR-go-1SG ‗I went to Bontoc.‘

 PLOC



pig]NOM F Bontoc

(12) ---  - PRET-TR-go-1SG [ART pig-GEN.1SG]ACC ‗I took my pig.‘ (13) -- - PRET-see-2SG ACC.1SG ‗You saw me.‘ (14) -  NOM.1SG COP ‗As for me, I went.‘

---

PRET-INTR-go-1SG

(15) -  ---  NOM.1SG COP PRET-TR-go-1SG [ART ‗As for me, I took your pig.‘



pig-GEN.1SG]ACC

(16) -      NOM.1SG ART taker PACC ART pig-GEN.2SG ‗I was the one who took your pig.‘ (Lit. ‗The taker of your pig was me.‘)

Seidenadel‘s grammar of Bontoc Igorot

159

(17)     - bring=GEN.2SG ART pig PDAT 1SG ‗Bring the pig to me.‘

Bibliography Adriani, N. (1893): Sangireesche spraachkunst. Leiden: A.H. Adriani. Aoyama, Katsura & Reid, Lawrence A. (2006): Cross-linguistic tendencies and durational contrasts in geminate consonants: An examination of Guinaang Bontok geminates, in: Journal of the International Phonetics Association 36(2), 145–157. Bell, Sarah J. (1976): Cebuano subjects in two frameworks. Ph.D. dissertation, Massachusetts Institute of Technology. Bell, Sarah J. (1983): Advancements and ascensions in Cebuano, in: Perlmutter, David (ed.), Studies in relational grammar I, Chicago: University of Chicago Press, 143–218. Blake, Frank R. (1904): Differences between Tagalog and Bisayan, in: Journal of the American Oriental Society 25, 162–169. Blake, Frank R. (1905): The Bisayan dialects, in: Journal of the American Oriental Society 26, 120–136. Blake, Frank R. (1906a): Contributions to comparative Philippine grammar, in: Journal of the American Oriental Society 27, 317–396. Blake, Frank R. (1906b): Expression of case by the verb in Tagalog, in: Journal of the American Oriental Society 27, 183–189. Blake, Frank R. (1917): The Tagalog verb, in: Journal of the American Oriental Society 37, 396–414. Blake, Frank R. (1925): A grammar of the Tagalog language: The chief native idiom of the Philippine Islands. New Haven: American Oriental Society. Bloomfield, Leonard (1917): Tagalog texts with grammatical analysis. University of Illinois Studies in Language and Literature – vol. III. Urbana: University of Illinois. Bloomfield, Leonard (1942): Outline of Ilocano syntax, in: Language 18, 193–200. Blust, Robert (2002): Notes on the history of focus in Austronesian languages, in: Wouk, Fay & Ross, Malcolm (eds.), The history and typology of Western Austronesian voice systems. Canberra: Pacific Linguistics, Australian National University, 63–78. Chamberlain, Alexander F. (1910): Review of The first grammar of the language spoken by the Bontoc Igórot, with a vocabulary and texts, mythology, folklore, historical episodes, songs, by Carl Wilhelm Seidenadel, in: American Anthropologist, New Series 12(2), 321–323. Comrie, Bernard (1988): Passive and voice, in: Shibatani, Masayoshi (ed.), Passive and voice. (Typological Studies in Language 16). Amsterdam/Philadelphia: Benjamins, 9–23. Conant, Carlos E. (1911): Review of The first grammar of the language spoken by the Bontoc Igorot, with a vocabulary and texts, mythology, folklore, historical episodes, songs, by Carl Wilhelm Seidenadel, in: Classical Philology 6(3), 365–367. Cooreman, Ann (1988): The antipassive in Chamorro: variations on the theme of transitivity, in: Shibatani, Masayoshi (ed.), Passive and voice. (Typological Studies in Language 16). Amsterdam/Philadelphia: John Benjamins, 561–593. De Wolf, Charles M. (1979): Sentential predicates: a cross-linguistic analysis. Ph.D. dissertation, University of Hawai'i. De Wolf, Charles M. (1988): Voice in Austronesian languages of Philippine type: passive, ergative, or neither?, in: Shibatani, Masayoshi (ed.), Passive and voice. (Typological Studies in Language 16). Amsterdam/Philadelphia: John Benjamins, 143–194.

160

Lawrence A. Reid

French, Koleen Matsuda (1987–1988): The focus system in Philippine languages: a historical overview, in: Philippine Journal of Linguistics 18–19, 1–27. Givón, Talmy (1979): On understanding grammar. New York: Academic Press. Himes, Ronald S. (1984–85): Allophonic variation and the Bontok-Kankanaey voiced stops, in: Philippine Journal of Linguistics 15–16, 49–56. Himmelmann, Nikolaus P. (1991): The Philippine challenge to Universal Grammar. (Arbeitspapier Nr. 15, Neue Folge). Institut für Sprachwissenschaft, Universität zu Köln. Himmelmann, Nikolaus P. (2002): Voice in western Austronesian: an update, in: Wouk, Fay & Ross, Malcolm (eds.), The history and typology of Western Austronesian voice systems. Canberra: Pacific Linguistics, Australia National University, 7–16. Kikusawa, Ritsuko& Reid, Lawrence A. (2003): A Talubin text with wordlist and grammatical notes, in: Journal of Asian and African Studies 65, 89–148. Liao, Hsiu-Chuan (2004): Transitivity and ergativity in Formosan and Philippine languages. Ph.D. dissertation, University of Hawai`i. Lopez, Cecilio (1928): Comparison of Tagalog and Iloko. Hamburg: J. J. Augustin. Lopez, Cecilio (1941): A manual of the Tagalog language. Manila: Institute of National Language. Naylor, Paz Buenaventura (1979): Linking, relation-marking, and Tagalog syntax, in: Naylor, Paz Buenaventura (ed.), Austronesian studies: Papers from the Second Eastern Conference on Austronesian Languages. Ann Arbor: Center for South and Southeast Asian Studies, University of Michigan, 33–50. Naylor, Paz Buenaventura (1995): Linguistic expressions of Asian basic cultural ethics. Lecture given at the Institute of Asian Cultural Studies, International Christian University, Tokyo, Japan. Naylor, Paz Buenaventura (1999): Nominal syntax in verbal predications. Paper presented at the Linguistisches Kolloquium, Johannes Gutenberg-University of Mainz. Naylor, Paz Buenaventura (2001): Nominal syntax in verbal predications: implications for Austronesian typology. Paper presented at the International Symposium on Malay and Indonesian Studies, Max Planck Institute, Leipzig. Naylor, Paz Buenaventura (2005): On the stative predicate: Tagalog ―existentials‖ revisited, in: Liao, Hsiu-chuan & Rubino, Carl (eds.), Current issues in Philippine linguistics and anthropology: Parangal kay Lawrence A. Reid. Manila: The Linguistic Society of the Philippines and SIL Philippines, 419–435. Payne, Thomas E. (1994): The pragmatics of voice in a Philippine language: actor-focus and goalfocus in Cebuano narrative, in: Givón, Talmy (ed.), Voice and inversion. (Typological Studies in Language 28). Amsterdam/Philadelphia: John Benjamins, 317–364. Reid, Lawrence A. (1963): The phonology of Central Bontoc, in: Journal of the Polynesian Society 72, 21–26. Reid, Lawrence A. (1965): Bontoc clause structure. MA thesis, University of Hawai`i. Reid, Lawrence A. (1970): Central Bontoc: sentence, paragraph and discourse. (Summer Institute of Linguistics Publications in Linguistics and Related Fields Publications No. 27). Norman: University of Oklahoma Press. Reid, Lawrence A. (1976): Bontok-English dictionary, with English-Bontok finder list. (Pacific Linguistics C-36). Canberra: Australian National University. Reid, Lawrence A. (1992): Guinaang Bontok texts. (Study of Languages and Cultures of Asia and Africa Monograph Series). Tokyo: Tokyo University of Foreign Studies. Reid, Lawrence A. (2002): Morphosyntactic evidence for the position of Chamorro in the Austronesian language family, in: Bauer, Robert (ed.), Collected papers on Southeast Asian and Pacific languages. Canberra: Pacific Linguistics, 63–94.

Seidenadel‘s grammar of Bontoc Igorot

161

Reid, Lawrence A. (2005): A cross-generational view of contact-related phenomena in a Philippine language: Phonology, in: Quakenbush, J. Stephen & Dayag, Danilo (eds.), Sociolinguistics and language education in the Philippines and beyond: Festschrift in honor of Ma. Lourdes S. Bautista. Manila: Linguistic Society of the Philippines and the Summer Institute of Linguistics. Reid, Lawrence A. (2006): On the origin of Philippine vowel grades, in: Oceanic Linguistics 45(22), 457–472. Reid, Lawrence A. & Liao, Hsiu-Chuan (2004): A brief syntactic typology of Philippine languages, in: Language and Linguistics 5(2), 433–490. Ross, Malcolm D. (1995): Reconstructing Proto-Austronesian verbal morphology: Evidence from Taiwan, in: Jen-kuei Li, Paul; Tsang, Cheng-hwa; Huang, Ying-kuei; Ho, Dah-an & Tseng, Chiuyu (eds.), Austronesian studies relating to Taiwan. Taipei: Academia Sinica, 727–791. Ross, Malcolm D. (2002): The history and transitivity of western Austronesian voice and voicemarking, in: Wouk, Fay & Ross, Malcolm (eds.), The history and typology of Western Austronesian voice systems. Canberra: Pacific Linguistics, Australia National University, 17–62. Shibatani, Masayoshi (1985): Passives and related constructions: A prototype analysis, in: Language 61(4), 821–848. Starosta, Stanley; Pawley, Andrew K. & Reid, Lawrence A. (1982): The evolution of focus in Austronesian, in: Wurm, Stephen A. & Carrington, Lois (eds.), Papers from the Third International Conference on Austronesian Linguistics – vol. 2: Tracking the Travelers. Canberra: Pacific Linguistics, 145–170. Tanangkingsing, Michael & Huang, Shuanfan (2007): Cebuano passives revisited, in: Oceanic Linguistics 46(2), 554–584. Topping, Donald M. (1973): Chamorro reference grammar. Honolulu: The University of Hawai‘i Press van der Tuuk, H. N. (1971[1864–67]): A grammar of Toba Batak. The Hague: Nijhoff. Wolff, John U. (1973): Verbal inflection in Proto-Austronesian, in: Gonzalez, Andrew (ed.), Parangal kay Cecilio Lopez: Essays in honor of Cecilio Lopez on his seventy-fifth birthday. Quezon City: Linguistic Society of the Philippines, 71–91. Wolff, John U. (1979): Verbal morphology and verbal sentences in Proto-Austronesian, in: Naylor, Paz Buenaventura (ed.), Austronesian studies: papers from the Second Eastern Conference on Austronesian languages. Ann Arbor: Center for South and Southeast Asian Studies, the University of Michigan, 153–167. Wolff, John U. (1996): The development of the passive verbs with pronominal prefix in Western Austronesian languages, in: Nothofer, Bernd (ed.), Reconstruction, classification, description: Festschrift in honor of Isidore Dyen. Hamburg: Abera, 15–40.

KLAUS ZIMMERMANN (BREMEN/GERMANY)

The Diccionario español-chamorro (1865) by Padre Fray Aniceto Ibáñez del Cármen: a historiographical characterization of a pedagogic-lexicographic discourse type in late colonial Austronesia Abstract This paper examines the bilingual Spanish-Chamorro Dictionary by Aniceto Ibáñez del Cármen in the context of other lexicographic works of the missionary linguistic tradition in the Spanish colonies. Situated not in the context of the descriptions of indigenous languages but focusing on the Spanish language instruction of the colonized and missionized population by exogenous linguists, the analysis suggests that, instead of being regarded as an objective account of a language‘s lexicon, this (and every) dictionary has to be considered in the light of their goals and the historical circumstances at the time of their conception. Elaborating the specific traits of these contexts and their underlying linguistic theories, this paper also points out the dictionary‘s technical characteristics as well as its documentary nature for the history of the Chamorro vocabulary.

1. Theoretical and methodological remarks My analysis draws upon two theoretical postulates: a dictionary is not an objective representation of the word pool of a language but rather a ―verbal phenomenon‖ (Lara 1997: 16–17 and passim), namely a discourse, i.e. a (discursive) construction of a part of a language. Dictionaries are based on the categorical construction of a distinction between the grammar and the lexicon of a language, they rely on the accepted underlying construction of the concept word, on the form and order of representation of what is considered a word, on a selection of the total number of lexical items, on the way of presenting the meaning or various meanings of a lexeme, on other types of information about the lexeme, and so on. Over the last five hundred years, scientific research about the lexicon and the methodological designs of its presentation has brought about a certain discourse-type, which is called ―dictionary‖ with various variants of lists of words, vocabularies and dictionaries.1 We consider the dictionary as discourse in Foucault‘s sense (1971). To conceive dictionaries as discourses not only in the ordinary sense but in the sense of Michel Foucault (1969, 1971) implies a historiographical analysis that will contribute to the ―archaeology‖ 1

At the very beginning of lexicography, i.e. with the documentation of Latin, there are two different models, that of Antonio de Nebrija (1979 [1492]) and that of Ambrosio Calepino (1502) (see Smith Stark 2009).

164

Klaus Zimmermann

of the construction of what was regarded as the ―truth‖ in each historical era. In our case the construction of a mode of representing the whole of existing words or expressions and their meanings is considered a ―truth‖ construction. Thus, our task consists in explaining a special instance of the dictionary as a discourse, i.e. the conception of what is assumed to be the ―truth knowledge‖ about the lexical part of language constructed by a certain author. Dictionaries, too, must be seen as products of authors who have political, colonial, religious aims. In the case of Missionary Bilingual Lexicography, which is a intercultural situation at a certain period of time, we must consider that it is an activity which has a specific cultural, linguistic and religious background. From the 16th up to the 19th century, Spanish Missionary linguists prepared the ground for research on languages that before had been alien and unknown to the researcher and they were the first to apply linguistic discourse models developed in the context of describing and teaching Latin in other languages. When constructing a linguistic discourse about a language that is not one‘s native tongue, the notion of intercultural perception is a relevant concept to describe academic activities (or, in some respect, to prepare these activities). My general hypothesis is that the concepts of interculturality and transculturality have explicative power not only when it comes to the perception of ―the other‖ in everyday situations, novels and travelogues, but also in the field of scientific research – above all, in those subjects which deal with foreign cultures. The activities undertaken by missionaries were not only descriptions of languages but they often included language planning (Cerrón Palomino 1997, Zimmermann 2005), which resulted in different forms and degrees of transculturation and translingualization in the cultures and languages described. I propose the following set of questions as a guideline for this type of study: 

 



What attitude did linguists with a religious background adopt when confronted with a foreign language, and, above all, with a different vision of the conceptual universe? Are there explicit testimonies? And, to what extent can we ultimately derive that attitude from the way they treated the languages (which would be visible primarily in the lexicographic treatment)? How can the fact that the description of meaning in a foreign language is an intercultural construction be dealt with adequately? Intercultural construction can be manifested in a variety of attitudes: crude ethnocentrism or an attitude of tolerance and conscious acceptance of the right of difference and of the dignity of the language of others, the latter being the objectivist description towards which linguists strive today. What attitudes can we detect in missionary linguistics, especially in the Diccionario of Ibáñez, and through what methods can we identify these? What impression about the vocabulary (and the language) in question results from the discourse about the lexical totality offered by non-native authors? What impression do we furthermore get about the meanings of each word-lexeme, the dif-

The Diccionario español-chamorro

165

ferences between the meanings of the two languages in question, and the implicit value of the languages? I would also like to point out that my intention as a historian of Language Sciences is not to analyse structures and changes of language itself. This would be an issue for historical linguistics. My concern is to examine the dictionary as an instance of a cognitive and discursive construction for a certain purpose in a certain historical period. To this end, I draw on the theory of the monolingual dictionary worked out by Lara (1997). He conceives of dictionaries as verbal phenomena, i.e. as discourses, which leads to conceiving of lexical entries as speech acts in the framework of pragmatics (i.e. the theory of speech acts developed by John Austin and John Searle in the 1960‘s). This perspective allows him to characterize word entries as answers to questions that ask for the meaning of a word. These questions are raised by members of the community speaking the language and therefore are assertive acts. In principle, dictionaries are collections of all the words of a language and the anticipated answers to this type of hypothetical question. However, their function is not limited to the description of the real and diverse meanings: In the framework of a normative construction of language, predominant in the time when this social invention was developed, they turn into instruments of language planning, prescribing which meanings, pronunciations and graphic representations are ―correct‖ and, in a quasi concealed manner, which ones are ―incorrect‖. Thus they can also be ascribed to be regulating (normative) acts directed to the verbal behaviour of the language community. At the level of speech act analysis, we find that the bilingual dictionary shares the characteristics as an assertive act, although the type of answer is not a definition, but one or several equivalents in another language, provided that the reader knows the meaning of the equivalent. Another aspect that concerns the dictionary as a construction in bilingual dictionaries is the belief that it represents equivalents of words in the two respective languages. It is thus of great importance to adequately familiarize oneself with the historical background of a particular dictionary, which involves gaining knowledge about the author‘s identity, the institutional context, the intended users and the proposed purposes of a particular dictionary; and naturally, the question about the extent to which it reached/reaches its goals has to be answered. Instead of criticizing a work written in the past in the light of the modern knowledge we now have about the language (and, depending on the point-of-view of the writer, possibly evaluating it positively or negatively), historiographical analysis aims at elucidating the work‘s purposes, circumstances, users and conceptual basis, highlighting its analytical and constructive achievements. The task is to discover the author‘s contribution to the knowledge of the language and its eventual analytical descriptive and explaining epistemes (epistemological constructions). A further task is to situate the work in the development of lexicographic theory and practice by pointing out the innovations brought about by the work (i.e. its author) on the one hand (cf. Zimmermann 1997, 2004) and the cases where former models were copied on the other.

166

Klaus Zimmermann

2. The history of Chamorro lexicography in the cultural and political context of the Marianas The first linguistic description of Chamorro dates back to 1565. It is a lexicographic text, a simple bilingual glossary, written by Esteban Rodríguez (?–1596), a seaman. It contains 67 entries of Chamorro words with their Spanish equivalences.2 Some examples: chamor baquimaqui churu tana azuiban ami enft tagayaya

amigo daca señor tierra pescado salado agua tamal madera

‗friend‘ ‗hand it over‘ ‗master, lord‘ ‗land‘ ‗salted fish‘ ‗water‘ ‗wrapped food‘ ‗wood‘

The second description of Chamorro is the Diccionario of Ibáñez (to be discussed below) and, in terms of completeness, it may be considered the first dictionary. Further dictionaries are those of Fritz (1908) and Lopinot [1876–1966] (1910), the latter being a Chamorro-German word list, the Chamorro-Wörterbuch. In 1932, the dictionary by Román de Vera (1878–1959) is published. The most recent one is that by Topping, Ogo and Dungca (1975). So far, all Chamorro dictionaries have been bilingual dictionaries; no monolingual dictionary3 has appeared yet.4 This has had rather practical reasons: most Chamorros speakers are literate in English while only a few Chamorros are able to read and write in Chamorro. Both this circumstance and the non-existence of monolingual dictionaries are significant in what we may call the language culture of a speech community, showing the prevailing external view of Chamorro. Potential users of bilingual dictionaries can be divided into two groups: 1. Non-native speakers of Chamorro (in former times, people with colonial or missionary purposes; nowadays those representing the cultural context and domination of the United States, of which Guam forms part). 2. Chamorro speakers who need dictionaries for communicating with foreigners. The result is that Chamorro native speakers who want to get acquainted with their native

2

3

4

A first description was made by Antonio Quilis (1988); he counts 66 entries. Rodríguez-Ponga (2008) counts 67 entries. The first monolingual dictionary in the history of Spanish linguistics is the Spanish dictionary by Covarrubias (2006 [1611]). For the history of Spanish lexicography, see Azorín Fernández (2001). The early editions of the Latin dictionary by Calepino (1502) were monolingual, later transforming into a hybrid type, combining explanations in Latin and equivalents in other languages. This is due to the fact that the Chamorro population today is bilingual, speaking Chamorro and English (Rodríguez-Ponga 2009: 17–18).

The Diccionario español-chamorro

167

language at a conscious level have to go via an external language. The outcome is an alienated relation of Chamorros to their mother tongue.5 It must be underlined that monolingual dictionaries have both a different type of user and a different purpose, for they serve as a means to foster cultural and linguistic development within a society (Lara 1997) and are a symptom of cultural and linguistic selfconsciousness.

3. The Ibáñez‘ Diccionario6 3.1. The Diccionario in the context of Missionary Linguistics There are two different – yet politically connected – areas of Missionary linguistic activities in colonial times of Spain: 1. the description of the languages of the colonized, which served to teach these languages to the missionaries. The background of this work was the conviction that the mission had to be realized in the language of the people addressed. 2. The elaboration of pedagogical textbooks to teach Spanish to the colonized (in relation with evangelization). The first one in Mexico was the Cartilla para enseñar a leer, nuevamente enmendada, y quitadas todas las abreviaturas que antes tenía (1569), attributed to Pedro de Gante.7 The dictionary by Ibáñez is cast in the tradition of the second area. The focus is on Spanish, i.e. not the native language, but with its implicit contrastive approach (due to its bilingual design), it gives also a lot of (unique) information about the state of Chamorro in the second half of the 19th century.

5

6

7

Imagine a situation in which an English youngster who does not know the meaning of an English word needs to speak German and therefore looks it up in a German-English dictionary for a German inter-lingual synonym of this word. In fact, this is a very frequent situation in the context of many indigenous languages, where up to now lexicographers of these languages produce bilingual dictionaries instead of monolingual dictionaries for native speakers. In a way, this is a continuing form of colonization or a remnant of colonial thinking. Even the fact that ―foreigners‖ are the lexicographers is not an argument against this claim. First, because the non-existence of native lexicographers is also a result of colonisation, and second, because the intended use of a dictionary for natives to give them information about their language in another language, i.e. not the native one, as the only source of information, is actually – although not necessarily being aware or intended – a continuing form of colonization. This section is closely related to Stolz (this volume) and his analysis of the same author‘s grammar; both, the dictionary and the grammar, belong together because they have the same purpose, namely that of teaching Spanish to Chamorros. For an evaluation see Bravo Ahuja (1977: 30–55). For other textbooks in colonial Mexico, see Contreras García (1985).

168

Klaus Zimmermann

3.2. A first characterization This is a unidirectional bilingual dictionary. The macrostructure starts from Spanish entries and gives what might be called equivalents in Chamorro. The fact that it was published in 1865 suggests that the author might have been familiar with earlier dictionary discourse traditions. There were a number of bilingual missionary dictionaries in the Americas and Asia as well as the monolingual dictionaries in Spain, beginning with the Tesoro (1611), elaborated by Sebastián de Covarrubias (1539–1613), and those edited by the Spanish Royal Academy, the one by Vicente Salvá and so on. Naturally, there were bilingual dictionaries of other languages that might have been known by the author and that thus served as models. All together, they constituted a sample of criteria and technical apparatus for dictionary design (although Ibáñez does not mention any). It is not easy to make a historiographical description and evaluation of Ibáñez‘ dictionary. In the foreword, there is no information from which we could derive what the author‘s goals were or what the work‘s purpose was. Neither does it contain any information about its conception or possible problems during its preparation. There is neither a discussion of the orthography used for Chamorro words nor a legend of the symbols used. The only metalexicographic indication Ibáñez gives (in the title) is that the dictionary is dedicated to the schools of the Marianas (―que dedica a las escuelas de Marianas‖), which points to the educational sector as the domain where the dictionary was to be used; i.e. the intended users were pupils and teachers of Spanish in Guam. At that time, Spanish colonial domination still existed. In fact, the Diccionario by Ibáñez is not a dictionary of Chamorro but a dictionary of Spanish for Chamorro native speakers.8 This is corroborated by the grammar written by Ibáñez, which is not a grammar of Chamorro but a grammar of Spanish for Chamorro speakers (cf. Stolz, in this volume). Nevertheless, it gives a lot of (implicit) information about Chamorro in the Chamorro equivalents. Another important characteristic of the dictionary is its bilingual character. It is an instance of the type of bilingual dictionary (1492, 1495) invented by Antonio de Nebrija (1441–1522). Bilingual dictionaries start from the theoretical assumption that there exist interlingual synonyms (words with the same semantic and grammatical value) or at least that the special situation of a language learner or translator legitimates establishing a near, if not total, equivalent of the meaning of a word or expression; possible deficiencies can be accepted in favour of the information‘s usefulness, even if the expected equivalence is not really given.

8

In this it is not misleading, as is the title of his Gramática (cf. Stolz, this volume). The title of the Diccionario is clear: Dictionary of Spanish-Chamorro.

The Diccionario español-chamorro

169

3.3. A comparison with other missionary dictionaries Because Guam was still under the colonial domination of Spain at the time of the publication of the Ibáñez dictionary, it is worthwhile to look at possible lexicographical traditions of the time and to compare the Diccionario both with the dictionary discourse invented and developed in this period and with the same missionary context. Although it is evident that the Diccionario was not directed to missionaries, it might be interesting to compare it with missionary dictionaries, looking for similarities or even lexicographic loans in the tradition of missionary linguistics on the one hand, and in the respective ideological and political context on the other. The abundance of religious terms and particularities that have been described for other dictionaries of the same tradition are found in the Spanish-Chamorro dictionary, too. This can be explained by the fact that the teaching of Spanish to Chamorro native people was not separated from an underlying religious concern. Instruction in schools was put in the hands of people in religious positions, who – at that time – obviously combined it with religious missionary aims. This is an ideological trait detectable even today in language teaching materials elaborated by religious people. In this respect, we can state that the Ibáñez Diccionario is different from the Spanish missionary dictionaries of indigenous languages produced in the 16th and 17th century in Latin America and Asia. The purpose of the latter was the description of the indigenous language through equivalents in Spanish. The intended users were Spanishspeaking European missionaries who learned the indigenous language. It was only later that reciprocal bilingual dictionaries were produced, and they followed the model created by Nebrija, who first elaborated a Latin-Spanish dictionary (1492) and some years later a Spanish-Latin dictionary (1495) as its complement. Both Ibáñez and the lexicographers in America and Asia followed his model. Most important lexicographers (although not all of them) in Latin America explicitly stated that they were following the Nebrijan tradition. The first lexicographer, Alonso de Molina (1514–1585) declares in the foreword of his Vocabulario (1557, 1571) that he follows the model of the bilingual dictionary brought about by Antonio de Nebrija and thus establishes an intertextual relation to his precursor. He also describes the intercultural difficulties encountered during the process of making a bilingual description of a language, that is, a dictionary dedicated to language learning by foreigners, in this case, Spanish-speaking friars who wanted to learn Nahuatl for missionary concerns (and not a dictionary for indigenous people who wanted to learn Spanish). This is why the first dictionary of Nahuatl is designed roughly in the same way: Spanish entries and Nahuatl equivalents. Although the macrostructure of the American colonial dictionaries and the one by Ibáñez is the same (Spanish entries explained in a colonized indigenous language), their aims are very different. Ibánez‘ aim was to help learning Spanish. American colonial dictionaries, in contrast, were supposed to help with learning and communicating in the

170

Klaus Zimmermann

respective indigenous language. Hence, the fact that Ibáñez was a friar and the author of other religious books such as El verdadero cristiano instruido and Devoción a San Francisco de Borja, patrón de Rota,9 does not entail that his Diccionario was basically an instrument of the missionary enterprise. It must be taken into account that missionary aims may have already been achieved in the second half of the 19th century. The type of religious instruction in the 16th and 17th century was quite different from that of the 19th century. The fact that the entries of Ibáñez‘ dictionary are Spanish is certainly due to the type of foreign language learning method in the Mariana Islands, where Spanish texts are read and worked with. We can therefore state that it is not a dictionary in the missionary tradition, written for Spanish missionaries, but – as mentioned above – a school book dictionary whose purpose was learning Spanish within a religious context. What can be stated is that the list of entries in Ibáñez‘ dictionary does not follow the list of the Diccionario de la Real Academia Española (DRAE) of his time. We find entries like adherir, aplomo, atrapar, banca, batir, desorientar, embalsamar, evaporar, mensaje, quinqué, útil, which Ibáñez‘ work does not contain while Salvá‘s does (1846). But it may be the case that Ibáñez took other dictionaries, e.g. those written in the Philippines for missionary purposes (cf. García-Medall 2009), as a model. All together, the list merely represents a (pedagogically) relevant selection characteristic of the country.

3.4. Language contact in Chamorro lexicography I am concerned here with the conception and history of lexicography, not with the history of Chamorro.10 However, since Chamorro is a highly mixed language, it is of theoretical and practical lexicographic interest to understand how loans from languages other than the prehispanic Chamorro sample are dealt with and how the eventual conflict between native Austronesian words and loans are treated and represented in the Chamorro equivalents of the Diccionario. Are there purist conceptions or not? This question is pertinent because purism already formed part of the linguistic and lexicographic discourse in the Spanish speaking world at that time, as is manifested in the existence of the Spanish Royal Academy, an institution whose motto was limpia, fija y da esplandor (‗cleans, fixes and gives splendour‘). The Diccionario by Ibáñez is the first dictionary which includes Chamorro after three hundred years of language contact between Chamorro and Spanish. As such it is a rich testimony of language contact at the lexical level. Rodríguez-Ponga (2009: 18) estimates that the number of Spanish transferences (loan words) accounts for about 50–

9

10

It must be noted that Ibáñez wrote this religious text in the Chamorro language. This can be seen either as a text for Christianized readers who have certain competence in reading Chamorro or as a text for teaching reading comprehension in Chamorro with the secondary aim of religious instruction. Cf. the recent description of the history of this language by Rodríguez-Ponga (2009: 31–70).

171

The Diccionario español-chamorro

60% of the Chamorro vocabulary.11 We can find many transferences from Spanish12 (and other languages such as Nahuatl13 via Spanish) to Chamorro in the ―Chamorro equivalents‖ of Ibáñez‘ Diccionario, which therefore, in some respects, can be seen as a documentation of the history and change of the Chamorro language through contact and of the history of transculturation taken place in this culture.14 We will see that the way of including transferences as a result of language contact not only represents a source for the study of contact-induced language change and the possibility of ―measuring‖ the degree of mixture that the lexicon had reached at Ibáñez‘ time; it also reveals the nonnative lexicographer‘s attitude towards these transferences, he, who was a member of the colonial donor culture and language, the Spanish one. An analysis of the treatment of transferences can be classified in the following manner: 1.

Simple transferences (loans) from Spanish (often assimilated to Chamorro phonetics and phonology). There are a great many of these, some of which may have displaced authentic Chamorro words,15 others are due to new cultural entities (things and concepts) which did not exist before contact with Spaniards and Spanish culture.

(1) New things and concepts Raza, s.f. Vino, s.m. Vinagre, s.m. Vaca, s.f. Torre, s.f. Tortilla, s.f. Tamarindo, s.m. Lana, s.f. Lata, s.f. Leche, s.f. Llave, s.f. Baile, s.m. Nota, s.f. 11

12 13 14

15

rasa Bino Binagle Baca Tore, Campanayo Titiyas Camaiéndô Lana Lata Leche Yabe Baila Nota

‗race‘ ‗vine‘ ‗vinegar‘ ‗cow‘ ‗tower‘ ‗tortilla‘ (unleavened flatbread) ‗tamarind‘ ‗wool‘ ‗can‘ ‗milk‘ ‗key‘ ‗dance‘ ‗notice‘

I use the following contact terminology: transferences are elements borrowed from a second language which became part of the native language (also called borrowing or loan); interference refers to elements of a native language which became part of an acquired second language. For more details and theoretical justification, see Zimmermann (2008). Cf. the list and etymological study of the first transferences made by Rodríguez-Ponga (2009: 73–92). For an analysis of transferences from Mexican languages, see Albalá Hernández (2001). The practice of transference impositions of Christian religious terms into languages in order to accomplish missionary goals is largely discussed in missionary theory in New Spain; cf. Zimmermann (2005, 2006). This cannot be proved by the documented words of Chamorro before 1865 because we do not have such documents; what we have is only the small list by Rodríguez. But we can suppose that certain concepts must have existed before; for example, god as a supernatural being (not necessarily as a monotheistic god).

172

Klaus Zimmermann

Novio, s.m. Noticia, s.f. Obispo, s.m. Orilla, s.f. Oro, s.m.

Nobio Noticia Obispo Oriya Oro

‗fiancé‘ ‗news‘ ‗bishop‘ ‗coast‘ ‗gold‘

There are documented transferences from Nahuatl via Spanish. Some examples: (2)

Tomate, s.m. Camote, s.m.

(3)

And from Caribbean languages Maíz, s.m. Maijs Caimán, s.m. Caiman

(4)

Tomates Camute

‗tomato‘ ‗sweet potato‘ ‗corn‘ ‗cayman‘

Transferred (imposed) religious terms

Limosna, s.f. Sacristán, s.m. Sacrificar, v.a. Sagrado, adj. Santo, adj. Sacerdote, s.m. Dios, s.m. Diablo, s.f. (sic) Santificar, v.a.

Limosna Saclestan Ofrese § fig. Jágadae Santos, Consagrao Santos Pale [< padre] Yuus [< dios] Lósefet [< luzifer] Nasantos, guatda [< guardar]

‗alms‘ ‗sexton‘ ‗to sacrifice‘ ‗sacred‘ ‗saint‘ ‗priest‘ ‗god‘ ‗devil‘ ‗to sanctify, to canonize‘

(5) Hybrid compositions from Spanish and Chamorro Visitador, s.m. Veterinario, s.m.

Gamanbisita A-amte cabayo

‗official inspector in colonial Spain‘ ‗veterinary‘

Not only do transferences in a dictionary give us information about language contact in the evolution of Chamorro, their lexicographic treatment also reveals the conceptual design of the dictionary, which becomes clear when we compare Ibáñez‘ Diccionario with the Chamorro-Wörterbuch by Fritz (1908). This bilingual dictionary also presents a rather simple structure. German words are paralleled to equivalent Chamorro words or expressions, sometimes alternative ones. But it is also a real contact dictionary, the author being quite aware of the overt contact design of his Wörterbuch: Enlisting Chamorro words including the transferences (loans form Spanish) and giving the Spanish source etymon in a separate column, the fact of the contact process becomes transparent (to a certain degree). This might have been due to the intended German-speaking users and/or dedicated to the philological interest of a cultivated person. German Garten Jäger Gott gütig Schnur Priester

Chamorro hadín, guétta kasadṹles jṹŏs gáibitú(a) kod‘dét pāle

Etymon (jardin), (huerta) (cazador) (dios) (virtud) (cordel) (padre)

‗garden‘ ‗hunter‘ ‗god‘ ‗kind-hearted‘ ‗cord‘ ‗priest‘

173

The Diccionario español-chamorro

3.5. Technical lexicographic patterns of Ibáñez‘ dictionary 3.5.1. Lexical variation In the Diccionario, we often perceive lexical variation of Chamorro in relation to the Spanish entry without further comments on synonymy or partial synonymy: Burlar, v.a. Bullir, v.n. Banquete, s.m. Aplicar, v.a.

Mofea, esitane, fababa, dague, guajlô Loclog, camten, Uminquieto, bologbog, § fig. Ulâ Fandango, guput Naye, fojñe, tojné

‗to joke‘ ‗to boil‘ ‗banquet, dinner‘ ‗to apply‘

When we compare the Chamorro equivalents with the entries in the Diccionario by de Vera (1932), we get the following equivalents for the case of burlar: mofea ‗mofar, insultar, encarnecer‘; esitane ‗mofar, burlarse; bromear, chancearse, hacer algo ironicamente‘; fababa ‗hacer traición, traicionar‘; dague ‗engañar, mentir; ardid, estratagema, fraude, artería‘; guajlô ‗faltar, fallar‘. In fact, these series of partial equivalents manifest possible equivalents depending on the context in which the word occurs. But it is the user who has to decide which of these the right alternative word is. Evidently, the alternative Chamorro equivalents are not distinguished or marked in stylistic or other terms (equivalents, approximations, or metaphoric use, for example). 3.5.2. Encyclopaedic and ethnological descriptions In some cases, Ibáñez gives an ethnological description, sometimes providing more than one Chamorro equivalent, but differentiating the meaning of each word: Tejido, s.m.

Tirar, v. a.

Divorcio, s.m.

Tinifeg § de carizo, fr. Segaale de hojas de coco que se usa para techar las casas, fr. Jigae § de hojas de coco para cubrir el caballete. Fr. Pagpanğ § de nina o nete para techar las casas. Fr. Tagon Sagüe § Yute, fagasñaijon, dagaonaijon § Paque § algo a alguno fr. Yotte § contra alguno fr. Dagao Ynápattan umásagua pot sentensian magas na pale

‗roof made of coconut branches‘

‗to shoot‘

‗divorce‘

It is interesting that parts of these descriptions are not made in Chamorro but in Spanish, as is the case for tejido and tirar. On the one hand, this is clearly non-systematic. In the case of tejido, it denotes a typical Austronesian architectural technique, but the word is Spanish. I suppose that, in this given case, the author thought of users other than Chamorro native speakers. The description seems overtly designed for Spanish speakers. However, this does not apply to the case of divorcio: This entry‘s description is

174

Klaus Zimmermann

realized in Chamorro, explaining what divorce is (according to the Catholic Church, as can be seen in the use of pale in the definition, which designates the Catholic priest). There is no information about whether there exists or existed a non- or pre-Christian type of divorce without the licence of an authority, like a priest. 3.5.3. Competing equivalents In some cases, we can find double equivalents in Chamorro that coexist or, in other words, compete; one of Chamorro provenience and the other from Spanish. Double equivalents = Variation of Spanish transferences and words of Austronesian origin Spanish Bien, adj. Concha, s.f. Virgen, adj. Virtud, s.f. Vecino, adj. Vicio, s.m. vaciar, v.a. Raspar, v.a. Razón, s.m. vivo, adj. vocablo, s.m. lengua, s.f. obediente, adj. Nudo, adj. Mezclar, v.a.

Chamorro Bien, guef, maulegja, mauleg Concha, táigonğ Bitjen, gasgas, ti maete, tataaimayuma Bittut § Minauleg Besino, jijot, taotao (y sumasagagui) Bisio, tinaelaye Basia, chuda Raspa, güesgües Jinaso, rason Lâlâ, bibo, macalamya Finijo, palabra Julá, lenğgua Osgon, obediente Tochong, nudo Nadaña, mescla

‗good‘ ‗shell‘ ‗virgin‘ ‗virtue‘ ‗neighbour‘ ‗vice‘ ‗to empty‘ ‗to scrape‘ ‗reason‘ ‗alive‘ ‗word‘ ‗tongue, language‘ ‗obedient‘ ‗naked‘ ‗to mix‘

In the case of double equivalents, transfers from Spanish typically appear first, and then a word of authentic Chamorro origin. This double system of Chamorro in the lexical domain is evidence of the diglossic structure of the Chamorro lexicon, one part constituting a higher variety, the other a lower one. Another hypothesis is that the author was a defender of the purist language ideal, revitalizing older Chamorro words that were out of use at the time of the dictionary. If that were the case, this Diccionario would constitute part of other normative and language planning efforts initiated by foreign lexicographers. This interventionist attitude (and activity) was prevalent in missionary dictionaries from the sixteenth century in Latin America, especially in the lexical domain of religion, which is why it cannot be reduced to the case of Ibáñez. In the cases of razón, vivo y vocablo, the author opted for a different order. I suppose that it is significant that the first of the equivalences is the one which is typically used. This type of question needs further investigation. In the case of lengua, Ibáñez gives two equivalences, but without any indication of the different value of each (these are

The Diccionario español-chamorro

175

Julá ‗tongue‘, lenğgua ‗language‘). We have two different concepts, here. In Chamorro, there are two different words, while, in Spanish, the two concepts are blended in one by metonymy. A short comparison to Román de Vera‘s Diccionario 1. I think that Ibáñez‘ procedure of presenting competing items (of Spanish and of Chamorro origin) is not a symptom of a purist attitude, but of a descriptive orientation. At that time, coexisting or competing words existed due to the language contact in process, and Ibáñez documents this situation. The comparison with Román de Vera (1932), another lexicographer with a real purist attitude, supports this hypothesis. De Vera developed a typographic distinction to mark the status of competing words, such as estranhero, extranjero, taotao huyoñg (signifying ‗man from outside‘). He first gives the Chamorro word loaned from Spanish, then the Spanish equivalent and after that the possible Chamorro expression, composed of authentic Chamorro linguistic material, written in small capital font.16 2. The fact that there are coexisting or competing words – one being a loan, the other still existing from authentic Chamorro (or a neologism created from authentic Chamorro material) – can be seen as a further indicator of the existence of two varieties in Chamorro, perhaps a higher and a lower one, the first being affected by loans, the other constituting a much purer variety. 3. In Ibáñez‘ Diccionario, we do not find indications of differences in the meaning of Austronesian-Spanish word pairs. Salas Palomo and Stolz (2008: 262) report that, today, only some older speakers distinguish between different meanings in these word pairs, young people do not. So perhaps, at the time of Ibáñez, the semantic differences still existed but he did not explicitly pay attention to them in his dictionary. 3.5.4. Culture specific entries and multi-verbal descriptions Since Wilhelm von Humboldt, it has been known that different languages conceptualize the world and its entities in different semantic forms (linguistic relativity). Dictionaries are a prominent place where these different conceptualizations can manifest. But how should they be represented? This is a challenge to bilingual lexicographers, because the central assumption (and, so to speak, the promise to the user of bilingual dictionaries) is that there exist bilingual equivalents. The theory of linguistic relativity is a challenge to this assumption. In Ibáñez‘ Diccionario, there are a number of different conceptualizations in the Chamorro language compared with Spanish. We can distinguish two forms of representation of this difference in concepts in both languages. 16

―De Vera was committed to promoting the authentic Chamorro word, rather than the borrowing from Spanish, even if the authentic Chamorro word had become archaic. This is evidenced by his choice of words in his translations, and by his Chamorro-Spanish dictionary. Many Chamorro words which would have fallen into disuse and obscurity have thus been preserved for future generations.‖ (http//www.guampedia.com/category/ 67-2-spanish-era/entry/123-father-roman-de-vera)

176

Klaus Zimmermann

First and foremost, we have to consider Chamorro concepts that do not exist in Spanish. In many bilingual dictionaries, those particularities are not indicated in the section where the other language is the entry language. In the Spanish lexicographic tradition, beginning with Nebrija, these concepts are taken into account even in the language where they do not exist by means of so called multi-verbal descriptions as entries (Azorín Fernández 2001, Zimmermann 2009). Such is the case with Ibáñez‘ work. One single Chamorro concept (represented by one word) is given an equivalent in the Spanish entry in the form of a multi-verbal description of the meaning of this single Chamorro word. Lexicographically, this is resolved by the repetition of the Spanish entry, which is either semantically close to the concept or part of it, with an additional explanation, each one having a different meaning, though in Spanish it appears as a sort of specification. In contrast to Spanish, in Chamorro the phonological and orthographic shape does not indicate that the equivalents are perceived as neighbouring concepts. In this respect, the lexical equivalence pattern of Ibáñez‘ Diccionario is broken. Examples of culture-specific entries and multi-verbal descriptions: comer v.a.r comer carne o pescado crudo fr. comer a dos carillos fr. doler v.n. doler la cabeza. fr. escaparse, v.r. escaparse de un riesgo fr. estender, v.a. estender el cuello para mirar algo. fr. limpiar, v. a. limpiar los polvos con la mano ú otro instrumento como barriendo. fr. limpiar el trasero cuando se proveyó ó después de provisto. fr. limpiar los dientes con el palito. fr. erizarse, v.r. erizarse las plumas al gallo de temor, fr. hediondez, s.f. hediondez. De agua corrompida. fr.

Cano, chocho § v.n. Chumocho § v.a. fig. Lachae, tucho Fag-la

‗to eat‘

Mumotmot pute Malinegy ilo Malago Umescapa Jutô, talâ Nğajagûe, lúlô

‗to munch with two cheeks‘ ‗to hurt‘ ‗headache‘ ‗to escape‘ ‗to escape from a risk‘ ‗to extend‘ ‗to stretch one‘s neck in order to look at something‘ ‗to clean‘ ‗to clean from dust by hand or by a broom‘

Nagasgas. sausan safè

‗to eat raw meat or fish‘

Etgüe

‗to clean one‘s backside‘

Güegüe

‗to clean the teeth with a toothpick‘ ‗to raise‘ ‗the cock raising the feathers because of fear‘ ‗stench‘ ‗the stinking of bad water‘

Cumalao § v.a. nacalao Dôle, jajatsa gohyaña Minitonğ Pinegsen

177

The Diccionario español-chamorro instrumento, s.m. instrumento para tejer petates, sacos, etc. fr. instrumento que usan los ladrones para limpiar las sementeras. fr.

Jeramienta, instrumento § Documento trastes Sii

‗tool‘

Fusiñas

‗tool used by thieves to clean the place of their robbery‘

‗tool to web petates‘

The examples chosen from Ibáñez‘ Diccionario cover words specifically from the culture of Guam. Such multi-verbal entries are not very abundant. When they occur, we can see that Ibáñez conceptualizes the bilingual semantic space from the Chamorro angle, providing a conceptual description in Spanish for the Chamorro cognitive semantic construction. As such, the Spanish entry is a description of the meaning of the Chamorro word. In these cases, the entry‘s macrostructure is conceived from the Chamorro conceptual point of view. This is a feature of an intercultural and contrastive lexicographic approach, producing a hybrid lexicographic mode, i.e. a semasiological design mixed with an onomasiological design. We can find the same procedure in Alonso de Molina‘s Vocabulario17 and in Nebrija‘s Dictionarium (1495), too. Often, the author gives the same equivalents in Chamorro with different word classes in Spanish: doler v.n. dolor s.m. doloroso adj.

pute pinite, pute pute, napinite

‗to hurt‘ ‗the pain‘ ‗painful‘

This is possibly due to the fact that Chamorro is a language of permeable word classes which are not explicitly marked by morphological features.

4. Summary and conclusion Ibáñez‘ Diccionario is a document of high transculturation and translingualization in the lexical domain, a result of more than three hundred years of contact. 



17

Compared to missionary dictionaries in Colonial America, this work is more standardized. The author has benefited from knowledge of the development of lexicographic instrumental techniques of the 18th and 19th centuries. This is clearly indicated by notations such as ‗fr.‘ and ‗fig.‘, the indication of word class, gender, etc. It is a school book dictionary for Chamorro children who learn Spanish. This statement finds its structural correspondence in the indication of grammatical information in the Spanish entries, whereas there are no such indications in the Chamorro synonyms.

This is analyzed in more detail by Zimmermann (2009).

178 

 





Klaus Zimmermann

The Diccionario has descriptive orientation, being a document of lexical variation, whilst precise indications of style, variety, use, etc. are absent. It does not offer a purist orientation. The indication of variants in the Chamorro section without any specification is due to the fact that Chamorro native speakers are meant to use the book. They want to understand the meaning of the Spanish word, the author presents variants, the readers can choose the ―best‖ one to translate or the one they use in their own variety. An example of this orientation is the case of divorcio, where the author gives an explanation for Chamorro speakers for the Catholic way of divorce. This leads to the conclusion that the presumed users were predominantly Chamorro children or adults who had to learn Spanish (or to read Spanish texts), and not Spanish friars, as was the case in Colonial Spanish America. There can still be found traits that are common in missionary dictionaries (and in Nebrija‘s dictionary) or dictionaries directed to Spanish native users: The multiverbal entries (which are semantic definitions, marked with ‗fr.‘ by Ibáñez) are clearly intended not for Chamorro people but for outsiders; their corresponding Chamorro expression is a mono-verbal lexical unit. The situation of colonial dominance had led to a process of replacement (a substitution of existing words) in the form of relexification. Taking the processes in other parts of the Spanish colonial empire as a reference, it can be assumed that parts of this relexification were imposed for reasons of cultural and religious purity. Finally, it also contains creations of equivalents proposed by the lexicographer. Being exogenous speakers, lexicographers deliberately intervene in the corpus of the described language. The macrostructure, i.e. the word list, however, does not represent the Spanish cultural system only; there are words that clearly refer to the Mariana‘s semantic universe, too. One example is tejido, whose meaning is described as a Guam architectural technique. The dictionary might be seen as bearing testimony to transcultural processes having taken place in the Marianas and the state of the transcultural results. It is not my concern to determine to what extent this is an objective and exhaustive testimony; but, ideologically, it is a construction of a semantic universe that implies and accepts transculturation. This aspect is worth being pointed out because dictionaries were often normative; they did not accept real linguistic and semantic givens but eliminated words or meanings that were not in accordance with their ideological convictions (normative dictionaries). In this case, however, it is evident that the lexicographer‘s tolerance to transferences resides in the fact that he is a member of the culture occupying the territory, i.e. he belongs to the colonizing and imposing culture (religion) and its language. The dictionary‘s descriptive orientation is manifest in religious terminology, too. Obviously, explicit goals of language planning for Chamorro cannot be perceived. The Diccionario‘s first aim is to serve as an instrument for teaching the Castilian language. With regard to Castilian, it does not intervene in an innovative manner. It copies Castilian dictionaries already existing in Spain and only adds specific

The Diccionario español-chamorro





179

concepts of the Mariana Islands. However, it indirectly contributes to the description and lexicographization of Chamorro (including an implicit normativization/standardisation) because it establishes for the first time equivalencies between the words of the two languages; this also provokes (new) constructions of meanings of the Chamorro language, although Chamorro is actually not the language that is focused. Ibáñez also creates a graphic form of Chamorro words, serving as an orthographic model at that time. We must suppose that this model was created by non-native speakers. The Diccionario is a documentation of conceptual-semantic differences between the two languages: After 300 years of cultural domination, these differences still existed. In the dictionary, however, we find only a reduced number of multi-verbal Spanish entries and of ethnological explanations. Some of these indications might be given for native speakers of Spanish consulting the dictionary. The Diccionario evokes an impression that there exist a high number of bilingual synonym equivalents. The question whether this hypothesis corresponds to semantic reality or whether there are slight semantic or stylistic differences remains to be considered, though. I assume that the impression that Ibáñez presents these items as equivalents without any further explanations is due to the fact that the dictionary represents a simplification of the real givens. They are known by Ibáñez but deleted for pedagogical reasons.

Bibliography Albalá Hernández, Paloma (2001): Voces indígenas anericanas en las lenguas del Pacífico, in: Zimmermann, Klaus & Stolz, Thomas (eds.), Lo propio y lo ajeno en las lenguas austronésicas y americanas: Procesos interculturales en el contacto de lenguas indígenas con el español en el Pacífico e Hispanoamérica. Madrid: Iberoamericana/Frankfurt: Vervuert, 279–294. Albalá, Paloma (2002): La labor lingüística de los españoles en las islas Marianas antes de 1940, in: Antón Burgos, Francisco Javier & Ramos Alonso, Luis Oscar (eds.), Traspasando fronteras: el reto de Asia y el Pacífico. Valladolid: Centro de Estudios de Asia, 645–651. Azorín Fernández, Dolores (2001): Los diccionarios del español en su perspectiva histórica. Alicante: Universidad de Alicante. Azorín Fernández, Dolores (2007): Salvá y la Academia española: ¿Dos maneras de concebir el diccionario de la lengua?, in: Bierbach, Mechthild; von Gemmingen, Barbara & Stork, Yvonne (eds.), La palabra atada: contribuciones sobre la evolución de diccionarios y gramáticas del español. Bonn: Romanistischer Verlag, 17–32. Bravo Ahuja, Gloria (1977): La enseñanza del español a los indígenas mexicanos. Mexico City: El Colegio de México. Calepino, Ambrogio (1502): Calepinus. Ad Librum. Mos est putidos: & nouus repertus: Ingens materia ut que at uideri: Proeclarusq[ue] liber: bunbusq[ue] totus: versus addere nominis probati: Mentitis titulis: tubore nullo: Obscuiq[ue] uiri: rudisq[ue] uatis: Auctor sic quasi tunc: bonusq[ue] fiat. Nullis mobile ueritas: fidesq[ue] est. Iis demptisa liber exeas aperta In uulgus facie: fauore nullo: Et

180

Klaus Zimmermann

graiis galeatus: & latinis Nam: credas: alii magis: q[ue] ipse Quoerunt auxilium: petas ab illis. Sed si flatus olet: proba: legasque. Reggio nell‘Emilia: Domnino Bertocchi. Cerrón-Palomino, Rodolfo (1997): La primera codificación del aimara, in: Zimmermann, Klaus (ed.), La descripción de las lenguas amerindias en la época colonial. Frankfurt: Vervuert/Madrid: Iberoamericana, 195–257. Contreras García, Irma (1985): Bibliografía sobre la castellanización de los grupos indígenas de la República Mexicana. Mexico City: Universidad Nacional Autónoma de México. Covarrubias Horozco, Sebastián de (2006 [1611]): Tesoro de la Lengua Castellana o Española. Madrid: Luis Sánchez [complete and illustrated edition by Ignacio Arellano and Rafael Zafra, Madrid: Universidad de Navarra/Iberoamericana / Vervuert/RAE]. Foucault, Michel (1969): Archéologie du savoir. Paris: Gallimard. Foucault, Michel (1971): L‘ordre du discours. Paris: Gallimard. Fritz, Georg (1908): Chamorro-Wörterbuch in zwei Teilen: Deutsch-Chamorro und Chamorro-Deutsch. Auf der Insel Saipan, Marianen, gesammelt. (Archiv für das Studium deutscher Kolonialsprachen 2). Berlin: Reimer. García-Medall, Joaquín (2009): Vocabularios Hispano-Asiáticos. Traducción y contacto intercultural. Soria: Exma. Disputación de Soria. Ibáñez del Cármen, P. Aniceto (1863): I magajet na Quisiliano manaéyac nu y debe ujongui, ugagao, ufatinas yan uchajlao yan y palo na devocion sija na merece ufanmatungo. Manila: Ramirez y Giraudier. Ibáñez del Cármen, P. Aniceto (1865a): Gramática chamorra. Manila: Ramirez y Giraudier. Ibáñez del Cármen, P. Aniceto (1865b): Diccionario español-chamorro. Manila: Ramirez y Giraudier. Ibáñez del Cármen, P. Aniceto (1887): Devoción as San Francisco de Borja, patron Luta: mapula y sacramentos sija ya jaf taimano ufanmaguef-recibe devoción as San Dimans y Mauleg na Ladron. Patron Maleso, yan doctrina na mapula. Manila: Amigos del País. Lara, Luis Fernando (1997): Teoría del diccionario monolingüe. Mexico City: El Colegio de México. Lopinot, P. Callistus (1910): Chamorro Wörterbuch enthaltend I. Deutsch-Chamorro, II. ChamorroDeutsch nebst einer Chamorro-Grammatik und einigen Sprachübungen. Hongkong: Typis Societatis Missionum ad Exteros. Molina, Alonso Fray de (2001[1555]): Aquí comiença vn vocabulario en la lengua castellana y mexicana. Edited by Manuel Galeote. Málaga [Analecta Malacitana, Anejo XXXVII]. Molina, Alonso Fray de (1970a[1571]): Vocabulario en lengua castellana y mexicana. Estudio preliminar de Miguel León-Portilla. 1st part. Mexico City: Porrúa. Molina, Alonso Fray de (1970b [1571]): Vocabulario en lengua mexicana y castellana. Estudio preliminar de Miguel León-Portilla. 2nd part. Mexico City: Porrúa. Moreno Fernández, Francisco (1994): Antonio de Nebrija y la lexicografía americana del siglo XVI. A propósito del Lexicon de Fray Domingo de Santo Tomás, in: Voz y Letra 5(1), 79–104. Nebrija, Antonio de (1979 [1492]): Lexicon hoc est Dictionarium ex Sermone Latino in Hispaniensem. Salamanca [facsimil edition prepared by G. Colón & A. Soberanas. Barcelona: Puvill. Nebrija, Antonio de (1951, 21989 [1495?]): Dictionarium ex hispaniensi in latinum sermonem o Vocabulario español-latín. Salamanca [facsimil edition of the Real Academia Española. Madrid. Quilis, Antonio (1988): El primer vocabulario conocido de las Islas Marianas, in: Lingüística Española Actual 10, 177–181. Rodríguez-Ponga, Rafael (1995): El elemento español en la lengua Chamorra (Islas Marianas). Ph.D. dissertation. Madrid: Facultad de Filología/Universidad Complutense de Madrid. Rodríguez-Ponga, Rafael (2001): Los numerales hispano-chamorras, in: Zimmermann, Klaus & Stolz, Thomas (eds.), Lo propio y lo ajeno en las lenguas austronésicas y amerindias. Procesos

The Diccionario español-chamorro

181

interculturales en el contacto de lenguas indígenas con el español en el Pacífico e Hispanoamérica. Frankfurt: Vervuert, 253–278. Rodríguez-Ponga, Rafael (2008): Esteban Rodríguez‘ vocabulary of the language of Guam (1565). Unpublished paper. Rodríguez-Ponga, Rafael (2009): Del español al chamorro: Lenguas en contacto en el Pacífico. Madrid: Ediciones Gondo. Safford, William E. (1903): The Chamorro language of Guam, in: American Anthropologist 5, 289– 311 and 508–529. Salas Palomo, Rosa & Stolz, Thomas (2008): Pro and contra Hispanisms: Attitudes of native speakers of modern Chamoru, in: Stolz, Thomas; Bakker, Dik & Salas Palomo, Rosa (eds.), Hispanisation. The impact of Spanish on the lexicon and grammar of the indigenous languages of Austronesia and the Americas. Berlin: Mouton de Gruyter, 237–267. Salvá, Vicente (1846): Nuevo diccionario de la Lengua Castellana, Paris. Smith Stark, Thomas (2009): Lexicography in New Spain (1492–1611), in: Zwartjes, Otto (ed.), Missionary linguistics IV: Lexicography. Selected papers from the Fifth International Conference on Missionary Linguistics, Mérida, Yucatán, 14–17 March 2007. Amsterdam/Philadelphia: Benjamins, 3–82. Stolz, Thomas (1998): Die Hispanität des Chamorro als sprachwissenschaftliches Problem, in: Iberoamericana 70(2), 5–38. Stolz, Thomas (this volume): The Gramática chamorra. Topping, Donald M.; Ogo, Pedro M. & Dungca, Bernadita C. (1975): Chamorro-English dictionary. Honolulu: University of Hawai‗i Press. Vera, Roman María de (1932): Diccionario Chamorro-Castellano. Manila: Imprenta y Litografía ―Germania‖. Zimmermann, Klaus (1997): Apuntes para la historia de la lingüística de las lenguas amerindias, in: Zimmermann, Klaus (ed.), La descripción de las lenguas amerindias en la época colonial. Frankfurt: Vervuert/Madrid: Iberoamericana, 9–17. Zimmermann, Klaus (2004): La construcción del objeto de la historiografía de la lingüística misionera, in: Zwartjes, Otto & Hovdhaugen, Even (eds.), Missionary linguistics/Lingüística misionera: Selected papers from the First International Conference on Missionary Linguistics, Oslo, March 13–16, 2003. Amsterdam/Philadelphia: Benjamins, 7–32. Zimmermann, Klaus (2005): Traducción, préstamos y teoría del lenguaje: la práctica transcultural de los lingüistas misioneros en el México del siglo XVI, in: Zwartjes, Otto & Altman, Cristina (eds.), Missionary linguistics II/Lingüística misionera II: Orthography and phonology: Selected papers from the Second International Conference on Missionary Linguistics, São Paulo, March 10–13, 2004. Amsterdam/Philadelphia: Benjamins, 107–136. Zimmermann, Klaus (2006): Las gramáticas y vocabularios misioneros: entre la conquista y la construcción transcultural de la lengua del otro, in: Máynez, Pilar & Rosario, Dosal G. María (eds.), V Encuentro internacional de lingüística en Acatlán. Mexico City: UNAM, 319–356. Zimmermann, Klaus (2008): Constructivist theory of language contact and the Romanicisation of indigenous languages, in: Stolz, Thomas; Bakker, Dik & Salas Palomo, Rosa (eds.), Aspects of language contact: New theoretical, methodological and empirical findings with special focus on Romanicisation. Berlin: Mouton de Gruyter, 141–164. Zimmermann, Klaus (2009): La construcción discursiva del léxico en la Lingüística Misionera: Interculturalidad y glotocentrismo en diccionarios náhuatl y hñähñu-otomí de los siglos XVI y XVII (Alonso de Molina, Alonso Urbano y autor anónimo 1640), in: Revista Internacional de Lingüística Iberoamericana 7.1(13), 161–186. Zimmermann, Klaus (2010): The invention of multiple comparison as a method to overcome ethnocentric perspectives in language description? New World missionary linguists and Wilhelm von

182

Klaus Zimmermann

Humboldt and their studies of Amerindian Languages, in: Stolz, Thomas; Ruigendijk, Esther & Trabant, Jürgen (eds.), Linguistik im Nordwesten: Beiträge zum 1. Nordwestdeutschen Linguistischen Kolloquium, Bremen, Oktober10–11, 2008. Bochum: Brockmeyer, 29–55.

THOMAS STOLZ (BREMEN/GERMANY)

The Gramática chamorra

Abstract This paper highlights a number of linguistically interesting aspects connected with the Gramática chamorra of Fray Aniceto Ibáñez del Carmen published in 1865 in Manila. The grammar describes Spanish (and not Chamorro) and was meant for the use in the public schools of the Mariana Islands during the final phase of Spanish colonial rule. Thus, it is also the first example of the indigenous language of these islands being used as a language of formal instruction. This fact alone singles out the Gramática chamorra as a very precious example of early (and perhaps premature) Ausbau of a language of the Pacific region.

1. Introduction1 The Gramática chamorra is remarkable for a number of reasons. First of all, it is the second book ever printed in Chamorro preceded only by I magajet na Quisiliano (Ibáñez del Cármen 1863), a bilingual Spanish-Chamorro collection of Christian wisdom.2 On the title page of Gramática chamorra, two dates are indicated, viz. 1864 and 1865. The latter refers to the year in which the book was published in Manila whereas the earlier date probably identifies the year in which the manuscript version was completed. Prior to the above two books, whatever was written in Chamorro remained unedited (e.g. two manuscripts by Ibáñez del Cármen, both written in 1860).3 Other early bits and pieces of Chamorro are scattered about in a variety of travel logs and reports (cf. Rodríguez-Ponga forthcoming, Stolz this volume). Secondly, the secular nature of the Gramática chamorra sets it apart from other early examples of literacy in Oceania because the bulk of the texts published in languages of this region in the 19th century have religious contents (Topping 1987) and the first printed books in Samoan (1836), 1

2 3

This article is closely related to Zimmermann (this volume) – a paper which focuses on the lexicological and lexicographical aspects of Ibáñez del Carmen‘s linguistic activities in the Marianas. I am grateful to Emilio Ridruejo for his comments on an earlier version of this contribution. Albalá (2001) provides an electronic edition of the original texts on CD-ROM. Ibáñez del Carmen (1887) is another religious text published in a bilingual Spanish-Chamorro edition.

184

Thomas Stolz

Tahitian (1838), Hawaiian (1839), Tongan (1844), Fiji (1856), Marshallese (1863) and Trukese (1898) are all translations of the Bible or of parts thereof. Also Ibáñez del Cármen (1863) conveys religious content. However, it is remarkable that his next book is of a worldly nature. A third interesting fact about the Gramática chamorra is connected to the misleading title. As a matter of fact, Gramática chamorra is a blatant misnomer because the language described in the grammar is Spanish! Chamorro serves as the language in which the grammatical properties of Spanish are described. In this way, Sanvitores‘ Chamorro grammar-cum-catechism of 16684 (Burrus 1954, cf. Winkler this volume) continued without serious competitor until the publication of Safford‘s (1903) writings. We do not know of any other case as early as this in which the language of the colonised is used in print to teach the language of the colonisers to the indigenous population. Moreover, the very existence of the Gramática chamorra suggests that during the last decades of Spanish rule over the Marianas, Chamorro was one of the languages of instruction in the schools (Pozuelo 1997).

2. Comparing the original with the translation The full title of the Gramática chamorra reveals that it is a translation and not an original: Gramática chamorra que traducida literalmente de la que escribió D. Luis Mata y Araujo dedica á las escuelas de Marianas con el fin de que los niños aprendan el Castellano el P. Fr. Aniceto Ibáñez del Cármen, cura párroco de Agaña [Chamorro grammar which P. Fr. Aniceto Ibáñez del Cármen, parish priest of Hagatña, literally translated from what D. Luis Mata y Araujo wrote and dedicates to the schools in the Marianas in order for the children to learn Spanish] The translator5 refers to the rather successful Spanish grammar-teaching book by D. Luis de Mata y Araujo which went through several editions throughout the 19th century (Urrutia 1986: 285).6 The 1842 edition of the Spanish original7 is entitled Nuevo epitome de Gramática Castellana ò metodo sencillo de enseñar la lengua castellana por los principios generales á la filosofía comun de las lenguas; arreglado tambien á la latina para facilitar su estudio. Puesto en 4

5

6

7

Burrus (1954: 934) makes the misleading statement that both the grammar and the catechism of Sanvitores are in Chamorro. The grammar is in Latin, only the catechism is in Chamorro. Note that Sanvitores‘ work was never printed before its edition by Burrus (1954). Albalá (2002: 649) assumes that Ibáñez del Cármen wrote the grammar in close co-operation with Padre Bernardo Palomo, a well-educated native speaker of Chamorro who was Ibáñez del Cármen‘s coadjutor in the parish of Hagatña at the time. Topping (1973: 5–6) believes erroneously that the Gramática chamorra and the Nuevo epitome are two Spanish grammars of Chamorro. The version we are referring to is a separate Cuban printing which is contemporary to the version published in Spain: Reimpreso en la Habana. Imprenta Cubana – Calle de la Lamparilla no. 90– 1842. Chances are that this separate printing deviates from the parallel publication in the metropolis because of the negligence of the printers who seem to have omitted part of the original by oversight.

The Gramática chamorra

185

dialogo siguiendo los principios de la Academia Española por D. Luis de Mata y Araujo, catedrático de latinidad, retórica y poética de los pages de S.M., individuo y examinador de la Academia latina [New sketch of Spanish grammar or simple method to teach the Spanish language via the general principles of the common philosophy of the languages; also arranged for Latin to facilitate its study, put into dialogical form following the principles of the Spanish Academy by D. Luis de Mata y Araujo, chair of Latin, Rhetoric and Poetics of the pages of H.M., member and examiner of the Latin Academy]. This longish title tells us two things, namely on the one hand, that the author has principled pedagogical, didactic and language-theoretical ideas he considers innovations and, on the other hand, that the contents are presented in dialogical form, i.e. two fictitious personae introduce the elements of Spanish grammar in a constant game of question and answer. Unsurprisingly, the translator copies the dialogical structure of the original. However, this methodological correspondence does not mean that the Gramática chamorra is a straightforward and faithful translation of the Nuevo epitome. To the contrary, there are various discrepancies between the two texts. These deviations of the translation from the original are telling in terms of what Ibáñez del Cármen considered useful information about Spanish for his Chamorro-speaking pupils. If one compares the two books under scrutiny, there are differences in size. The original comes in the octavo size whereas the translation is in quarto. Superficially, this difference seems to account for the different length in terms of number of pages. The original comprises 88 paginated pages, the translation has exactly 50. However, the original and the translation also differ in the realm of content. In the first edition of Nuevo epitome, there was a prologue (propagating the necessity of studying Spanish) which the author dropped from the second edition. This change is explained in a short Advertencia on page 1. Neither the Advertencia nor the original Prólogo are reflected in the translation. Similarly, the translator handles the footnotes included in the original in different ways. For instance, the first three footnotes which give additional information as to the notions of etimología, conjugacion and género neutro (pp. 4–6 in the original) with reference to Latin (and Ancient Greek) are left out altogether in the translation. However, the omission of footnotes cannot be motivated solely by the fact that some of them refer to languages other than Spanish. Ibáñez del Cármen does not deem it necessary either to translate the footnotes dealing with the problems posed, under pluralisation, by proper nouns and nouns whose singular terminates in -as and -is in Spanish. He also passes over Mata y Araujo‘s remarks on the variable use of the singular definite articles la and el with feminine nouns with initial a-. Apart from the fact that Mata y Araujo does not formulate any valid rule for this alternation, the footnote refers to a passage of the original which is left out entirely in the translation, viz. 16 lines which deal with the problem of determining the correct gender of a noun if the article el can combine with nouns of both genders (p. 9 of the original). The same fate is shared by various other footnotes of the original: morphonology of -ísimo (p. 11), rules for the correct use of os and vos (p. 19), the relation between the definite article and the forms

186

Thomas Stolz

of the pronouns of the 3rd persons used as complements of prepositions (p. 19), the central role of the verb in the syntax (p. 22), the arguments against the category of reciprocal verbs (p. 22), the definition of the infinitive as the nominal form of the verb (p. 23), the figurative use of gerunds (p. 26), the irregularly formed passive participles (p. 26), the facultative use of subject pronouns (with additional reference to Latin) (p. 27), additional auxiliary verbs (p. 28), the use of tener que as a modal verb of obligation (with additional reference to Latin and French) (p. 41), the didactic advice to let the pupils conjugate certain irregular verbs (p. 47), the explanation as to the supposed preposition *so (p. 50), two notes on the morphonology of the conjunctions y and o (p. 51), additional complex conjunctions (p. 52), the distinction of two types of superlatives (p. 58), two notes on case government in Latin (p. 59), the infrequency of the present participle in Spanish (p. 60), the explanation of the use of the case labels familiar from Latin (p. 60), the distinction between logical and grammatical structure (p. 62), the identity of word-classes in Spanish and Latin (p. 62), another methodological advice for the teacher to work on the basis of repetition (p. 64), a lengthy note on the use of the gerund as attribute with cross-reference to later parts of the Nuevo epitome (p. 65), the defence of the notion of sintáxis figurada (p. 67). Two additionally ―missing‖ footnotes constitute different cases, however. In the original (p. 24), the author suggests an alternative terminology for the Spanish tense system and expresses his hope that ―la Academia española admita esta distincion.‖ This note introduces among other things, the label presente relativo in lieu of pretérito imperfecto (which Mata y Araujo nevertheless uses in the main text of his 1842 edition). However, the translator employs only those alternative designations. This choice of terms is Ibáñez del Cármen‘s because in a later edition of the Spanish original, the traditional terminology remains unaltered (Emilio Ridruejo, p. c.). Furthermore, the original contains a wordy footnote (p. 15) which explains the rules for determining the grammatical gender of Spanish nouns. The text of this footnote is not lost in the translation because the translator integrates this passage directly into the main body of the text. In sum, only very few footnotes of the original survive in the translation, viz. exactly three, none of which contains explanations. In addition, there are two footnotes which are unique to the translation. What footnotes there are in the Gramática chamorra give alternative readings or alternative word-forms. With so many footnotes missing, the difference in length of original and translation is at least partly explicable. However, what is more decisive is the decision of Ibáñez del Cármen not to translate the final two parts of the Nuevo epitome. The victims of this decision are Parte III: De la ortografia (pp. 73–83) and Parte IV: De la prosodia (pp. 84–88). Thus, the users of the Gramática chamorra are left in the dark about the spelling rules (including the euphonic pronunciation) of Spanish.8 More generally, all infor8

This is especially irritating because in the introductory chapter (p. 3), Ibáñez del Cármen gives the following question-answer pair: P. ¿Cuantas pattes guaja gui gramática? R. Cuatro: rudimentos pat elementos (na jagas mafanaan etimología) sintaxis, prosodia yan ortografia. This is a relatively close translation of the corresponding fragment of a dialog in the original (p. 4, the asterisks indicates a footnote in the original): P. Cuantas son las partes de la gramática? R. Cuatro; rudimentos

The Gramática chamorra

187

mation about segmental and suprasegmental phonological matters is simply omitted. Thus, the translation can be characterised as selective. It can be speculated that the translator passes tacitly over all those issues raised by Mata y Araujo which he, Ibáñez del Cármen, considers superfluous for the teaching of Spanish as a foreign language under the conditions of a long neglected colonial outpost of Spain in the Pacific. This explanation holds for all the learned excursions referring to general linguistics and/or the classical languages. Why orthography, pronunciation and prosody are likewise counted out is less transparent. It cannot be ruled out completely that the translator had not enough energy or time to continue with the translation when he reached the end of part II. The time factor should not be underestimated as a better command of Spanish was to be enforced as part of the policy of Hispanisation adopted by Spain for her remaining overseas territories in the 19th century, meaning: the teaching of Spanish according to an approved method should begin as soon as possible. Moreover, Albalá (2002: 649) mentions the possibility that the publication of the Gramática chamorra is the direct outcome of a visit the Bishop of Cebú paid to Guam in March 1865. In his function as the superior of Ibáñez del Cármen, the Bishop may have ordered the parish priest to print an early version of the manuscript. Before we present some specimens of the text of the Gramática chamorra, it is necessary to glance at the internal organisation of the book. The original comes in parts which in turn are subdivided into chapters. The chapters are then further organised in sections and these again in sub-sections. While parts and chapters are identified by consecutive numbers, all divisions below these levels lack numeration. Parts, chapters, sections and sub-sections also have headings which identify their subject matter. Ibáñez del Cármen makes do without the level of parts such that the book is divided into chapters on the highest level. However, he stops counting with Capítulo 3 (p. 5) and then has no further chapters for the remaining 45 pages, only sections and sub-sections, whereas, in the original, there are also Capítulo 4 (Del verbo) and Capítulo 5 (Del adverbio). Like these chapters, the entire Parte II: De la sintaxis is demoted to the rank of a section in the translation. This demotion does not necessarily imply further cuts in length. We interpret this inconsistency on the part of the translator as a sign of his desire to finish his task on time. In the translation, Mata y Araujo‘s introductory chapter (Capítulo unico: Nociones de la gramatica en general, p. 2–6) is collapsed with the initial paragraphs of his Primera parte (p. 7). This is not incidentally so because of the translator‘s negligence. It is far more probable that Ibáñez del Cármen decided beforehand to re-organise the text without the Partes-level. If so, it makes perfect sense to unite the above passages because the initial paragraph of the Primera parte also treats general aspects of (Spanish) grammar and thus ties up with the topic of the previous section. ó elementos, á lo que se llama etimologia*, sintáxis, prosodia i ortografia. ―Q. How many parts of the grammar are there? A.: Four: basic or elementary notions, which are called etymology, syntax, prosody and orthography.‖ Thus, the user of the Gramática chamorra might expect at least some paragraphs dealing with prosody and orthography – an expectation that is not met, however.

188

Thomas Stolz

However, the questions connected to the general lines of organisation of the Gramática chamorra are only moderately interesting compared to the enormous task of the translator to reduce Chamorro to writing. In the absence of any compelling evidence that there was an established tradition of Chamorro literacy prior to the 1860s, we are led to assume that Ibáñez del Cármen is actually the pioneer of written Chamorro. It is doubtful that Sanvitores‘ rendering of Chamorro dating back to the mid-17th century had any bearing on Ibáñez del Cármen‘s solutions as the discrepancies between the two authors outweigh their similarities by far.9 With the experience of composing the two unedited manuscripts of 1860 in Chamorro, the translator did not have to invent a consistent writing system out of the blue when he set out to translate the Nuevo epitome. The major decisions as to how Chamorro sounds should be represented graphically and how utterances should be divided into words had already been taken prior to the translation. However, Ibáñez del Cármen faced another huge problem in 1864, namely the task of adapting Chamorro to the technical discourse of European philologers and schoolgrammarians.

3. Some observations on the noun With a view to evaluating the achievements of Ibáñez del Cármen in this area, we take a closer look at a section of the Gramática chamorra and compare it to the original version. To this end, we have chosen the section dedicated to the noun (Del nombre, pp. 7– 8 in the original = Y nombre, pp. 4–5 in the translation). The section is presented in the following way. We reproduce the passage from the Grammatica chamorra in bold italics. Each turn of one of the two partners in dialog is followed by the equivalent passage from the Spanish original in plain type. An English translation is provided separately at the end of the dialog.10 The abbreviations P. and R. stand for pregunta ―question‖ (= Q.) and respuesta ―answer‖ (= A.), respectively. The signs * and ý indicate where the original has footnotes left out in the translation. Note also that Mata y Araujo makes use of different fonts and types to distinguish explanatory text (plain type) from examples (italics). This practice is not followed in the translation. For obvious reasons, we keep the orthography of our sources. 9

10

For instance, Sanvitores consistently employs the letter to indicate the glottal fricative [h] whereas Ibáñez del Carmen opts for . This difference may be explained alternatively by changes in Spanish phonology and orthography posterior to Sanvitores‘ writings (Emilio Ridruejo, p.c.). On the possibility that Sanvitores‘ command of Chamorro was rather poor and greatly influenced by his prior exposure to Tagalog in the Philippines, cf. Rodríguez-Ponga (1999: 523, note 14). Burrus (1954: 936) quotes from Sanvitores‘ own report according to which, with the help of a Tagalog speaker once shipwrecked on Guam, he had the grammar ready before he even reached the islands. This English translation follows the Spanish original except for those parts of the section which are missing from the original. The Chamorro version is commented upon in due course below.

189

The Gramática chamorra

Y nombre Del nombre P. ¿Jafa nombre? Qué es nombre? R.Un agúyon na finijo na sirbe para ufanaan yuje sija y mammagajetna sustancia pat calang sustancia, como patgon, lamasa, átbot, minauleg: ayo mina mafanaan sustantivo. Una palabra declinable que sirve para nombrar las cosas ú objetos que son sustancias reales ó figuradas; v.g. niño, mesa, árbol, virtud; por lo cual se llama sustantivo. P. ¿Jaf nai madibibide y nombre? En qué se divide el nombre? R.Gui propio, comun pat apelativo, abstracto yan colectivo. En propio, comun ó apelativo, i abstracto. P. ¿Jaf nombre propio? Qué es nombre propio? R.Ayo y una cosaja janamatútunģo, como Pedro, Jagatña, Asan. El que significa una cosa sola ó un solo individuo; v.g. Pedro, Madrid, Rioseco, Analucia. P. ¿Jaf nombre comun pat apelativo? Qué es nombre comun ó apelativo? R.Ayo y mina matútun ģo sin udestingue todo y clase pat todo y jechuran jaf sija, como gâga, tautau, sonģsonģ. El que significa confusamente todo el género ó toda la especia; v.g. animal, hombre, villa. P. ¿Jaj nombre abstracto? Qué es nombre abstracto? R.Ayo y janamatútun ģo y jechuran tautau pat y ánti, taigüije mojon y de potsija mangaigui: como guiniflii, minaipe, minauleg. El que significa las cualidades de los cuerpos ó del espíritu como subsistentes por sí; v.g. amor, calor, virtud. P. ¿Jaf nombre colectivo? ---R.Y naan ni y gaigui qui singular ya jasasanģan y inetnon megae sija: como ejérsito, docena, manada, catdume. ---P. ¿Mangai singulat yan mangai plurat todos y nombresija? Tienen todos los nombres numero singular i plural. R.La mayot patte mangai iyo nu y dos número menos y propiosija na manaiplurat; guaja locue apelativosija na manai singulat, como albricias, trévedes, puches, víveres.

190

Thomas Stolz

La mayor parte de los nombres tienen los dos números; esceptúanse los propios, que no tienen plural:* hay tambien algunos apelativos que carecen de singular; como albricias, puches, trébedes, vísperas, &c. P. ¿Jaf taimano fatinanasña y plurat? Cómo se forma el plural en los nombres? R.Y finolago sija na nombre manmámagpo gui singulat nu y vocal ni y manacádada pronunsiaña, nu y vocal ni y manánaco pronunsiaña pat consonante: y nombre ni y cadada ultimoña na vocal maañáñade un s para umafatinas y plurat v. g. guine, pluma, mano, manafanhuhuyonģ, plumas, manos; ya y nombre ni y anaco na vocal pat consonante ultimóña, maañade nu y es na silaba v. g. guine, alelí, razon manmanajújuyong, alelíes, razones. Los nombres castellanos terminan en singular en vocal breve, en vocal aguda ó en consonante; al nombre que termina en vocal breve se añade una s para formar el plural; v.g. de pluma, mano; plumas, manos; i al que termine en vocal aguda ó consonante se añade la sílaba es; v.g. de alelí, alelíes; de razon, razones.ý English translation: On the noun. Q.: What is a noun? A.: A declinable word which serves to name the things or objects which are real or figurative substances; e.g. niño ‗child‘, mesa ‗table‘, arbol ‗tree‘, virtud ‗virtue‘; for this reason it is called substantive. Q.: Into what classes is the noun divided? A.: Into proper, common or appellative, and abstract. Q.: What is a proper noun? A.: That which signifies a single thing or a single individual; e.g. Pedro, Madrid, Rioseco, Andalucia. Q.: What is a common or appellative noun? A.: That which signifies the entire class or species indiscriminately; v.g. animal ‗animal‘, hombre ‗man‘, villa ‗small town‘. Q.: What is an abstract noun? A.: That which signifies the qualities of the bodies or of the mind as they exist on their own; e.g. amor ‗love‘, calor ‗heat‘, virtud ‗virtue‘. Q.: What is a collective noun? A.: The noun that is in the singular and expresses the assemblage of many; like ejérsito ‘army’, docena ‘dozen’, manada ‘herd’, catdume ‘swarm’. Q.: Do all nouns have the numbers singular and plural? A.: The bulk of the nouns have both numbers; exceptions are proper nouns which do not have a plural: there are also some appellative nouns which lack the singular; like albricias ‗payment for a favour ‗, puches ‗porridge‘, trébedes ‗tripod‘, vísperas ‗vespers‘. Q.: How is the plural of nouns formed? A.: Spanish nouns end in a short vowel, a stressed vowel or a consonant; to form the plural, an s is added to the noun which ends in a short vowel; e.g. pluma ‗feather‘, mano ‗hand‘; plumas ‗feathers‘, manos ‗hands‘; and the syllable es is added to those nouns which end in an accented vowel or a consonant; e.g. alelí ‗stock (botany)‘, alelíes ‗stocks‘; razon ‗reason‘, razones ‗reasons‘.

The Gramática chamorra

191

3.1. Comments What strikes the eye most is the gap in the original text (marked by – in the Spanish part and by bold italics in the English translation). The question and answer as to the nature of collective nouns are accidentally missing from the edition of the Spanish original we have access to. At the same time, the Chamorro text contains the passage in question and thus we conclude that the translator made use of a different, that is, complete edition or printing of the Nuevo epitome. Since the remainder of the above section attests no further gaps on either side, we proceed with the evaluation in the following order. First, we comment upon orthography and phonological matters. Then we look at the morphology and syntax of the translation. Lexicon, semantics and idioms come in third place. We round up our discussion with some general remarks on Ibáñez del Cármen‘s achievements. For obvious reasons, we have to restrict our evaluation to a relatively small selection of phenomena. 3.1.1. Orthography/phonology Ibáñez del Cármen applies the orthographic conventions of 19th century Spanish.11 This is an unremarkable fact and reflects the general practice of the early missionary linguists and their contemporaries. This orientation towards the Spanish model has the detrimental effect that the highly frequent glottal plosive /Ɂ/ of Chamorro which has no equivalent in Spanish has no place in the translator‘s orthography. In modern orthography, the glottal plosive is represented by . Thus, what in the translation occurs as para ufanaan ‗s/he will name‘ yields para u fanna‘an in present-day Chamorro. The verb form contains the noun na‘an ‗name‘ with an intervocalic glottal plosive. Ibáñez del Cármen makes do with just noting the two vowels: naan. Admittedly, in intervocalic position, the glottal plosive can be understood as somehow implied by diæresis. However, this is different for cases like finolago ‗word‘ (today: fino‘lagu) in which the syllable-final glottal plosive is not reflected in writing. Other phonemes foreign to Spanish are better represented in the Gramática chamorra. The glottal fricative /h/ for instance, is consistently represented by (which in Spanish refers to the velar fricative /x/). This solution has not survived into modern times: words like yuje ‗thing‘ are written yuhi nowadays i.e. the has given way to . The only special character Ibáñez del Cármen adds to the Spanish alphabet is the i.e. a basic with a superscript apostrophe (referred to in Spanish as a tilde). This formidable invention of his (which copies the conventions the Spaniards used in the Philippines) occurs only in combination with a preceding . Thus, this is the translator‘s way to mark the velar nasal /ŋ/ as in sonģsonģ ‗village‘. In modern orthography, /ŋ/ does not need any diacritics: songsong. 11

For obvious reasons, we cannot dwell upon the relative success Ibáñez del Carmen‘s ―colonial‖ orthography had in the years before the first modern orthography was made official in 1971. Suffice it to refer to Topping‘s (1973: 57–8) critical but nevertheless surprisingly moderate judgment.

192

Thomas Stolz

The Spanish variant ways of writing the velar plosives /k/ and /g/ is copied faithfully: /k/ is written except preceding front vowels where is used, whereas /g/ comes as unless /e/ or /i/ follows; if so, is the option. This system has been abandoned so that nowadays /k/ is always written and /g/ always , independent of the phonological context: old calang ‗like‘ = modern kalang, old qui ‗that‘ = modern ki, old patgon ‗child‘ = modern patgon, old gui ‗in‘ = modern gi. The latter makes it necessary that the semivowel /w/ be orthographically marked by a diaeresis: as in old taigüije ‗that is the way‘ = modern taiguihi. For loans from Spanish, the translator overwhelmingly adopts the Spanish orthography and thus transfers other variant graphemes into the nonAustronesian part of the Chamorro lexicon when he writes sustancia ‗substance‘ and docena ‗dozen‘ as Chamorro words (and not as Spanish examples!). Elsewhere in the above section, he employs pronunsiaña ‗its pronunciation‘ and thus creates an unordered alternation of and in front of . We assume that the latter example provides evidence that the translator is adhering to the seseo. The use of is ambiguous. On the one hand, if occurs on its own, it represents the common article, nowadays written /i/ and thus has the value of a vowel (as in y hechuran tautau ‗the human body‘ = modern i hechuran taotao). Everywhere else, conforms to the modern usage i.e. it represents the voiced alveolar affricate /dz/ (as in yan ‗and‘ = modern yan). What else is worth noting in the realm of orthography and phonology? In polysyllabic words, Ibañez del Carmen has mostly final mid-high vowels where the modern orthography prescribes the corresponding high vowels: old ayo ‗that‘ = modern ayu, old catdume ‗crowd‘ = modern katdumi. Beyond that we also find some instances of unassimilated liquids in Spanish loans. If a Spanish word contains /l/ or /r/ in syllable-final position, this liquid is usually turned into the voiceless dental plosive /t/ in Chamorro, cf. patte ‗part‘ < Spanish parte in the translation. Ibáñez del Cármen obeys this rule if the position of the Spanish liquid is word-final: plurat ‗plural‘ < Spanish plural. The one instance of singular in lieu of regular singulat ‗singular‘ < Spanish singular can be attributed to oversight. However, word-internally he allows for exceptions such as sirbe ‗to serve‘ < Spanish sirve ‗it serves‘ which corresponds to modern setbe ~ sietbe. Likewise, the last example shows there are some instances of orthographic in the translation. We do not know the phonetic realisation that was on the translator‘s mind when he wrote down the words. However, since modern Chamorro has no voiced labio-dental fricative /v/ and loans from Spanish and English equipped with this phoneme in the donor language are usually subject to betacism i.e. Chamorro /b/ replaces original /v/ as in modern baitamina ‗vitamin‘ < English vitamin, we assume that the Spanish words were also pronounced with a voiced labial plosive in lieu of the fricative no matter what their Castillian pronunciation was supposed to be. Furthermore, patte ‗part‘ is among the rather few instances of words in which Ibáñez del Cármen seems to notice the existence of ambisyllabic geminate consonants. More often than not, geminates are represented by single graphemes as in locue ‗too, also‘ in lieu of modern lokkue‘.

The Gramática chamorra

193

3.1.2. Morphology/syntax The most striking fact about the morphology of the translator‘s rendering of Chamorro is his tendency to consider certain classes of syntagmatically adjacent elements to form one word whereas the modern orthography regularly decomposes these supposed units into sequences of more than one word. There are two contexts in which this difference between old and new comes to the fore. The first area to explore is plural marking in the noun phrase. The translator makes abundant use of the post-nominal plural marker sija (modern siha, identical with the 3rd person plural pronoun) whose statistical prominence in the Gramática chamorra might pass as unnecessary overuse among native speakers. For modern usage, Topping (1973: 234) states that ―[t]he plural marker siha is often omitted.‖ Ibáñez del Cármen‘s predilection for overt plural-marking is perhaps due to interference from his native Spanish. Apart from this covert Hispanisation in terms of quantity, the translator also oscillates between two ways to treat the relation between the plural marker and its head-noun. In many cases, head-noun and sija are separated by a blank as in yuje sija ‗things‘ (modern yuhi siha) which is fully in line with modern practice. However, equally often, the Gramática Chamorro joins the head noun and its plural marker: y nombresija ‗the nouns‘. In the above section, this strategy of uniting a head noun and its plural marker in one word is restricted to Spanish loans. We can only speculate that the final -s of the correct Spanish plurals of these loans (such as los nombres) and the initial sibilant of the plural marker facilitated the coalescence of the words. The plural marker thus, at least in writing, has alternative morpheme statuses: a free morpheme with Austronesian head nouns and a bound morpheme with Spanish-derived elements. In a way, one may understand this indecision of the translator as another Hispanisation phenomenon in the making. Interestingly, the joining of sija with an immediately preceding element is also witnessed by potsija ‗on their own‘ which of course is not a case of number marking on a noun. Potsija (modern pot siha) reflects the Spanish PP por si ‗by him-/herself/themselves‘ which is a closely knit syntactic unit. Since Chamorro does not have a dedicated set of reflexive pronouns (like Spanish se ~ si, for instance), this univerbation may be an attempt of the translator‘s to differentiate between pot sija ‗by/because of them‘ and ―reflexive‖ potsija ‗by/because of themselves‘. The second type of univerbation is located in the verb phrase and manifests itself in the combination of preverbal pronouns and the verb. In the above text, we encounter six instances in which Ibáñez del Cármen joins the preverbal pronominal to the immediately following verb. These instances are distributed over two verbal constructions, namely the fully transitive ergative in the realis (―non-future‖) and the irrealis (―future‖) which is independent of the transitivity of the clause. Two examples of those are provided in (1)–(2). (1.1) and (2.1) present the sentences as construed by Ibáñez del Cármen whereas (1.2) and (2.2) give their potential modern equivalents. Boldface identifies the word-forms to be looked at more closely.

194

Thomas Stolz

(1) Ergative realis (1.1) Ayo y una cosaja janamatútunģo, un-a cos-a-ha ja-na-ma-tú-tunģo that CA one-F thing-F-only 3SG.ERG-CAUS-PASS-RED-know como Pedro, Jagatña, Asan. like Pedro Hagatña Asan ‗That which designates one thing alone, like Pedro, Hagatña, Asan.‘ (1.2) Ayu i ha namatútungo’ una kosa(s) ha‘, komu Pedro, Hagatña, Asan. (2) Irrealis (2.1) Ayo y that

mina matútunģo sin udestingue ma-tú-tunģo sin u-destingue CA reason PASS-RED-know without FUT.3-distinguish

todo y clase all CA class ‗That is why the entire class is designated without differentiation [literally: without distinguishing].‖ (2.2) Ayu i mina matútungo‘ todu i klasi sin u distengge. For Ibáñez del Cármen, the pronominal ergative marker, i.e. the so-called hu-type pronoun forms part of the verb-form. The same holds for the special pronominal elements used in the irrealis. Thus, janamatútunģo ‗s/he is making them being known‘ is a morphologically complex word-form which corresponds to an orthographic two-word syntagm in modern Chamorro, namely ha namatútungo‘ where ha ‗s/he; it‘ appears as a separate word. Likewise, the translator‘s udestingue ‗s/he will distinguish‘ nowadays yields two orthographic words: u distengge. Pre-verbal pronominals are some kind of prefix i.e. bound morphemes in the Gramática chamorra. In modern practice, they are treated like free morphemes separated from the remainder of the verbal complex (Topping 1973: 106). However, since nothing can be intercalated between the pre-verbal pronominals and the verb, experts of Chamorro like Cooreman (1987: 36) and Chung (1998: 26–27) suggest that these elements are prefixes indeed. In a way, Ibáñez del Carmen thus was far ahead of the times when he adopted a practice that modern analysts find more suitable to the facts of Chamorro grammar than the prescribed spelling rules. In general, the translator tends to lump together (unstressed) clitics and bound morphemes whereas the currently valid orthography forms an umbrella-class which includes both free morphemes and clitics. This has repercussions on the definition of word boundaries and as a consequence also for the definition of the syntactic word in Chamorro. The author of the Gramática chamorra clearly has an idea of the notion of word that differs considerably from the presently established norm. In (1.1), we find further supporting evidence of Ibáñez del Cármen‘s attitude towards clitics. The wordform cosaha ‗only (one) thing‘ can be further analysed into (modern) kosa ‗thing‘ and

The Gramática chamorra

195

the limitative intensifier ha‘ ‗only; too; very‘ which is currently written separately from its neighbour to the left with which however it forms a complex phonological word. We understand now that Ibáñez del Cármen‘s representation of Chamorro is not random and heterogeneous. To the contrary, it is the result of serious reflection upon the structural properties of the language and thus bears evident signs of systematicity. This general impression holds even in the face of the differential treatment of Austronesian and Spanish nouns under pluralisation (but cf. below). To conclude this sub-section, we comment upon the transfer of the morphological properties of Spanish into Chamorro. Example (1.1) again provides some food for thought. There is the syntagm una cosaha ‗one thing only‘. This syntagm illustrates obligatory Spanish gender agreement: the feminine noun cosa ‗thing‘ requires the feminine form of the cardinal numeral uno/una ‗one‘. Chamorro, like most of its relatives, does not belong with the class of gender marking languages. There is no morphologically relevant distinction of masculine and feminine as, for example, in Spanish. However, the massive intrusion of nouns (and adjectives) from the donor language Spanish, the parallel introduction of the Spanish numerals and the gendersensitive indefinite article un/una have created a situation in which the partial adoption of the Spanish agreement system was an option (Stolz 1998). Rodríguez-Ponga (2001: 261) demonstrates that the gender (and number) distinctions of Spanish un(os)/una(s) ‗one; a/an; some‘ are exceptional in written Chamorro and – like una kosa ‗something‘ – can be considered lexicalisations, i.e. fixed expressions. Similarly, formal Spanish plurals as in the quantifier phrase todos y nombresija ‗all the nouns‘ (modern todu i nombre siha) give rise to the impression that within the noun phrase, there is morphological agreement, because todos ‗old‘ contrasts with todo in todo y clase ‗the entire class‘. This looks like a straightforward distinction of singular and plural of todo(s) according to the number of the quantified noun. This distinction has not survived into modern Chamorro. Related to this issue, it is revealing to see that Ibáñez del Cármen avoids the plural in y dos número ‗the two numbers‘ where Spanish would require the noun to be overtly marked for plural: los dos números. However, the translator liberates himself from the Spanish model and employs the Chamorro rule according to which a quantified noun normally does not take a plural marker. If we take into account that the Gramática chamorra was the sole basis on which (Spanish) grammar was taught in the Marianas, chances are that Ibáñez del Cármen‘s version of Chamorro was accepted as a kind of normative pattern by those who attended school in the last three decades of Spanish dominance. Given this, one might agree with Rodríguez-Ponga (1999)12 that, at least in certain domains of grammar, Chamorro was more hispanised in the 19th century than in later times because grammatical rules of Spanish interfered with those inherited from pre-colonial times.

12

Without necessarily subscribing to his hypothesis of the Creole-status of 19th century Chamorro.

196

Thomas Stolz

3.1.3. Lexicon/semantics Generally, the amount of lexical hispanisms in the Gramática chamorra is high even for written Chamorro. This high incidence of hispanisms comes as no surprise because the translator had to create a language for special purposes from scratch, namely philological terminology and ways of saying things. This explains the use of bona fide hispanisms with a philological reading such as vocal ‗vowel‘, silaba ‗syllable‘, nombre ‗noun‘, número ‗number‘, etc. Apart from these technical terms however, Ibáñez del Cármen does not seem to resort to hispanisms beyond what is needed. Most of the nontechnical hispanisms found in the above text are still in use today, e.g. old dibide ‗to divide‘ = modern dibidi, old = modern sin ‗without‘, old = modern dos ‗two‘, old = modern menos ‗less‘, old añade ‗to add‘ = modern añadi, etc. Interestingly, the philological terminology introduced by Ibáñez del Cármen has not been as successful in its entirety as old vocal has been replaced by buet < English vowel, old singulat has given way to modern sensiyu ‗singular‘ < Spanish sencillo ‗simple‘ and old plurat has been ousted by modern motmot (Austronesian) ‗full‘, etc. Nevertheless, modern technical terms in the realm of philology are still overwhelmingly Hispanic in shape and meaning as evidenced by the terminology employed in Rivera (1998). Furthermore, Ibáñez del Cármen successfully resists the temptation to simply transfer certain words from the Spanish original into his Chamorro version. For instance, he correctly judges it inappropriate to take Spanish examples to illustrate the various categories of nouns. What he does instead is translate them into Chamorro or find a suitable example from Chamorro (as, for example, the locally more familiar toponyms Jagatña and Asan in lieu of Madrid and Rioseco).13 The examples of the Spanish original are kept only where morphological properties peculiar to Spanish are at issue. This means that the translator divides the section on nouns into two distinct parts the first of which is assumed to address general aspects of all human languages which also apply to Chamorro and thus can be illustrated with Chamorro words, whereas the second part focuses on typically Spanish structural traits. In this way, the Gramática chamorra willy-nilly also describes some elements of Chamorro, albeit only fragmentarily. Our final example is a semantic misunderstanding of the translator. To better understand the problem, consider (3). (3)

13

Semantic misunderstanding Ayo y janamatútunģo ja-na-ma-tú-tunģo that CA 3SG.ERG-CAUS-PASS-RED-know

That many of these Chamorro terms are loans from Spanish themselves – like átbot (modern atbot) ‗tree‘ < Spanish árbol – suggests that these hispanisms had been fully integrated into the lexicon of Chamorro by the time Ibáñez del Cármen wrote the Gramática chamorra.

197

The Gramática chamorra

y CA

jechuran jechura-n form-LINK

tautau

pat

y

person

or

CA

ánti spirit

‗That which designates the human body or the spirit.‘ This clause opens the answer to the question about the definition of abstract nouns. One wonders what the human body (jechuran tautau) has to do with abstract concepts like love, heat and virtue mentioned in the examples. In point of fact, Ibáñez del Cármen must have misunderstood the meaning of las cualidades del los cuerpos literally ‗the qualities of the bodies‘ in the original. Mata y Araujo uses the expression cuerpos ‗bodies‘ in the sense of ‗concrete objects‘. This venerable usage14 of the term is either unfamiliar to Ibáñez del Cármen or escapes his notice because of the time pressure under which the translator is working. Of course, the problems raised by this misinterpretation also have an effect on the continuation of the clause because it is unclear how the reader is meant to understand y anti ‗the spirit‘ in this context. To assume that in the 19th century Chamorro jechura (< Spanish hechura ‗shape, figure, stature‘)15 had a meaning corresponding to Spanish calidad (old cualidad) ‗quality‘ is certainly too far fetched.16 There is a second instance of the word under scrutiny, namely in todo y jechuran jaf siha translating the original toda la especie ‗the entire species‘. The Chamorro construction can be rendered as ‗the entire form(s) as they are‘. However, this second instance of jechura(n) does not shed any light on the problem illustrated by (3). A likewise too remote possibility to save the translation is to take a detour via the additional meaning ‗doing(s), product(s)‘ of Spanish hechura such that y jechuran taotao would translate as ‗the doings of man‘. On the one hand, this would be in line with the dictionary translation Ibáñez del Cármen (1865b: 48) provides for Spanish hechura, namely jechura and finatinas. Among the meanings of modern fina‘tinas, we find ‗action, deed‘. Whether this solves more problems than it creates is a question we cannot answer here. Thus, we concede that Ibáñez del Cármen might have made a mistake.

14

15

16

This choice of words has a long tradition. We find it as early as Donatus (4th century AD) in his Ars minor whose wording and methodology exerted a strong influence upon countless generations of pre-modern grammarians. In Schönberger‘s (2008: 12) edition of Donatus the noun is defined as Pars orationis cum casu corpus aut rem proprie communiterue significans. ―A declinable part of speech that designates a body or a thing in a special or a general way.‖ There are a number of parallel cases to Spanish hechura where adaptation into Chamorro has brought forth an initial /h/ for the otherwise ―silent‖ Spanish such as modern Chamorro hereria ‗blacksmith‘s workshop‘ = old jerreria < Spanish herrería. Normally, an initial /h/ in a Spanish loan corresponds to the voiceless velar fricative /x/ of Spanish. However, Emilio Ridruejo (p.c.) informs me that especially in southern varieties of Spanish, hechura (occasionally pronounced with an initial glottal fricative) still has a meaning range which includes the general aspects of form, shape and quality of things.

198

Thomas Stolz

3.2. Summary evaluation It is clear that Ibáñez del Cármen‘s Gramática chamorra is not flawless. There is evidence of haste, carelessness and misunderstanding. However, these negative aspects are by no means so numerous as to allow us to disparage the Gramática chamorra. If we consider that the Spanish original too is not an example of exactitude and accuracy, the shortcomings of Ibáñez del Cármen‘s translation become less significant. As a matter of fact, our sample text is relatively close to native Chamorro as one can expect of a technical topic such as this. The quality of the translation thus can be judged as surprisingly high (by this we do not mean brilliant!). We emphasise that the Gramática chamorra is a great achievement. Its purpose of propagating the knowledge of Spanish notwithstanding, it is first and foremost a document of an indigenous language that up to the publication of the Gramática chamorra still was in need of a literary tradition. From this perspective, the Gramática chamorra is even a milestone that deserves to be re-edited and studied in detail.

4. Conclusion The Gramática chamorra suggests that it is worthwhile linguistically to have a closer look at the publications in Chamorro provided by the Spanish missionaries in the 19th century. Not only do we need a very close reading of and comments on the entire text of this grammar of Spanish, but it is also necessary to scrutinise the other bilingual publications by Ibáñez del Carmen (1863: 1887) in order to find out about the possible features that Chamorro had in the last half century of Spanish rule over the Marianas. This together with Sanvitores‘ grammar and the lexicographic work of Ibáñez del Carmen (Winkler this volume, Zimmermann this volume) will provide the ideal grounds for a veritable historical grammar of Chamorro and thus fill a gap in our knowledge of the developments which have led to the present state of affairs.

Abbreviations CA CAUS ERG F FUT

common article causative ergative feminine future

LINK PASS RED SG

linker passive reduplication singular

The Gramática chamorra

199

Bibliography Albalá, Paloma (ed.) (2001): Obras clásicas para el estudio del la antigua Oceanía Española (Colección Clásicos Tavera, Serie II Temáticas para la historia de Iberoamérica, vol. 6). Madrid: DIGIBIS. Albalá, Paloma (2002): La labor lingüística de los españoles en las islas Marianas antes de 1940, in: Antón Burgos, Francisco Javier & Ramos Alonso, Luis Oscar (eds.), Traspasando fronteras: el reto de Asia y el Pacífico. Valladolid: Centro de Estudios de Asia, 645–651. Burrus, E. J. (1954): Sanvitores‘ grammar and catechism in the Mariana (or Chamorro) language (1668), in: Anthropos 49, 934–960. Chung, Sandra (1998): The design of agreement: Evidence from Chamorro. Chicago/London: The University of Chicago Press. Cooreman, Ann M. (1987): Transitivity and discourse continuity in Chamorro narratives. Berlin/New York: Mouton de Gruyter. Ibáñez del Cármen, P. Aniceto (1863): I magajet na Quisiliano manaéyac nu y debe ujongui, ugagao, ufatinas yan uchajlao yan y palo na devocion sija na merece ufanmatungo. Manila: Ramirez y Giraudier. Ibáñez del Cármen, P. Aniceto (1865a): Gramática chamorra. Manila: Ramirez y Giraudier. Ibáñez del Cármen, P. Aniceto (1865b): Diccionario español-chamorro. Manila: Ramirez y Giraudier. Ibáñez del Cármen, P. Aniceto (1887): Devoción as San Francisco de Borja, patron Luta: mapula y sacramentos sija ya jaf taimano ufanmaguef-recibe devoción as San Dimans y Mauleg na Ladron. Patron Maleso, yan doctrina na mapula. Manila: Amigos del País. Mata y Araujo, Luis de (1842): Nuevo epitome de gramatica castellana ò metodo sencillo de enseñar la lengua castellana por los principios generales á la filosofía comun de las lenguas. La Habana: Imprenta Cubana. Pozuelo, Belén (1997): El sistema educativo en las islas Marianas en el siglo XIX, in: García Abásolo Antonio (ed.), España y el Pacífico. Córdoba, Madrid: Dirección General de Relaciones Culturales y Científica, 325–342. Rivera, Maria A. T. (1998): Sostånsian i Chamoru-ta. Hagatña: Gubietnamenton Guåhan. Rodríguez-Ponga, Rafael (1999): ¿Qué de hablaba en las islas Marianas a finales del siglo XIX?, in: Talaván, Miguel Luque; Pacheco Onrubia, Juan José & Aguado, Fernando Palanco (eds.), 1898: España y el Pacífico. Interpretación del pasado, realidad del presente. Madrid: Asociación Española de Estudios del Pacífico, 521–526. Rodríguez-Ponga, Rafael (2001): Los numerales hispano-chamorras, in: Zimmermann, Klaus & Stolz, Thomas (eds.), Lo propio y lo ajeno en las lenguas austronésicas y amerindias. Procesos interculturales en el contacto de lenguas indígenas con el español en el Pacífico e Hispanoamérica. Frankfurt/M.: Vervuert, 253–278. Rodrìguez-Ponga, Rafael (forthcoming): Esteban Rodríguez‘s vocabulary of the language of Guam (1565), in: Stolz, Thomas; Fischer, Steven; Schneemann, Christina & Dewein, Barbara (eds.), The early documents on and in Chamorro. Colloquium on the occasion of the foundation of CHiN – Chamorro Linguistics. An International Network. Bochum: Brockmeyer. Safford, William E. (1903): The Chamorro language of Guam, in: American Anthropologist 5, 289– 311 and 508–529. Schönberger, Axel (ed.) (2008): Die Ars minor des Aelius Donatus. Lateinischer Text und kommentierte deutsche Übersetzung einer antiken Elementargrammatik aus dem 4. Jahrhundert. Frankfurt/M.: Valentia. Stolz, Thomas (1998): Die Hispanität des Chamorro als sprachwissenschaftliches Problem, in: Iberoamericana 70(2), 5–38.

200

Thomas Stolz

Stolz, Thomas (this volume): German and Dutch contributions to Chamorro studies (1800–1920). Topping, Donald M. (with the assistance of Bernadita C. Dungca) (1973): Chamorro reference grammar. Honolulu: University of Hawaii Press. Topping, Donald M. (1987): Literacy in the Pacific Islands, in: Interchange 18(1/2), 48–59. Topping, Donald M.; Ogo, Pedro M. & Dungca, Bernadita C. (1975): Chamorro-English dictionary. Honolulu: University of Hawaii Press. Urrutia, Hernán (1986): Conocimiento, lenguaje y gramática en la obra de Andrés Bello (1781–1865), in: Quilis, Antonio Morales & Niederehe, Hans-Josef (eds.), The history of linguistics in Spain. Amsterdam/Phildelphia: John Benjamins, 263–287. Winkler, Pierre (this volume): Subject, topic, passive and perspective in Functional (Discourse) Grammar and in Philippine Missionary Grammar. Zimmermann, Klaus (this volume): The Diccionario español-chamorro (1865) of Padre Fray Aniceto Ibáñez del Cármen: a historio-graphical characterization of a pedagogic-lexicographical discourse type in colonial Austronesia.

THOMAS STOLZ (BREMEN/GERMANY)

German and Dutch contributions to Chamorro studies (1800–1920)

Abstract This paper addresses linguistically-relevant studies of Chamorro written in German and Dutch until the aftermath of the 1st World War. Owing to the relatively sizable bibliography of German contributions in this area, only a selection of particularly prominent titles is reviewed. For each of the chosen sources, only one subject matter is more closely scrutinised (numerals or phonology/orthography). The achievements and shortcomings of the early grammarians of Chamorro are discussed. It is shown that a re-edition of the texts under review is urgently called for.

1. Introduction In the first two decades of the 20th century, Chamorro studies experienced a peak of intensity. During the entire period of Spanish colonialism in the Marianas (1565–1898/9), only three linguistically relevant works saw the light of day, namely the famous grammar of Sanvitores in 1668 (Burrus 1954; cf. Winkler this volume), the Gramática chamorra and the Diccionario español-chamorro by Ibáñez del Cármen (1865a–b; cf. Zimmermann and Stolz this volume). At the end of the Spanish-American war, the United States took possession of Guam in 1898 as part of their war booty and Spain subsequently sold out her remaining island colonies in Oceania to the German Kaiser in 1899, the new colonial masters of the Marianas took stock of the economic, ethnic, socio-cultural and geographic properties of their new territories. This interest – dictated mainly by the politico-military necessities of exerting absolute control over hitherto unknown people and areas – had a positive side-effect on the linguistic study of Chamorro. Both US and German authorities aimed at cutting off the continuation of Spanish cultural influence in the islands and thus were reluctant to make use of Spanish for too long as a means of communication with the indigenous population. As long as English and German were not commonly known among the islanders, Chamorro was the only politically correct language of communication. In the absence of sufficiently resourceful descriptive and didactic material on Chamorro from the Spanish period, there was thus no other option for the colonial administration than the intensification of work in the field of Chamorro studies. For the American side, the ground-breaking articles by Safford (1903, 1905), republished as a

202

Thomas Stolz

monograph only shortly afterwards (Safford 1909), and the dictionary-cum-grammar by Von Preissig (1918) give testimony to the activities of the Americans in connection with the local language. These early American contributions to the study of Chamorro are relatively well known to the wider international circles of scholars who take an interest in Chamorro. This is different however with those publications of the same period which are not in English. There are a number of interesting articles and books written by authors with a German or Dutch background. Some of these publications are (partly) in Chamorro, if not in German or Dutch. This article focuses on those publications whose fate was to almost fall into oblivion only because of being written in languages which are currently far less important in linguistics than international English. It remains to be seen whether or not these forgotten texts have anything to offer to the linguist of Chamorro. This can be decided only after a thorough evaluation of the texts. Independent of their purely scientific or historic value, they will certainly be interesting to those members of the contemporary Chamorro speech-community who take an interest in the past of their language. For reasons of space, I cannot discuss at length all the texts that fit this description. I concentrate on a selection of particularly prominent publications, namely Fritz (1903), Lopinot (1910c) and Kats (1917) whose opinions on Chamorro phonology are scrutinised. However, I start with a forerunner of these grammarians in section 2.1. Wherever appropriate, I additionally comment upon other publications of the same period. The texts are presented in chronological order. I begin with the German contributions (section 2).1 The Dutch grammar of Chamorro is reviewed in section 3. The conclusions are drawn in the final section 4.

2. German contributions2 2.1. An early forerunner The very first German text I know of dealing with Chamorro belongs to the vast oeuvre of Adelbert von Chamisso (*1781 Boncourt/France, ý1838 Berlin), the famous FrenchGerman romantic poet-explorer, whose writings were popular throughout the 19th century. He participated in the Russian Pacific expedition (the so-called Romantsov expedition which lasted from 1815–1818) whose ships reached Guam in 1817.3 Chamisso wrote a literary travel log in which he describes the situation in the Marianas at the time 1

2

3

For a detailed presentation of the most sizable German text on Chamorro (subsequent to the period under scrutiny in this article), the grammar entitled Die Chamoro Sprache by Costenoble (1940), I refer the reader to Stolz et al.(this volume). For a comprehensive bibliography of everything that was written by Germans or in German about the Marianas and related subjects during the existence of the German colony (1899–1914/20), cf. Spennemann (2004). Chamisso sailed under the command of Captain Otto von Kotzebue on the brig Rurik.

German and Dutch contributions to Chamorro studies

203

of his visit there.4 Linguistically more interesting however is the fact that Chamisso, besides a grammatical sketch of Hawaiian (Elbert & Pukui 1979: 1–3) he wrote as a result of the same expedition, also provides a comparative word-list of Chamorro and a variety of other languages of the Micronesian area. This word-list bears the title Vokabularium der Dialekte Chamori (Marianeninseln) und von Yap, Ulea und Radack [Vocabulary of the dialect Chamori (Mariana Islands) and those of Yapo, Ulea and Radack]. It is included in the appendix to his travel log, the Bemerkungen und Ansichten ―Notes and opinions‖ republished by Max von Sydow (1907c: 67–82). This vocabulary is based on the answers to a small questionnaire in which the equivalents of German numerals and various basic concepts are given in the indigenous languages.5 In a short introductory note, Chamisso explains that he uses German orthography (with an occasional modification) for representing the sounds of Chamorro and other languages – and he is aware of the pitfalls of this practice.6 Since Chamisso‘s work predates the period in focus by well over 80 years, I make do with a look at the Chamorro numerals as presented by Chamisso (Sydow 1907c: 68–69). The numerals are interesting because Chamisso‘s data fill a gap in the chronological overview of Rodríguez-Ponga (2001: 263–271) who tries to determine when exactly the inherited Austronesian numeral system of Chamorro gave way to the Spanish-derived system which is valid nowadays. First of all, Chamisso presents only the Chamorro equivalents for the first ten numerals and the tens up to 100 and the expression for 1,000 whereas he provides more data for two of his sample languages. As he does not tell the reader how he came by the data, we cannot tell whether or not the huge gaps in his list reflect the inability of his informant(s) to remember certain numerals in their Chamorro shape. The absence of entries might just as well be incidental or occasioned by lack of time. I reproduce Chamisso‘s Chamorro numerals in (1). As a matter of fact, Chamisso‘s inventory of ―basic‖ numerals combines numerals from two sets. His first ten reflects those numerals which serve the purpose of counting/quantifying inanimates. Their recurrent ending -(gu)ijai (later -[g]iyai) bears testimony to this. From 20 to 90, however, the numerals are indeed those which belong to the ―basic‖ set whereas 100 and 1,000 are neutral as to the semantic class of the noun they quantify. It strikes the eye that there are two supposedly synonymous expressions (1) Chamorro numerals according to Chamisso 4

5

6

In his report Reise um die Welt (‗Journey around the world‘), Chamisso has a chapter on the Marianas (focusing on Guam) where he describes the natural resources of the islands, social and historical aspects and some remarkable events (Sydow 1907a: 390–400). In the separate diary, Chamisso reports on his personal experience during his visit on Guam where he stayed from 23 November until 29 November, 1817 (Sydow 1907b: 225–231). For the first ten numerals, Chamisso also includes data from Tagalog, Pampango, Bisaya, and Palauan compiled from earlier sources. Chamisso‘s decision may have many different reasons. I am not completely sure that his use of the German orthography can be taken as evidence against the existence of a tradition of writing Chamorro (the Spanish way) in the early 19th century.

204

Thomas Stolz numeral

―basic‖ (cf. below)

―length measurement‖ (cf. below)

1

Hatjijai

Tac hatjun

2

Huguijai

Tac hugua

3

Totguijai

Tac tulum

4

Fatfatai

Tac fatum

5

Limijai

Tac lima

6

Gonmijai

Tac gonum

7

Fedguijai

Tac guijai 7

8

Gualguijai

Tac gualum

9

Siguiajai

Tac siguam

10

Manutai

Tac manud

20

Huguanafulu

30

Tulungafulu

40

Fatfatnafulu

50

Limangafulu

60

Gonumnafulu

70

Fitinafulu

80

Gualungafulu

90

Siguanafulu

100

Manud ~ Gatus

1,000

Tjalan

of 100 in Chamisso‘s list. However, in most of the other listings of the pre-Hispanic Chamorro numerals, 100 is given as gatus ~ gatos. Chamisso‘s manud looks suspiciously like the expected ―basic‖ numeral 10, viz. manut ~ manot which is reported in the other sources. However, in 1668, Sanvitores (Burrus 1954: 950) assumes that gatos represents the value 1,000 (alternatively expressed by monot nasaran ‗literally: ten hundreds‘) whereas 100 is given as acha nasaran ‗literally: one hundred‘ (Rodríguez-Ponga 2001: 264). On the basis of this evidence, one could speculate that the obscure manud for 100 in Chamisso‘s list is not a blatant mistake but perhaps the reflex of an erstwhile numeration practice in which the value 100 could also be expressed by multiplication i.e. by 10 x 10 of which manud ‗ten‘ is a relic of a presumably more extended construction. Among the numerals used for measurements of length, Tac guijai ‗seven long objects‘ is especially noteworthy as it deviates from all other examples of this value and 7

In Sydow‘s edition of Chamisso‘s texts, this numeral is misspelled Gnalguijai (Sydow 1907c: 68). From the other numerals connected to the value 8, it is clear that the should be corrected to .

German and Dutch contributions to Chamorro studies

205

class. It is exceptional because it combines the two ―class markers‖ tac (~ tag-) would indicate the class of long objects whereas guijai (~ -giyai) would indicate the class of inanimates. According to other inventories of the numerals, the marker of inanimates is not permitted in combination with tac (~ tag-). Moreover, in Tac guijai the numeral value 7 is nowhere mentioned explicitly otherwise there should be some reflex of fed(~ fit-) as e.g. in tagfitun ‗seven long objects‘ (Lopinot 1910c: 7, Costenoble 1940: 262). It is of course no wonder that a linguistic field-work session of maximally six days cannot lead to absolutely flawless results. Unfortunately, it is impossible to say with absolute certainty whether or not Chamisso leaves out the Spanish numerals deliberately. Since his goal was to compare the indigenous languages of various islands with each other, chances are that he discarded all non-native elements as they would have skewed the results.

2.2. Governor Fritz In 1899, Germany bought the Northern Marianas together with the Carolines, Palau, and Yap from Spain. The main reason for the purchase certainly was the wish of the Imperial government to show off. Whatever economic, strategic or demographic hopes the people in charge might have had, they were proven vain rather soon as there never was a German immigration nor a garrison to speak of and the budget allotted to the colonies in Micronesia was always too small to allow for any major investments. However, the first and only governor (―Bezirksamtmann‖)8 of Saipan (responsible for the entire Northern Marianas), Georg Fritz (*1865 Alzey, ý1944 Alzey) was very active in promoting the cause of the Chamorros whose culture and language he described and protected against foreign influence. He remained in office in Saipan from 1899/ 1900– 1907.9 Of his writings, the ethnographic study Die Chamorro. Eine Geschichte und Ethnographie der Marianen ‗The Chamorro. A history and ethnography of the Marianas‘ (Fritz 1901) was translated into English in 1989 whereas most of his linguistically interesting books and articles are still awaiting re-edition, translation and critical appraisal. The first step in this direction has been taken. At the end of his career in the Marianas, Fritz (1907) published the Kurze Geschichte der Marianen ‗Short history of the Marianas‘ in Chamorro with a German interlinear version and free translation. This text is reproduced, modernised, analysed and commented upon in Stolz (2007). However, there are many 8

9

―District officer‖ is the more appropriate term in English. I use governor for Fritz‘s position because of his seemingly omnipotent position in the Marianas. In 1908/9, Fritz was ordered to fill in temporarily the vacant position in Ponape. In 1909/10, he was stationed in Yap but was still responsible for the Marianas. In 1911, he retired because of the failure of his policy in the Carolines where a rebellion against German rule broke out. Later on, Fritz became involved with the Nazi movement whose anti-Semitic ideology he shared (Hardach 1991: 80–82).

206

Thomas Stolz

more texts which can be credited to Fritz,10 namely a grammatical sketch (Fritz 1903), a dictionary (Fritz 1904),11 and the Chaifi legend in Chamorro (Fritz 1906) as well as some unpublished material12 and articles in response to the public insinuations of the German expatriate Costenoble. Apart from the problem that these texts require a good command of German from the reader, there is another purely technical problem which surely has prevented their re-edition, namely their inaccessibility. The journals in which the articles were printed for the first time have long ceased to exist and thus are hard to come by in libraries outside Germany. Nevertheless, I am convinced that it is worth the effort to prepare them for an annotated re-edition and translation. For the present purpose, a closer look at the grammatical sketch (Fritz 1903) must suffice.13 The text is 27 pages long. It appeared in the then well renowned Mitteilungen des Seminars für Orientalische Sprachen an der Friedrich-Wilhelms-Universität Berlin when Fritz had been in office in Saipan for a little over three years. On the very first page of the text, Fritz (1903, 1) mentions the sources on and in Chamorro he had access to, namely the three publications by Ibáñez del Cármen (1863, 1865a–b [cf. Zimmermann and Stolz this volume]).14 To what degree Fritz was able to liberate himself from his Spanish predecessor still needs to be determined.15 One thing he makes clear right at the outset: he wants to break with the Spanish past and thus rejects the established Spanish colonial orthography under the pretext that it is less suited than the German orthography to represent the sounds of Chamorro. However, he immediately hastens to add that special characters and diacritics are needed where Chamorro and German diverge too much phonologically (Fritz 1903: 2). In this way, Fritz‘s way of representing Chamorro sounds graphically is much less dependent than Chamisso‘s upon German writing traditions. The sketch is divided into ten sections starting with orthography and phonology (‗Schreibweise und Aussprache‘) and continuing with article and noun (‗Artikel und Substantiv‘), adjective (‗Adjectiv‘), numeral (‗Zahlwort‘), pronoun (‗Pronomen‘), interjection (‗Interjection‘), verb (‗Verb‘), adverb (‗Adverb‘), preposition (‗Präposition‘), and conjunction (‗Conjunction‘). The division follows 10

11 12

13

14

15

In addition to his work on and in Chamorro, Fritz (1911) also published a dictionary of Central Carolinian. The dictionary sold so well that a second edition appeared in 1905. In the Saipan archives, for instance, there is a 16-page manuscript of Fritz‘s entitled Chamorro Sprachübungsbuch which the Bezirksamtmann employed in the public school of Saipan where he himself taught (Hiery 2001). Moreover, the colonial historian Hermann Hiery (p.c.) claims that Georg Fritz also wrote a trilingual diary in Chamorro, Spanish and German. The manuscript is said to be in the possession of our colleague at the University of Bayreuth. For Costenoble‘s (1940) derisory remarks on Fritz‘s supposedly restricted mastery of Chamorro, cf. Stolz et al. (this volume). Fritz (1903: 2) also mentions in passing the existence of some governmental decrees of the Spanish period in Chamorro. This grammatical sketch appeared in print simultaneously with the grammar by the US governor of Guam, Safford (1903). Thus, any influence of one author on the other can be ruled out.

German and Dutch contributions to Chamorro studies

207

largely the established order of word-class distinctions in traditional German schoolgrammar. Syntax and other areas of the language are not dealt with. With almost nine printed pages (Fritz 1903: 12–20), the section on the verb is the most sizable. Interjections, adverbs, prepositions and conjunctions are presented in the form of simple lists. In other parts, relatively brief explanations accompany selections of examples with German translations. The terminology and categories Fritz employs are familiar from German philology. In the following paragraphs, I dwell a little on the most salient aspects of one of those sections of the Chamorro-Grammatik in which the author expresses his opinions in lieu of simply enumerating examples. There are not many sections of this kind, practically only the first two sections, namely the explanation of the spelling and pronunciation and the presentation of the categories of the noun. For reasons of space, I make do with a review of Fritz‘s section on orthography and pronunciation. Orthography/Pronunciation: The supposed superiority of the German orthography notwithstanding, Fritz feels the necessity to explain how the orthographic signs have to be pronounced. The first complication of the orthography arises from the distinction of three qualities of low vowels. Three graphemes are diacritically distinguished: 





, the corresponding sound is explained by Fritz as follows: ―nicht so nahe dem o wie das englische a, etwa wie in dem süddeutschen Beiwort halt (eben): galågo.‖ [not so close to the o as the English a, almost like the South German adverb halt (eben): galågo].16 Obviously, there is no close resemblance in standard German and Fritz refers to a foreign language (English) and regional varieties of German. Phonetically, the is meant to represent a slightly back low vowel. , which is said to be identical to German as in sÄsÄ (modern: sasa‘ ‗to spread apart‘). The German represents an obligatorily long low front vowel [æː]. Since vowel length is marked by a macron in Fritz‘s orthography (cf. below), the identity of the Chamorro sound and its German counterpart is not meant to hold for quantity. , which is described as ―ein sehr reines a, welches durch folgendes m oder n nicht getrübt werden darf: häme, hämio, länia.‖ [a very pure a which must not be obscured by a following m or n: häme, hämio, länia].17 The caveat is probably directed against pronunciation habits of German native speakers. The vowel could perhaps be understood as a central low vowel [a].

The distinction of several qualities of low vowels is a remarkable achievement because there is a phonemically relevant front-back distinction // ≠ // in Chamorro (Topping 1973: 19) which traditionally was reflected neither by the Spanish colonial orthography nor by any of the subsequent orthographies until the 1980s when the grapheme (at first for some time ) was established for the low back vowel (Underwood 1983: 16 17

The modern Chamorro rendering of the example is ga‘lågu ‗dog‘. The modern equivalents of the examples are ha‘me ‗warm‘, hamyo ‗you (plural absolutive)‘, laña‘ ‗expletive to express feelings ranging from mild surprise to complete disgust‘.

208

Thomas Stolz

Appendix E). Thus, Fritz‘s basic idea ultimately became real. However, his other proposals were less fortunate. First of all, the distinction of two further a-qualities raises problems. One problem is of a rather practical nature, as Fritz introduced two distinct signs which are hard to tell apart. There is with two superscript parallel strokes and ―dotted‖ with the typical German Umlaut-diaeresis. Even for a careful reader with schooling in German orthography, the differences are not salient enough to be noticed without the help of a looking-glass. What is more, there are additional superscript signs which mark different quantities of vowels. The macron on , , , , and indicates long quantities whereas the breve on , , , , and indicates short quantity.18 Vowel length is not phonemic in modern Chamorro (Topping 1973: 16–24)19 and there is no evidence that this was different at the turn of the 20th century, so the introduction of purely phonetic distinctions20 into the orthography appears uneconomic from a modern point of view.21 On closer inspection, the supposedly long mid-high vowels represented by and correspond to the closed high vowels /i/ and /u/ in modern Chamorro, respectively. Where Fritz writes hämē ‗we, us (emphatic exclusive)‘ the long final vowel is raised in modern hami; similarly Fritz‘s ūlō ‗head‘ corresponds to modern ulu. Fritz had to struggle with a problem that is still keeping phonologists of Chamorro busy: The phonetic output of the non-low vowel phonemes is also a matter of debate. For the most part they are predictable in native words: when stressed in closed syllables they are mid, in open syllables they are high (Cooreman 1987: 23). Fritz thus was entangled in the rather intricate phonology of vowel quality variation which depends not only on syllable structure and prosody but also partly on the etymology of the words under scrutiny, as Spanish loans do not always conform to the Austronesian-based system (Topping 1973: 57). To some extent, the short vowels marked by a breve in Fritz‘s orthography are also involved in the above problem.22 In words like ūdŏ ‗dumb‘ (Fritz 1903: 3), the final occurs in an open syllable just as the long in ūlō ‗head‘. In both cases, however, the modern versions of the words have final /u/: udu and ulu. At the same time, Fritz (1903: 2) mentions that the final position in (polysyllabic) Chamorro words is 18

19

20

21

22

It is understood that none of these diacritics has ever formed part of German orthography. The same negative statement applies to the distinction between and . Chamorro vowels may be phonetically long when they occur in open syllables bearing stress, for instance. Cf. [sa:gw] ‗spouse‘, ['so:b] ‗noodle soup‘ (Chung 1983: 36, Klein 2000: 83). Sandra Chung (p.c.) suggests that what Fritz took to be a quantity-based distinction is the tense-lax distinction which is in no way correlated to differences in length of the vocalic units. Of course, I do not intend to criticise anachronistically: Fritz wrote a quarter of a century before structural phonology developed and he was certainly not a phonologist avant la lettre. However, his native German is a language in which vowel length is distinctive and bears a heavy functional load. Thus, Fritz was somehow prepared to perceive different quantities and assign them a prominent status. Fritz does not illustrate the short vowels with examples from Chamorro in the phonology section.

German and Dutch contributions to Chamorro studies

209

prone to segmental reduction such that among others, is ―kaum hörbar‖ (‗hardly audible‘). Thus, the final short vowels marked by a breve reflect this optional reduction.23 In a small number of words, Fritz employs the breve for non-final vowels. In open syllables, the breve characterises stressed vowels immediately adjacent to the glottal fricative /h/ – be it to its left or to its right. Examples of this are hĭta ‗we, us (emphatic exclusive) and sĭha ‗they, them‘. However, this is not the only context in which the breve occurs. Fritz‘s example ŏdda ‗dirt‘ (modern: odda‘) shows us the way. Fritz knows that there are geminate consonants as he states that written double consonants have to be ―stets getrennt auszusprechen‖ (‗always pronounced separately‘) whereby he refers to the ambisyllabicity of the geminates. However, he does not always hear the geminates himself and thus transcribes some words as having a short vowel in a stressed open syllable as in i săkĕ ‗the thief‘ (modern: i sakke), i sădok ‗the river‘ (modern: i saddok), and yŏmok ‗fat‘ (modern: yommok).24 On the other hand, Fritz does not always use the breve even where he recognises geminate consonants as in hásso ‗to think‘ (modern: hasso) (Fritz 1903: 19). Unfortunately, Fritz is not particularly careful when it comes to employing his own conventions such that in a good part of the Chamorro-Grammatik the diacritics introduced in the introductory section become ever rarer (without disappearing altogether). The common article (modern: i) is introduced with a breve i.e. as ĭ (Fritz 1903: 3). However, this extremely frequent functional element is consistently written without the breve in the remainder of the ChamorroGrammatik. The proper article (modern: si) occurs more often with the breve as in sĭ Pedro estába guíne ‗Pedro was here‘ (Fritz 1903: 4) but the breve is occasionally missing too as in si júos gaige gi tautau ‗God stands above humans‘.25 The acute accent26 indicating stress site is relatively consistently employed but, again, with an occasional exception as in i kapitat Alemania ‗the capital of Germany‘ (Fritz 1903: 20) with no accents at all as opposed to i kapitá(t) Alemania (Fritz 1903: 4) with at least one stress site identified. This inconsistency problem is aggravated because from Fritz‘s admittedly rather short description, the reader expects that each vowel sign in the transcriptions hosts at least one superscript diacritic. However, this expectation is not fulfilled by Fritz‘s own representation of Chamorro words. First of all, there are a majority of vowel signs with23

24

25

26

Cf. the alternation of final breve-ă and unmarked in i hāgă ‗the daughter‘ and i hāga sĭha ‗the daughters‘ (Fritz 1903: 5). Fritz‘s failure to recognise the geminates is perhaps explicable as influence from his native German where some kind of ambisyllabic geminates seem to exist (the issue is controversial [Caratini 2007]), without being phonetically easily discernible (Barry et al. 1999). These geminates normally go hand in hand with a preceding stressed short vowel as in Koffer ‗suitcase‘ [().] with phonetically optional syllable coda after the stressed vowel. The modern versions of the sentences are: Si Pedro estaba guini and Si Yu‘os gaige gi taotao. The latter is literally ‗God exists in/at/on man.‘ The final words of the section under review are dedicated to accent signs: ―Die Betonung wurde durch Accente bezeichnet.‖ (‗The stress site was indicated by accents.‘).

210

Thomas Stolz

out any diacritics such as gualo ‗field‘27 (modern: gualo‘ ‗farm; toiled land‘) and many more. Nowhere in the section on orthography and phonology however, is there any indication of how vowel letters without diacritics have to be pronounced. Perhaps, Fritz assumed that wherever he did not make any special comments things had to be like in German. However, this assumption is not borne out by the other examples given by Fritz (1903: 2) because the digraphs and are explicitly characterised as ―Doppellaut wie im Deutschen‖ (‗diphthongs like in German‘),28 meaning: if the German pronunciation for certain orthographic solutions is presupposed, there would be no need to mention the digraphs at all. Be that as it may, the occurrence of adds a fourth low vowel beside , , and which together with the low vowels with length distinctions and yield a sizable inventory of six distinct signs. Apparently, signs of low vowels which bear a macron or a breve cannot be combined with any of the other superscripts (except the acute which marks stress site). Does this justify the conclusion that , , and are neutral as to quantity or always of the same quantity? In any case, it remains unclear which quality the specifically marked long/short low vowels supposedly have – back, central or front. The German examples given probably only illustrate length distinction.29 Outside the realm of vowels, the first consonant Fritz (1903: 2) illustrates is the labio-velar // written as is also the modern practice. This case is telling. Fritz‘s example is guine ‗here‘ (modern: guini). The reader is told that has to be pronounced ―wie gw‖ (―like gw‖) and thus the sample word is re-transcribed as gw ně. Fritz could have opted for a digraph * in lieu of such that the explanation would have been superfluous. However, at the end of the section under scrutiny Fritz states that – among other graphemes (or phon[em]es?) – ―w…giebt es nicht‖ (―there is no w‖). This is true of course if * corresponds phonetically to German which represents the voiced labio-dental fricative [v] which is indeed absent from modern Chamorro. However, the w-symbol used in the re-transcription of Fritz‘s guine cannot represent the German fricative. It must be either a bilabial approximant /w/ or the strong labialisation of the preceding obstruent. Topping (1973: 34 and 37) demonstrates that this /w/ also occurs in combination with other consonants (as in puengi ‗night‘, etc.) and could best be analysed as forming a complex single phonological unit with the preceding consonant (thus yielding a smallish set of labialised stops and nasals). This wider distribution must have escaped Fritz‘s notice. Moreover, his ―rule‖ that is pro27

28 29

This form contrasts with guālo ‗toiled land‘ (Fritz 1903: 3). This would yield a minimal pair (in fictitious modern spelling) guālo‘ ≠ gualo‘ with related but nevertheless different meanings. In the modern dictionary, there is however only one polysemous entry gualo‘ (Topping et al. 1975: 82) which combines all the meanings which are distributed over two supposedly distinct words in Fritz‘s Chamorro-Grammatik. Neither the Chamorro examples nor the German examples illustrate the digraphs/diphthongs. is said to be pronounced like the low vowel in German Haar, for , the German example is lachen.

German and Dutch contributions to Chamorro studies

211

nounced like gw is imprecise because it does not comprise the crucial condition under which it applies, namely the presence of a vowel immediately to the right of . Otherwise words like guma ‗house‘ (modern: guma‘) would lose their stressed vowel and start with inadmissible consonant clusters. The next consonant, , is said indirectly to be always a voiced plosive, the glottal fricative is always ―scharf gehaucht‖ (―sharply breathed‖). For these two speech sounds, only German examples are given because they are depicted as sounding just like in German (cf. above). More interesting remarks occur in connection with . According to Fritz, has to be pronounced ―sehr weich, hält die Mitte zwischen dem deutschen und dem französischen j (etwa dj)―.30 This description is phonetically inaccurate as the German represents the palatal approximant [j] whereas French is the grapheme that stands for the voiced post-alveolar fricative (sibilant) []. It is difficult to locate the actual range of Chamorro pronunciations of (modern: ) ―in the middle‖ between the two European comparees.31 The approximate identification of with the pronunciation dj is indicative of the presence of a plosive element which is absent from both the German and the French phonemes. Guessing from the ancillary transcription dj, one is tempted to equate Fritz‘s with either the voiced palatal plosive [] or the palatalised voiced dento-alveolar plosive []. One must read the -j in dj as connected to French [] to get to the palato-alveolar affricate [] which, in modern Chamorro, competes with the post-alveolar affricate [] (Topping 1973: 30–31). Since all these affricates are foreign to German (including its regional varieties) and French,32 Fritz had to circumscribe in lieu of explaining. There are two lines in the text on the velar nasal. Fritz introduces the digraph (= modern ) whose pronunciation is illustrated by German examples only (namely singen and Hang). From those, it is clear that the velar nasal [] is meant. The next line is dedicated to the more formidable trigraph – again with only a German example. The German adjectival compound strenggläubig contains the sequence which phonologically is cut in two by the syllable boundary such that [] is the coda of the first and [] the onset of the following syllable. It is not absolutely clear to me why Fritz felt compelled to mention at all. The (pre-glottal) voiceless plosives are represented as expected by

, and , respectively. No examples are given – not even from German as the typical German pronunciation of these phonemes is strongly aspirated.33 To Fritz‘s ear, the Chamorro 30

31 32

33

[very soft, keeps in the middle between the German and the French j (approximately dj)]. Fritz repeatedly refers to French in his description of Chamorro sounds because at the beginning of the 20th century, French was still widely known among the élite. The best candidate for this in-between position would be [] – the voiced palatal fricative. Fritz also knew Spanish very well as he had worked for some time in South America before he became a civil servant. Fritz (1903: 2) says ―nicht das deutsche, durch ein nachgehauchtes h verhärtete p, t, k‖ (‗not the German p, t, k which are made harder by a following breathed h‘).

212

Thomas Stolz

pronunciation comes closer to the French pronunciation or that of speakers of the Saxonian dialect in Germany. In both cases, the plosive is unaspirated. Next in line is the trigraph (modern: ) which is said to be similar to German . The equivalence is by no means absolute. What is written in German has more of a sequence [] + [] because the voiceless denti-alveolar plosive and the voiceless palato-alveolar fricative are heterosyllabic if in intervocalic position. They do not form a phonological unit. Fritz himself is not content with the above equation because he adds in brackets that what he means by represents ―ch in ich mit vorgesetztem t‖ (‗ch in ich with a pre-posed t‘). This boils down to a combination of the voiceless dento-alveolar plosive with the voiceless palatal fricative []. This comes relatively close to [] which is one of the fairly wide range of variants of the voiceless palato-alveolar affricate: [] ~ [] ~ []. In square brackets, Fritz comments on a problem arising from the introduction of . The ethnonym/glossonym Chamorro would have to be written Tchamorro which however, would violate the long established local traditions which he prefers to respect. The rhotic consonant written is identified (again without examples) as a phonological Hispanism as it occurs only in Spanish loanwords where it is pronounced ―ganz weich‖ (‗very softly‘). What is meant exactly by this description is far from clear. The supposed ―softness‖ might result from the absence of multiply trilled rhotics (the long of Spanish, for instance). This suggests that the rhotic was pronounced more like a tap than like a trill. The next paragraph in Fritz‘s account is dedicated to those elements which are weakened in pronunciation such that they are almost inaudible. Reduced audibility is indicated by brackets around the grapheme representing the weakened segment. Fritz notices a tendency towards weakening with one vowel and three consonants, namely (e), (g), (k), and (t). Weakening happens most often ―am Ende‖ (‗at the end‘). It is not unlikely that Fritz refers to the absolute word-final position. On the other hand, of the six examples he adduces, four illustrate word-internal, syllable-final weakening. All examples are of (t) about which Fritz says that it often corresponds to the Spanish rhotic or lateral liquids. Accordingly, five examples are of Spanish origin as e.g. o(t)den ‗command‘ (modern: otden) < Spanish orden and ánhe(t) ‗angel‘ (modern: anghet) < Spanish angel. In the main body of the Chamorro-Grammatik, bracketed signs are only sparingly employed, however. In the end, it does not strike me as particularly helpful for the feasibility of orthography to indicate the optional reduction of segments (cf. below). The glottal plosive // is represented by a grapheme of its own, namely the apostrophe . Fritz has no name for this consonant. What is more, does not represent a full-blown consonant but serves as a kind of diaraesis marker. In his own words Fritz explains that ―bedeutet, dass der vor und nach dem Apostroph stehende Laut nicht verschmolzen werden dürfen‖ [means that the sounds which stand before and after the apostrophe must not be coalesced]. His Chamorro example is l ‘ě ‗to see‘ (modern li‘e‘) of which he says that it has to be pronounced like li-e and never lije i.e. the vowels are

German and Dutch contributions to Chamorro studies

213

separated by a hiatus which is not to be filled by a glide. For Fritz, there is a break between the vowels, not a segment. That is why he continues the description by stating that even in the absence of , ―Laute und Wörter scharf abzusetzen und nicht zu verschleifen sind‖ [sounds and words are to be kept strictly apart and must not be coalesced].34 Fritz considers this a typical trait of Chamorro. His failure to recognise the glottal plosive as a distinct consonant is unsurprising as for a German native speaker the glottal plosive is an integrated component of syllable-initial vowels.35 However, in the remainder of his grammatical sketch, there is occasionally evidence of words whose transcription involves a post-vocalic which precedes a consonant as in ō‘son ‗tired, bored‘ (modern: o‘son) (Fritz 1903, 4).36 However, in most of the cases of syllable-final //, Fritz does not recognise the glottal plosive, cf. tchāgo ‗far‘ (modern: chago‘) where word-final is missing. On the next two lines, Fritz tells the reader that , , , , and are absent from Chamorro. The signs he refers to are graphemes of the German alphabet. The same holds for the four units , , and mentioned in the next line. In contrast to the former, however, the latter four are considered ―überflüssig‖ (‗superfluous‘). This differential treatment of the two groups of elements might be understood as reflecting their differences in status: , , , are just ―letters‖ of the German alphabet whereas , , , , , are letters which represent distinct speech sounds. Besides the problematic discussed above, the exclusion of is somewhat surprising because it represents the voiceless dento-alveolar affricate [] which happens to be the closest German equivalent of the Chamorro palato-alveolar affricate // as it corresponds to one of its allophonic realisations. What can be the appropriate judgement? To my mind, Fritz involuntarily mixed academic interests with those of practical language teaching and learning. Going by the title given to the section in Fritz‘s Chamorro-Grammatik, the goal should have been to provide Chamorro with as good an orthography as possible and explain the graphemes to the user in a way that allows speedy comprehension. The plot is spoiled however, because, as an orthography, Fritz‘s spelling rules are too impractical. The main disadvantage is the co-existence of obligatory and optional graphemes. The system is overburdened with allographs: the breve for instance is by no means compulsory and may disappear without explanation. Moreover, the identification of weakened segments is interesting from the point of view of phonetics and phonology but not for the average language learner. In addition, the many diacritics are no longer useful when they must be stacked (for instance in the case of stressed long or breve vowels). Digraphs and 34

35

36

This rule creates problems with the interpretation of digraphs/polygraphs in Fritz‘s system, for instance with the diphthongs Whether or not the German glottal plosive is phonemic is a controversial issue in the literature (Maas 1999: 224–225). Elsewhere (Fritz 1903: 17), it erroneously has glottal plosives in tcho‘tcho‘ ‗to eat‘ (modern: chocho) which in this way becomes the same as tcho‘tcho‘ ‗to work‘ (modern: cho‘cho‘).

214

Thomas Stolz

trigraphs could have been avoided at least in some cases (for instance ). For practical usage, the orthography is too exact whereas, for the purposes of descriptive linguistics, there is too little explanatory text. However, it would be unduly unfair to blame Fritz for his layman‘s handling of the admittedly intricate task of creating an orthography which not only is in line with the phonology of the language but also easy to learn. This task is always rather demanding and it was further complicated by the politically inspired desire to distance the ―German orthography‖ as far as possible from the earlier ―Spanish orthography‖. To the credit of Fritz, it must be said that he makes some correct and interesting observations in terms of phonology. His orthographic solutions, however, are far from being optimal. On the other hand, the Chamorro-Grammatik was a first for Fritz. Given that he had no special schooling in descriptive and field linguistics, what he did was daring. With a view to evaluating his achievements conclusively, future studies have to check whether the decisions taken in the Chamorro-Grammatik remained in force also in the later work of Fritz dedicated to Chamorro.

2.3. Father Callistus37 Father Callistus Lopinot (*Geipoldsheim/Alsace 1876, ýRome 1966) was a member of the Capuchin Order (Rhenish-Westphalian Province). He was appointed parish priest to Saipan in 1907 where he stayed only until 1909 when he was sent to Palau and later on to a variety of other locations. His dealings with governor Fritz proved to be rather problematic such that Fritz forestalled Lopinot‘s transfer to Guam in 1910. It is assumed that Fritz gave word to the governor of Guam that Lopinot was not a trustworthy person (Hardach 1990: 181–197). This animosity between the two characters explains to some extent why the Capuchin cleric passes silently over Fritz‘s work on Chamorro in his own Chamorro-Wörterbuch (Lopinot 1910c). What is more, Lopinot deliberately deviates from the orthographic conventions favoured by his adversary. Before I look more closely at this issue, it should not go unmentioned that Lopinot is the (co-)author of at least seven books and booklets with religious contents all of them written in Chamorro (Fischer & Lopinot 1910a–c, Lopinot 1910a–b, Lopinot & Schuster 1909) which are awaiting their annotated re-edition (Hezel 2001: 567). Lopinot‘s major contribution to the description of Chamorro is his ChamorroWörterbuch of 1910. This is not a dictionary alone as it also contains a grammatical sketch (= appendix 1: 28 pages)38 and a number of so-called exercises (―Sprachübungen‖ = appendix 2: 5 pages). The dictionary consists of a German-Chamorro part (86 37

38

Father Callistus signs his publications only with his clerical name Callistus. I refer to him with his secular last name Lopinot in this contribution even though this name is not mentioned on the cover pages of most of his publications. Without looking at the details, Lopinot‘s grammatical sketch can be judged superior to Fritz‘s.

German and Dutch contributions to Chamorro studies

215

pages) and the corresponding Chamorro-German part (85 pages). Contrary to what Spennemann (2004, #1005) says about the lack of references to other sources39 in Lopinot‘s dictionary, the author states in the foreword (Lopinot 1910c: i): Es ist dann in Anhang I eine Chamorro-Grammatik beigefügt, besonders deshalb, weil eine richtige Abhandlung des Chamorro-Verbums in deutscher Sprache bisher noch nicht erschienen ist. Die Grammatik stützt sich auf ein Manuscript von Msgr. Jose Palomo; auch ist die Arbeit von Saffort [sic!] im American Anthropologist benützt worden [Then there is, in appendix I, a Chamorro grammar; I attach it especially because a full-blown treatment of the Chamorro verb in German has not been published yet. The grammar is based on a manuscript by Monsignore Jose Palomo; in addition, the study by Safford an American Anthropologist has been consulted]. Since Safford (1903) refers to Jose Palomo as his major informant, the two grammars – Lopinot‘s and Safford‘s – are candidates for a future detailed comparison. It is indicative of the unfriendly relations between cleric and governor that Lopinot refers to all other sources except those written by Fritz. This telling silence makes the ChamorroWörterbuch even more interesting because of its competition with the ChamorroWörterbuch by Fritz (1904) published only six years earlier. The detailed comparison of the two dictionaries however, has to be postponed until a separate publication. According to the ―Vorwort‖ (Lopinot 1910c: i), the dictionary‘s purpose is predominantly of a practical nature. It is meant to allow Europeans to learn Chamorro. In addition, it provides Mariana Islanders40 with a practical tool to learn German. This is the reason why the Chamorro language had to be represented in its actual spoken form which includes the most widely used Hispanisms. In a purely academic dictionary, the Hispanisms would not be taken into consideration.41 On page II of his ChamorroWörterbuch, Lopinot very briefly introduces the ―Schreibweise‖ (‗spelling‘) used in the dictionary. In contrast to Fritz (cf. above), Lopinot pays his respects to the Spanish past when he says: Weil die spanische Sprache ganz mit dem Chamorro verwachsen, wurde inbezug auf die Schreibart teilweise an der Tradition festgehalten und ist in spanischen Wörtern j = h zu lesen (z.B. consejo) [Owing to the fact that the Spanish language is completely intertwined with Chamorro, the traditions as to the

39 40 41

Spennemann‘s statement is correct as far as a formal bibliography or list of references is concerned. ―[D]ie Eingeborenen der deutschen Inseln‖ [the indigenous people of the German islands]. Note however that Lopinot‘s dictionary contains a large number of outdated Chamorro terms ―die nicht mehr gesprochen oder sogar nicht mehr verstanden werden‖ [which are no longer uttered and even no longer understood].

216

Thomas Stolz

spelling have been respected at least partially and in Spanish words one should read j = h (e.g. consejo)].42 (Lopinot 1910c: ii) This is an important decision whereby the layering of the Chamorro lexicon is retained visibly with orthographic means, i.e. words of Spanish origin which contain a jota in their correct Spanish form keep the also in Chamorro. Since in Austronesian words, the glottal fricative is represented by in Lopinot‘s orthography, the division of the Chamorro lexicon into a Spanish part and a Chamorro part is obvious. For this phoneme, Lopinot (1910c: ii) states that it is aspiriert; in einigen Fällen, besonders am Ende einer Silbe tritt die Aspiration recht deutlich hervor; z.B. sahyau [aspirated; in some cases, especially at the end of a syllable, the aspiration is relatively clearly prominent; e.g. sahyau]43 Note first that, in contrast to Fritz,44 Lopinot is aware of the presence of // in positions other than syllable onset. There are no grounds to assume that and were pronounced differently in early-20th century Chamorro. I take it that both graphemes represented the glottal fricative [] (perhaps with allophonic variants in syllable-final position). Thus, Lopinot sacrifices homogeneity to the etymological principle of treating Spanish-derived elements differently from autochthonous etyma. This is also the reason why Lopinot re-introduces the typically Spanish allographs to represent the voiceless velar plosive //: is employed in front of the high vowels represented by and whereas is written elsewhere. Another orthographic Hispanism is the grapheme to represent the palatal nasal //. The pronunciation of is given as similar to German i.e. a palato-alveolar affricate // is the intended item. These Spanish-inspired spelling rules however, apply to the entire lexicon – and not only to the Spanish part thereof. Somewhat surprisingly, Lopinot deviates further from etymological principles when he declares that certain Spanish graphemes, digraphs and combinations are not simply transferred into Chamorro. This affects Spanish which becomes Chamorro , Spanish which yields in Chamorro, the Spanish sequences / which are rendered as in Chamorro and the Spanish digraph to which Chamorro corresponds. In the absence of any explanation for this differential treatment of Spanish graphemes (and the associated speech sounds), it is too hard to guess why the jota is kept but the zeta is not, for instance. I hypothesise that the variety of Spanish that is responsible for the Hispanisms in Chamorro resembles Philippine Spanish because this is the region with which the Marianas had the closest

42

43 44

The sample word would be *konsehu in modern Chamorro. However, Topping et al. (1975: 112) only have konseha ‗to advise‘. Rodríguez-Ponga (1995: 399) reports konséhu ‗council, advice‘ in his inventory of lexical Hispanisms in modern Chamorro. The modern form of sahyau is sahyao ‗quick, fast‘. In Fritz (1903), not a single syllable-final glottal fricative is acknowledged.

German and Dutch contributions to Chamorro studies

217

contact during the Spanish dominion over the islands. Betacism,45 seseo46 and the development of the palatal lateral // into the palatal glide/approximant // are typical of many non-Castilian varieties of Spanish, for instance Philippine Spanish (Quilis 1992: 124–133). Chances are that , and were felt to be purely orthographic complications without distinct phonetic output. In contrast, Spanish was perceived as preserving its distinct pronunciation47 and thus deserved a grapheme of its own. However, this does not explain why Lopinot chose to represent the same speech sound by a different grapheme in Austronesian words. As to the Spanish sequences /, their representation by Chamorro reflects a regular sound change to which Hispanisms equipped with liquids are subject under Chamorrisation, namely: LIQUID > PLOSIVE[+dental,–voice] /__$ i.e. Spanish syllable-final liquids are represented by voiceless dental plosives. Thus, the process is not dependent upon the presence of an adjacent //. In the dictionary, Lopinot often observes the rules, as in ayunat ‗fast‘ (modern: ayunat) < Spanish ayunar, but not always, as there are violations of the rule like atbol ‗tree‘ (modern: atbot) < Spanish arbol (Lopinot 1910c: 94–95). Lopinot provides some additional information in connection to the graphemes he uses. For he specifies that it is ―ein eigener Nasallaut und viel näselnder als unser ng‖ [a nasal sound of its own and much more nasalised then our ng] (Lopinot 1910c: ii). There is also the special character which is ―ähnlich dem englischen a in wall, doch nicht ganz so tief‖ [similar to the a in English wall, but not so low]. This sign thus corresponds to the introduced by Fritz (cf. above). For the mid-high and high vowels, Lopinot observes that ―e und i, o und u können in sehr vielen Fällen verwechselt werden z.B. déquiqui, díquiqui; tútuhun, tútuhon etc. …‖ [in very many cases, e and i, o and u can replace each other, e.g. déquiqui, díquiqui; tútuhun, tútuhon etc. …].48 The German verb verwechseln normally has the meaning ‗to mix up‘; it is also possible to use it with a less negative (albeit non-standard) connotation, namely ‗to exchange, to substitute‘. Both readings are possible here. In the final sentence of this sketchy overview, Lopinot states that ―[i]m übrigen ist die deutsche Schreibweise zu Grunde gelegt.‖ [as for the rest, the German spelling is the basis]. To sum up, Lopinot develops an orthography which combines elements of the Spanish colonial system with new ideas (some of which resemble solutions put forward by Fritz [1903]) and the rules of German orthography. With the Spanish part, the combination contains one ingredient more than Fritz‘s orthography. It seems doubtful that this combination yields good results. Nevertheless, Lopinot‘s proposal is superior to Fritz‘s at least in one aspect: there are 45

46

47

48

Basically, the term ‗betacism‘ refers to the generalisation of the plosive allophone [] of // to the detriment of the fricative allophones [] ~ []. In Spanish dialectology, seseo describes the merger of the voiceless interdental fricative /θ/ with the voiceless dento-alveolar sibilant //. With the proviso, that in non-metropolitan varieties of Spanish, the Castilian voiceless fricative // tends to be turned into the less prominent glottal fricative //. The examples correspond to modern díkike‘ ‗small‘ and tutuhon ‗start‘.

218

Thomas Stolz

not as many diacritics in Lopinot‘s spelling system and he does not deem it necessary to mark phonetically reduced segments. Clearly, 18 lines cannot do justice to the intricacies of Chamorro phonology and the ways and means to represent it orthographically. Lopinot is rather parsimonious with his explanations and illustrative examples are scarce, too. His proposal is also heterogeneous. If the Spanish allographs ~ for // are transferred to Chamorro, what happens to the parallel case ~ for //? A glance at the letter G in the Chamorro-German part of the dictionary (Lopinot 1910c: 121–125) reveals that Lopinot must have been undecided as to this issue because he oscillates between gia ~ guia ‗to guide‘ (modern: giha) < Spanish guía ‗guide‘, for instance. Moreover, in guëco ‗hollow‘ (modern: gueku) < Spanish hueco, Lopinot uses the diaresis without prior instruction on its use (stress site outside the penultimate is indicated by the acute accent). In another matter, Lopinot is clearly lagging behind Fritz. The cleric does not notice the glottal plosive // such that ‗to work‘ (modern: cho‘cho‘) and ‗to eat‘ (modern: chocho) are the same for him, namely chocho (Lopinot 1910c: 108), in intervocalic position, the glottal plosive is not represented either: lii ~ lie ‗to see‘ (modern: li‘e‘) (Lopinot 1910c: 134). Like his compatriot, Lopinot fails to recognise the consonantal geminates: deca ‗to present something with a stick‘ (modern: dekka‘ ‗to poke‘) (Lopinot 1910c: 110). To sum up, the two German sources discussed in the previous paragraphs have not succeeded in solving the riddles posed by Chamorro phonology. The personal antipathy between the two protagonists brought forth two competing German orthographies which coexisted for some time with the traditional Spanish orthography in the Northern Marianas and with the American orthography on Guam. Neither Fritz‘s nor Lopinot‘s spelling systems are satisfactory such that they can be employed successfully in the schools. Phonologically, Fritz‘s orthography is more interesting because its inventor tries hard to represent the sounds he heard in a consistent way. This strife for precision leads to an orthography which is hard to handle in everyday practice. Lopinot, on the other hand, ignores the phonetic subtleties, which is fine if one aims at creating a feasible spelling system. However, he spoils everything by way of lumping together Spanish, German and newly created conventions. Since Fritz dropped out from the colonial service in 1911 and afterwards turned into a vociferous adversary to the idea of regaining the overseas German colonies,49 the use of Fritz‘s orthography was discontinued very soon after its invention. Lopinot however, employed his orthography in a variety of Chamorro publications until the 1st World War put an end to the German period in the Marianas.

49

Fritz embraced the Nazi ideology and thus supported the idea of gaining new territories in the European East for the (seemingly endlessly) growing German population.

German and Dutch contributions to Chamorro studies

219

3. Het Tjamoro The Dutch grammar of Chamorro entitled Het Tjamoro van Guam en Saipan vergeleken met eenige verwante talen [The Chamorro of Guam and Saipan compared to some related languages] is a very peculiar book. There are several aspects which make Het Tjamoro special. First of all, the author of Het Tjamoro, Jacob Kats (1875–1945),50 is responsible for only 40% of the book because about 190 pages (including the preface) of 313 pages or 60% of the entire text go to the credit of Johann Christoph Gerhard Jonker (1857–1919).51 How is this hidden co-authorship of Johann C. G. Jonker to be explained? In 1913, Kats submitted a manuscript version of his part of the book to the internationally renowned Koninklijk Instituut voor de Taal-, Land- en Volkenkunde van Nederlandsch-Indie [Royal Institute for the Linguistics, Geography and Ethnology of the Dutch Indies] whose scientific board did not unanimously approve of the quality of this scholarly monograph. In the ―Vorrede‖ [foreword] signed in 1915, Jonker thus felt obliged to defend the original manuscript as much as possible against the objections raised by the board members. No matter how hard he tried there remained enough flaws to call for a thorough revision. Jonker himself revised the manuscript52 and this led him to write the extensive part of ―Opmerkingen‖ [Notes] of which there are altogether 284, i.e. more than 2 per printed page of the original text. In these notes, Jonker very often corrects the opinions, analysis and data presented by Kats. The scientific board of the Koninklijk Instituut criticised Kats‘s study not only because of its lack of originality, its repetitive style and superfluous (qua well-known) examples from languages other than Chamorro but also because Kats ignored some other publications dealing with Chamorro.53 Indeed, Kats refers exclusively to Safford (1903) and Fritz (1904), all other publications – including Safford (1905), the complete separate printing of Safford‘s work (1909), Fritz‘s grammar (1903) and subsequent papers by Fritz as well as Lopinot (1910c). Of the Spanish contributions, Kats is aware only of Ibáñez del Cármen (1887) 50

51

52

53

Jacob Kats worked as a school teacher and headmaster in the Dutch East Indies from 1897 onwards and became a renowned scholar of Javanese literature and drama. I owe this information to Matthew Cohen (p.c.). Johann C. G. Jonker was a leading expert of the languages of the Dutch East Indies and of Burmese. He authored numerous grammars, dictionaries, anthologies and scholarly articles dedicated to Javanese, Timorese, and other languages of the Indonesian archipelago. ―Aan den ondergeteekende werd opgedragen deze verkortingen aan te brengen en tevens enkele «Opmerkingen», welke van het begin af noodzakelijk werden geacht, aan het werk toe te voegen. Evenwel deze opdracht bleek bij de uitvoering niet zoo eenvoudig te zijn.‖ [The signed writer was given the task to execute these cuts and even add to the work some notes which were considered necessary right from the start. However this assignment proved not so easy when it came to putting it into practice] (Kats 1917: v). Jonker‘s revision notwithstanding, some of Kats‘s mistakes have passed unnoticed such as the year 1768 given as the beginning of the Spanish settlement in lieu of the correct 1668 (Kats 1917: 3). The board members especially disapproved of Kats‘s passing silently over Conant (1911).

220

Thomas Stolz

without making use thereof.54 Jonker (Kats 1917: viii) downplays the importance of such gaps when he admits that Kats ignored the existence of Fritz (1903) but adds that ―veel heft dit evenwel niet geschaad‖ [it did not do much harm, anyway]. However, in the ―Opmerkingen‖, Jonker refers to Fritz (1903) more often than not, such that one gets the impression that Kats could have avoided many mistakes if he had consulted the Chamorro-Grammatik.55 Furthermore, my antiquarian copy of Kats (1917) bears the all too visible marks of a very thorough previous reader who comments upon practically each and every paragraph with handwritten marginal notes in four different colours.56 This reader shows his or her strongly felt dissatisfaction with Kats‘s exposition and line of argumentation on many occasions. The disagreement with the author of Het Tjamoro is such that Kats is accused of plagiarism and some of the notes give evidence of foul language (especially in the notes section where Jonker reveals how many amendments Kats‘s original needed). In contrast to Fritz (1903, 1904) and Lopinot (1910c)57, Kats‘s goals are purely academic as he wants to contribute something substantial to the comparative linguistics of Austronesian. Thus, the book is replete with parallels found in up to 21 other Austronesian languages from Malagasy to Samoan. Kats (1917: 4–5) complains about two things. First of all, his failure to get access to any sizable Chamorro text made it difficult for him to check the validity of the information provided in his only two sources. Secondly, Kats blames both Safford (1903) and Fritz (1904) for their inaccurate, heterogeneous and careless transcription of Chamorro.58 Everything that Kats writes about Chamorro is based on the comparison of Safford‘s and Fritz‘s data and analyses although he is sometimes rather critical of both of his sources and even puts forward views different from theirs. In agreement with the previous two sub-sections, I give an overview of Kats‘s treatment of Chamorro phonology. With almost 25 full pages, his chapter on ―Klankstelsel‖ [Sound system] is already relatively long (Kats 1917: 5–30). To this chapter of the original, one has to add the 56 pages of notes (Kats 1917: 128–184) by Jonker referring to phonological issues. On closer scrutiny, this seemingly rich description of Chamorro phonology diminishes 54

55 56

57

58

Jonker (Kats 1917: 127) notes that Kats did not quote this title bibliographically correctly because he left out at least eight words. Jonker did not know Lopinot (1910c), nor the minor papers by Fritz. The comments are mainly in Dutch with the occasional German phrases and some expressions in English. The previous reader himself/herself probably adhered to omnicomparativism i.e. s/he searched for global etymologies making use of incidental similarities of roots in languages so disparate as Latin and Chamorro because Chamorro saina ‗Lord, father, parents‘ seems to resemble Latin senex ‗old‘ (= handwritten note on page 66). Kats (1917) is very critical of Fritz (1904). However he grants that the Chamorro-Wörterbuch is good enough for practical purposes. Already at this point, the angry previous reader scribbled at the top of page 5 that ―[z]onder G. Fritz zou Kats nog geen 60 blz. [blaadzijden] vol hebben gekregen.‖ [Without G. Fritz, Kats would not have filled more than 60 pages]. One is tempted to add that without Safford, not even these 60 imaginary pages would have ever been written.

German and Dutch contributions to Chamorro studies

221

considerably because the bulk of the data, analyses and arguments belong to comparative Austronesian studies. Only four pages of the original (Kats 1917: 5–9) and seven of the notes (Kats 1917: 128–135) are relevant for synchronic Chamorro phonology. Since Kats is completely dependent upon his predecessors, he enters the realm of Chamorro phonology by way of comparing Safford‘s and Fritz‘s orthographic conventions and their explanations of Chamorro speech sounds which Kats sometimes considers unclear. Thus, he employs a third transcription system (with some idiosyncratic additions), namely one that is widely used in Dutch Austronesianist studies of his time. Kats (1917: 5–6) claims that Fritz (1904) does not define the grapheme phonologically – an assumption which is corrected by Jonker (Kats 1917: 128) who refers to Fritz (1903: 2) where is described as a low back vowel (cf. above). Since Kats did not know of Fritz (1903), he also ignored Fritz‘s phonetic illustration of other graphemes such as , , etc. (cf. the corrections by Jonker [Kats 1917: 128–129]). In point of fact, Kats refrains from making any definitive statement about the phonetic value of the signs he uses to represent Chamorro sounds. In a way, he lets the graphemes ―speak for themselves‖. In connection to the discussion of the correct phonetic interpretation of Safford‘s grapheme , Kats quotes Safford‘s opinion according to which should be pronounced like the English voiced palato-alveolar affricate [] although he considers this a ―corruption‖ via Spanish influence of an erstwhile palatal approximant []. Kats (1917: 6) is apparently uncommitted in this matter because ―[h]oe het zij, ik geef hier dit teeken weer door j‖ (―be that as it may, I represent this sign here by j‖). His choice to replace Safford‘s by is perhaps interesting as Kats‘s digraph is suggestive of a place of articulation different from the one associated with . The latter is described (following Safford [1903] again) as corresponding to the in English church i.e. it is a voiceless palato-alveolar affricate //.59 Kats‘s however, invites two rather different interpretations. It could be taken to represent the voiceless palatal plosive // or the palatalised voiceless dento-alveolar plosive //. Neither of those counts among the known realisations of the unit under scrutiny, though. Why his transcriptions are better than those of his predecessors, Kats does not say. He simply makes do with stating what signs he employs. The only synchronically interesting paragraph in Kats‘s phonology section treats of the glottal plosive // (Kats 1917: 6). Kats notices that ―[b]ovendien geeft Safford nog een «guttural» klank aan door het teken ^‖ [in addition, Safford assumes a guttural sound marked by the sign ^]. From the remainder of this paragraph, it seems that Kats accepts 59

Kats (1917: 6) also quotes Safford (1903) on the supposed similarity of the pronunciation of Chamorro with the pronunciation of German . This equivalence is of course wrong as the German trigraph represents the voiceless palatal fricative []. This error results from a misunderstanding of the segmental structure of the German sample word Rutscher [.] which contains a heterosyllabic sequence of the voiceless dento-alveolar plosive // in the coda of the initial syllable and the adjacent onset of the final syllable filled by //. In allegro speech, the two segments coalesce such that they may yield optionally a phonetic affricate [].

222

Thomas Stolz

Safford‘s category of ―guttural vowels‖ i.e. the glottal plosive is not assigned segmental status. However, Kats also notices that the vowels are guttural sometimes where a former post-vocalic consonant was lost. Jonker (Kats 1917: 129) supports this view with comparative evidence from other Austronesian languages. Jonker also wonders why a word like guma ‗house‘ (modern: guma‘) is never marked for the guttural in Safford (1903) (the same applies to Fritz [1903]). Nevertheless, Jonker too treats the ―guttural‖ as a feature of the preceding vowel and thus denies the consonant the status of a phoneme. Kats‘s brief review of the sound system of Chamorro continues with a reference to Safford‘s observation (Safford 1903: 294) that mid-high and high vowels are often hard to distinguish one from the other (Kats 1917: 7). Kats‘s explanation for this phenomenon is insufficient and partly also faulty and thus Jonker (Kats 1917: 130) has to set things right in a lengthy note. On the remaining pages (Kats 1917: 7–9), the Umlaut phenomena, vowel syncope and apocope of word-final velar plosives are addressed. What strikes me as odd is Kats‘s silence about the putative quantity correlation with vowels marked by diacritics in Fritz‘s orthography (cf. above). Length distinctions are also abundant in the Chamorro-Wörterbuch (Fritz 1904) and thus Kats must have noticed the many macrons and breve signs. Jonker (Kats 1917: 129) has a note on Fritz‘s (1903) conventions to distinguish long and short vowels. However, Jonker too is content with quoting his source without weighing the pros and cons of a quantity correlation in Chamorro. Reading Kats‘s phonology section, one is none the wiser of the Chamorro sound system because the Dutch author does not fill the gaps left by his predecessors, nor does he rectify their mistakes. Given that Kats‘s main interests lay elsewhere (namely in the realm of comparative [historical] Austronesian linguistics), it is unfair to blame him for not clarifying the issues of synchronic Chamorro phonology. Moreover, the section devoted to phonology accounts only for a quarter of the original (and slightly less than a third of the notes). It is not unlikely that Kats has more to say on other subjects. We can be certain of the quality of Het Tjamoro only after a thorough examination of the remainder of the text.

4. Conclusions In the preceding sections, I have highlighted a selection of the early contributions by Germans and Dutchmen to the study of Chamorro. It is clear that in this article, I could only touch upon some interesting issues whereas some authors, many titles, and many more topics of grammar and lexicon were mentioned only in passing, if at all. The at times rather low editorial quality of the sources and their often doubtful analyses notwithstanding, the above texts and those alluded to are worthwhile studying in more detail. The value of these texts for the present and future generations of Chamorro native speakers and the linguistic community is undeniable. There is a plethora of data

German and Dutch contributions to Chamorro studies

223

which document the stage Chamorro was in a century ago. This snapshot of the past includes words – dialectal, regional or outdated – which nowadays are precious for those who want to keep Chamorro vital by way of enlarging its lexicon, for instance. The same holds for word-forms (in terms of phonological shape and morphological make-up), syntactic constructions, word meanings and idiomatic expressions including some forgotten Spanish elements. Thus, the old sources provide useful tools for any attempt at Ausbau of Chamorro which avoids the copying of English patterns. At the same time, the work by Fritz, Lopinot and Kats teaches the modern speech-community something about the history of their language. For the professional linguist, the same texts are interesting for three reasons. First of all, these sources make it possible to look at Chamorro also from a diachronic point of view (albeit with a relatively restricted time depth). Secondly, at least in some subject matters, the older sources contain more information than the most recent descriptions of Chamorro (although this is certainly untrue for phonology as the above discussion amply shows). Thirdly, in terms of the history of thought, the way the early grammarians of Chamorro were struggling to provide adequate interpretations of the facts is fascinating in itself. On the one hand, the diversity of opinions demonstrates that the facts are not self-explanatory in the sense that they only invite one interpretation. On the other hand, ideologies and various facets of Eurocentrism (including Americocentrism) come to the fore when the Spanish traditions are cut off for political reasons and the spelling rules of one‘s own language are transferred to Chamorro. These and many other aspects call for an annotated re-edition and translation not only of the texts mentioned in this article but also of the many other manuscripts and publications devoted to Chamorro before the 1950s.

Bibliography Barry, William J.; Klein, Cordula & Köser, Stephanie (1999): Speech production evidence for ambisyllabicity in German, in: Phonus 4, 87–102. Burrus, E. J. (1954): Sanvitores‘ grammar and catechism in the Mariana (or Chamorro) language (1668), in: Anthropos 49, 934–960. Caratini, Emilie (2007): There is no ambisyllabicity in German, in: Proceedings of ConSOLE XIV, 37–60. Chamisso, Adelbert von (1818a): Von Radack nach Guajan, in: Sydow, Max von (ed.), Chamissos Werke in fünf Teilen. Berlin: Deutsches Verlagshaus Bong & Co., 225–232. Chamisso, Adelbert von (1818b): Die Marianeninseln – Guajan, in: Sydow, Max von (ed.), Chamissos Werke in fünf Teilen. Berlin: Deutsches Verlagshaus Bong & Co., 97–105. Chamisso, Adelbert von (1818c): Vokabularium der Dialecte Chamori (Marianeninseln) und von Eap, Ulea und Radck, in: Sydow, Max von (ed.), Chamissos Werke in fünf Teilen. Berlin: Deutsches Verlagshaus Bong & Co., 67–82. Chung, Sandra (1983): Transderivational relationships in Chamorro phonology, in: Language 59(1), 35–64. Conant, Carlos Everett (1911): Consonant changes and vowel harmony in Chamorro, in: Anthropos 6, 136–146.

224

Thomas Stolz

Cooreman, Ann M. (1987): Transitivity and discourse continuity in Chamorro narratives. Berlin, New York: Mouton de Gruyter. Costenoble, Hermann (1940): Die Chamoro Sprache. ‗s-Gravenhage: Nijhoff. Elbert, Samuel H. & Kawena Pukui, Mary (1979): Hawaiian grammar. Honolulu: University of Hawaii Press. Fischer, Paulus & Lopinot, P.C allistus (1910a): Catesismon i Dottrina Cristiana mapublica para i Islas Marianas gi fino Chamorro pot i mamale Capuchino. Hongkong: Typis Societatis Missionum ad Exteros. Fischer, Paulus & Lopinot, P. Callistus (1910b): Gebetbuch in der Chamorro Sprache.. Hongkong: Typis Societatis Missionum ad Exteros. Fischer, Paulus & Lopinot, P. Callistus (1910c): Katechismus in der Chamorro Sprache. Hongkong: Typis Societatis Missionum ad Exteros. Fritz, Georg (1901): Die Chamorro. Eine Geschichte und Ethnographie der Marianen, in: Ethnologisches Notizblatt 3(1), 25–110. Fritz, Georg (1903): Chamorro Grammatik, in: Mitteilungen des Seminars für Orientalische Sprachen an der Friedrich-Wilhelms-Universität 6, 1–27. Fritz, Georg (1904): Chamorro-Wörterbuch. Berlin: Georg Reimer. Fritz, Georg (1906): Der Chaifi, in: Mitteilungen des Seminars für Orientalische Sprachen an der Friedrich-Wilhelms-Universität 9, 178–180. Fritz, Georg (1907): Kurze Geschichte der Marianen, in: Mitteilungen des Seminars für Orientalische Sprachen an der Friedrich-Wilhelms-Universität 10, 218–228. Fritz, Georg (1911): Die zentralkarolinische Sprache. Grammatik, Übungen u. Wörterbuch der Mundart der westlich von Truk liegenden Atolle, insbesondere der Saipan-Karoliner. Berlin: Georg Reimer. Hardach, Gerd (1990): König Kopra. Die Marianen unter deutscher Herrschaft 1899-1914. Stuttgart: Franz Steiner. Hezel, Francis X. (2001): Deutsche katholische Missionen in Mikronesien, in: Hiery, Hermann Joseph (ed.), Die Deutsche Südsee 1884–1914. Paderborn: Schöningh, 558–569. Hiery, Hermann Joseph (2001): Schule und Ausbildung in der deutschen Südsee, in: : Hiery, Hermann Joseph (ed.), Die Deutsche Südsee 1884–1914. Paderborn: Schöningh, 198–238. Hiery, Hermann Joseph (ed.) (2001): Die Deutsche Südsee 1884–1914. Paderborn: Schöningh. Ibáñez del Cármen, P. Aniceto (1863): I magajet na Quisiliano manaéyac nu y debe ujongui, ugagao, ufatinas yan uchajlao yan y palo na devocion sija na merece ufanmatungo. Manila: Ramirez y Giraudier. Ibáñez del Cármen, P. Aniceto (1865a): Gramática chamorra. Manila: Ramirez y Giraudier. Ibáñez del Cármen, P. Aniceto (1865b): Diccionario español-chamorro. Manila: Ramirez y Giraudier. Ibáñez del Cármen, P. Aniceto (1887): Devoción as San Francisco de Borja, Patron Luta: mapula y sagramentos sija ya jaf taimano ufanmaguef-recibe devoción as San Dimas, y mauleg na ladron, patron Maleso, yan doctrina na mapula. Manila: Imprenta Amigos del País. Kats, J[acob] (1917): Het Tjamoro van Guam en Saipan vergeleken met eenige verwante talen (met opmerkingen van Prof. Mr. Dr. J.C.G. Jonker). ‘s-Gravenhage: Martinus Nijhoff. Klein, Thomas B. (2000): >Umaut< in Optimality Theory. A comparative analysis of German and Chamorro. Tübingen: Max Niemeyer. Lopinot, P. Callistus (1910a): Debosionario Leblon Tinaitai Para Islas Marianas gi fino Chamorro pot i mamale Capuchino. Hongkong: Typis Societatis Missionun ad Exteros. Lopinot, P. Callistus (1910b): Historia Sagrada i Antiguo Yan i Nuevo Testamento. Freiburg im Breisgau: B. Herder.

German and Dutch contributions to Chamorro studies

225

Lopinot, P. Callistus (1910c): Chamorro Wörterbuch enthaltend I. Deutsch-Chamorro, II. ChamorroDeutsch nebst einer Chamorro-Grammatik und einigen Sprachübungen. Hongkong: Typis Societatis Missionum ad Exteros. Lopinot, Callistus (1910d): I Yiniusan na pinelo na Sacramenton Cumonfesta. Mapiula i tautau Marianas. Hongkong: Typis Societatis Missionun ad Exteros. Lopinot, P. Callistus & Schuster, Dr. I. (1909): Historia Sagrada i Antiguo Yan i Nuevo Testamento. Freiburg im Breisgau: B. Herder. Maas, Utz (1999): Phonologie. Einführung in die funktionale Phonetik des Deustchen. Opladen/Wiesbaden: Westdeutscher Verlag. Quilis, Antonio (1992): La lengua española en cuatro mundos. Madrid: Editorial MAPFRE. Rodríguez-Ponga, Rafael (1995): El elemento español en la lengua chamorra (Islas Marianas). Unpublished PhD-Thesis. Madrid: Universidad Complutense. Facultad de Filología. Rodríguez-Ponga, Rafael (2001): Los numerales hispano-chamorras, in: Zimmermann, Klaus & Stolz, Thomas (eds.), Lo propio y lo ajeno en las lenguas austronésicas y amerindias. Procesos interculturales en el contacto de lenguas indígenas con el español en el Pacífico e Hispanoamérica. Frankfurt/M.: Vervuert, 253–278. Safford, William E. (1903): The Chamorro language of Guam – I–IV, in: American Anthropologist 5, 289–311 and 508–529. Safford, William E. (1905): The Chamorro language of Guam V, in: American Anthropologist 7, 305–319. Safford, William E. (1909): The Chamorro language of Guam. Washington D.C.: W.H. Lowedermilk. Spennemann, Dirk H. R. (2004): An annotated bibliography of German language sources on the Mariana Islands. Saipan: Division of Historic Preservation. Sydow, Max von (1907a): Chamissos Werke in fünf Teilen. Vierter Teil: Reise um die Welt I. 1. Theil: Tagebuch. Berlin: Deutsches Verlagshaus Bong & Co. Sydow, Max von (1907b): Chamissos Werke in fünf Teilen. Fünfter Teil: Reise um die Welt II. Berlin: Deutsches Verlagshaus Bong & Co. Sydow, Max von (1907c): Chamissos Werke in fünf Teilen. Fünfter Teil: Reise um die Welt II. 2. Theil: Bermerkungen und Ansichten. Berlin: Deutsches Verlagshaus Bong & Co. Stolz, Thomas (2007): The Kurze Geschichte der Marianen by Georg Fritz. A commented re-edition, in: Schrader-Kniffki, Martina & Morgenthaler García, Laura (eds.), La Romania en interacción: entre istoria, contacto y política. Ensayos en homenaje a Klaus Zimmermann. Frankfurt: Vervuert, 307–349. Stolz, Thomas; Schneemann, Christina; Dewein, Barbara & Chung, Sandra (this volume): The mysterious H. Who was the author of Die Chamoro Sprache? Topping, Donald M. (with the assistence of Bernadita C. Dungca) (1973): Chamorro reference grammar. Honolulu: University of Hawaii Press. Topping, Donald M.; Ogo, Pedro M. & Dungca, Bernadita C. (1975): Chamorro-English dictionary. Honolulu: University of Hawaii Press. Underwood, Robert A. (1983): Kuenta – Report. Kumision i Fino‘ Chamorro. Agana: Government of Guam. Von Preissig, Edward R. (1918): Dictionary and grammar of the Chamorro language of the Island of Guam. Washington: Government Printing Office. Winkler, Pierre (this volume): Subject, topic, passive and perspective in Functional (Discourse) Grammar and in Philippine Missionary Grammar. Zimmermann, Klaus (this volume): The Diccionario español-chamorro (1865) of Padre Fray Aniceto Ibáñez del Cármen: a historiographical characterization of a pedagogic-lexicographical discourse type in colonial Austronesia. Stolz, Thomas (this volume): The Gramática chamorra

THOMAS STOLZ, CHRISTINA SCHNEEMANN, BARBARA DEWEIN & SANDRA CHUNG (BREMEN/GERMANY & SANTA CRUZ/USA)

The mysterious H. Who was the author of Die Chamoro Sprache?

Abstract This article tries to establish the identity of the author of Die Chamoro Sprache, which was published in 1940. It is demonstrated that previous attempts to identify H. Costenoble‘s full first name have yielded wrong results. Evidence assembled from the history of the Costenoble family points to Hermann Costenoble as the only likely candidate for the author. The article concludes with a brief discussion of some interesting aspects of Costenoble‘s grammar.

1. Introduction This article addresses a surprisingly intricate issue: the identification of the author of the most sizeable grammar ever of Chamorro. What at the outset seemed to be a minor task, to be handled on the side, has proved to require the expertise of private investigators, in our case the joint efforts of four linguistically trained ―detectives‖ and their historically and genealogically versed ―informants‖ on three continents.1 With a view to preparing Die Chamoro Sprache for an annotated re-edition and translation, we had intended to provide as much bio-bibliographical information as possible on the author. However, this means that we had to know the identity of the author in the first place – and establishing this identity gave us a very hard time. The initial H. of the author H. Costenoble poses a problem because it has been spelled out differently by various Chamorro specialists who came after Costenoble (cf. below). From a purely linguistic standpoint, the issue might seem to be relatively marginal: linguistics is not normally about individuals and their family trees, but about languages and their structures. Nevertheless, coming to grips with the historical truth is important for the Chamorro community, whose members want to know about their past – and this includes knowledge about who said what when about their language. Moreover, knowing the identity of the author may also be helpful when it comes to interpreting the analyses, arguments and attitudes expressed in a given text. This is certainly the case with H. Costenoble and Die Chamoro Sprache. 1

We are grateful to Andy Airriess, Marlies Salazar, and Dirk H. Spennemann for their (intercontinental) helping hands.

228

Thomas Stolz et al.

Finally, because the history of linguistics is a discipline concerned with the development and spread of linguistic ideas, it is also concerned with identifying the propagators of those ideas. In contrast to its predecessors, i.e. Sanvitores 1668 (Burrus 1954), Safford (1903, 1905), Fritz (1903), Lopinot (1910), Kats (1917) and Von Preissig (1918), Die Chamoro Sprache is a veritable in-depth study of the indigenous language of the Mariana Islands. Before its publication in 1940, information about Chamorro was somewhat sketchy and not always philologically accurate. H. Costenoble‘s descriptive grammar, with its 550 printed pages, provides an unprecedented wealth of data, analysis and insights. Because of this richness, Die Chamoro Sprache supplements Topping‘s (1973) reference grammar in many respects and deserves to be re-published in a format that can reach a wider audience. Of course, Die Chamoro Sprache is also special in some ways, and not all of the ideas of its author meet the standards of linguistic reasoning. That is why the planned re-edition cannot make do with a faithful reproduction of the original text and an English translation. There is ample need for accompanying commentary. Section 3 of this article highlights a few selected characteristics of Die Chamoro Sprache. Before turning to that, we would like to shed some light on the author of the grammar under scrutiny. Step by step, section 2 assembles the evidence for a definitive identification of the mysterious H.2

2. In search of an author… 2.1. Where the problem starts In an article dedicated to the clarification of Chamorro direction terminology, Solenberger (1953: 140) argues that [b]y far the most complete and linguistically informed grammar of Chamorro is that by Hertha Costenoble [our emphasis], who, according to her own preface, lived in Guam as a child between the years 1905 and 1913. Information from contemporary Saipanese indicates, however, that she, or at least some members of her family, were in Saipan during part of the German period. On the same page, Solenberger quotes extensively from Die Chamoro Sprache and refers to the author consistently with the feminine pronoun (she, her). Nowhere in his text does Solenberger mention how he came to conclude that the author was Hertha Costenoble. Since, at the time of writing, Solenberger was the district anthropologist of 2

More light is shed on this issue in a recent article by Salazar (2009). She had studied in the Philippines herself, where she met Cecilio Lopez (cf. below), who told her of his personal experience with H. Costenoble. Thus Marlies Salazar knew the identity of H. Costenoble even before we set out to investigate the issue.

Who was the author of Die Chamoro Sprache?

229

Saipan, he might have had access to archival material in which the authorship of Die Chamoro Sprache was disclosed in some way. But as will become clear below, this scenario is highly unlikely. We simply do not know Solenberger‘s motivation for ascribing the German grammar of Chamorro to Hertha Costenoble. Much later, Solenberger‘s unexplained choice was readily accepted by Chung (1998: 11), who claims that [t]he most detailed of the traditional works are the grammatical sketch by William Edwin Safford (1903-5) and the much longer grammar by Hertha Costenoble [our emphasis] (1940), both of which are excellent. (See Solenberger 1953 on the identity of the author of the latter work.) Stolz (2007, 311, fn. 7) follows Solenberger (1953) and Chung (1998) when he observes that [t]he identity of H. Costenoble is still a matter of dispute. The initial H. could stand for either Hermann or Hertha, both of whom lived in the German colony and later went to Guam. As the author states that she learnt Chamorro as a child, Hermann can be ruled out because of his age [our emphasis]. The majority of linguists working on Chamorro who feel the necessity to quote from Die Chamoro Sprache, or have it at least in their bibliography, are more cautious as to the identity of the author and do not spell out the first name (e.g. Cooreman 1987). Topping (1973) always avoids spelling out Costenoble‘s first name. However, Topping (1973: 6) assumes the author to be male: H. Costenoble was the last of the German grammarians who worked on Chamorro. Having learned to speak the language as a child on Guam (from 1905 to 1913), he is the first of the grammarians to have a high degree of fluency in the language at least by his own account. His work, Die Chamoro Sprache (1940), the result of over fifteen years‘ work, is a complete grammar, filling 550 pages and containing a vast amount of illustrative material. Unfortunately, its contents are not accessible to most Chamorros and Americans because they are written in German. Also, many of his examples are rejected by Chamorro speakers because they are ungrammatical [our emphasis]. Although Topping seems not to know Costenoble‘s exact first name, he is certain of Costenoble‘s male gender. This is in line with Spennemann (2004), whose bibliography of German language sources on the Mariana Islands lists Die Chamoro Sprache (under #994) and identifies the author as Hermann H.L.W. Costenoble. As we demonstrate below, this Hermann H.L.W. Costenoble is identical to the Hermann Costenoble in the above quote from Stolz – and neither he nor Hertha can be the author of Die Chamoro Sprache, for reasons exposed below.

230

Thomas Stolz et al.

2.2. The Costenoble family tree Thanks to the genealogical information provided by Andy Airriess on the internet (http://www.gencircles.com/users/airriess/1), the family tree of the Costenobles can be traced back to the early 17th century. There was also a Dutch branch of the family whose members called themselves van Costenoble. However, the surname is neither genuinely German nor Dutch. It is likely that the Costenobles came originally from a Romancespeaking area from which they immigrated to the German Empire. The earliest name mentioned is Jacques van Costenoble, born in Belle/Flanders, in the first half of the 17th century. From Flanders, the family went to Frankenthal in the Palatinate, then to Leiden and Sudenburg-Magdeburg, back westwards to Herford/Westphalia, back to Magdeburg and then to Jena in Thuringia. The Costenoble men in the 18th century had a predilection for marrying French wives which – together with the French first name of the earliest Costenoble – may be taken as evidence that they had old cultural ties to France.3 Chances are that the Costenoble immigrants were French Huguenots (expatriates); at least for the later generations, the marriages were celebrated in the local Protestant churches. Among the Costenobles we find inter alia a famous German stage-actor and writer of the 18th century (= Karl Ludwig Costenoble, Herford), one of the leading German literary publishers of the 19th century (= Hermann Costenoble, Jena) and a famous female painter and illustrator of the early 20th century (= Anna Costenoble, Danzig). Already these few names are indicative of an interesting, culturally active family. The Costenobles we are especially interested in are those of the Jena branch. Hermann Ludwig Wilhelm Costenoble4 (*4 March, 1867 in Jena, ý1936 in Leyte/Philippines) is the head of the first German family to settle in the newly acquired German colony, the Mariana Islands, in 1903. He married Gertrude Blum in 1892 in Rheinfelden, a town divided in two by the Swiss-German border. Their son Hermann Costenoble – the first of their nine children – was born on the German side of the border on 15 January, 1893; his sister Gertrude was born in the Swiss part of Rheinfelden on 20 December, 1893. In quick succession, three further siblings were born whose first names have an initial H: Hertha (*1895), Hilde (*1896) and Herbert (*1898). Together with their elder brother Hermann and their father Hermann Ludwig Wilhelm, they form a group of five candidates for the author of Die Chamoro Sprache, as H. Costenoble could refer to any of them. The other four children – Gertrude, Erich, Wilhelm, Fritz, and Edith – can be excluded as candidates for the author.

3

4

The name Costenoble is reported variously in the northern Départements of modern France. However, there are also Costenobles in Switzerland and elsewhere. Ballendorf (2001: 752–753) claims that the original name was von Costenoble and that the ―von‖ was dropped after the Costenobles had moved to Guam.

Who was the author of Die Chamoro Sprache?

231

2.3. The Costenobles in the Marianas and the Philippines The Costenobles arrived in Saipan in March 1903 and settled near Talofo‘fo‘.5 It had been hoped that they would be the first of a much larger contingent of German settlers (Hardach 1990: 129). Governor Fritz‘s hopes for a smallish community of up to 50 settlers soon turned out to be in vain.6 The number of German residents remained minimal throughout the existence of the colony. Moreover, in 1904, slightly more than a year after their arrival in Saipan, the Costenobles left the German Marianas and moved to American Guam (Hardach 1990: 130).7 The Costenobles had baptised their son Fritz (born on Saipan shortly after their arrival) after his god-father Georg Fritz, the Governor. This optimistic beginning was followed by disillusion and journalistic invectives by the Costenobles‘ pater familias. In 1905, Hermann L.W. Costenoble published an article in a widely read German magazine accusing Georg Fritz of a pro-Chamorro policy which supposedly was detrimental to the prosperity of the German settlement.8 In 1906, Edith, the ninth child of the family, was born on Guam; Wilhelm died there in 1910 at the age of only nine. Gertrude Costenoble née Blum, the mother of the Costenoble offspring, died on Guam in 1915. All five members of the family with an H-initial first name re-settled on Guam. However, their father, who at that time was 37 years old, does not fit the autobiographical description given in the preface (―Vorwort‖) to Die Chamoro Sprache, where the author reminisces that [m]eine Kenntnisse des Chamorro erwarb ich mir auf Guam als erwachsenes Kind, im täglichen Verkehr mit den Eingeborenen, zwischen den Jahren 1905 und 1913. (Costenoble 1940a, v) [I acquired my knowledge of Chamorro as a grown-up child on Guam from 1905 to 1913, in daily dealings with the indigenous population]. Even the somewhat awkward wording erwachsenes Kind (literally, ―grown-up child‖; the intended reading most probably was Heranwachsender or Jugendlicher ―adolescent; 5

6

7

8

Wherever the name Costenoble is mentioned in scientific work on the history of the Marianas, reference is made to Hermann L.W. Costenoble i.e. to the head of the family (Hardach 1990, Farrell 1991, Rogers 1995, Ballendorf 2001). In point of fact, Governor Fritz was decidedly against mass immigration; Hiery (2001: 18) terms his policy ―eine Antieinwanderungspolitik‖. Fritz discouraged potential immigrants from coming to the Marianas and ―jene, die sich an seinen Rat nicht gehalten hatten und doch nach Saipan oder Rota gekommen waren, ließ er buchstäblich am eigenen Leibe spüren, daß die deutsche Verwaltung nicht dazu da war, deutsche Siedler zu unterstützen‖. [he let those who ignored his advice and had come to Saipan or Rota nevertheless experience personally that it was not the duty of the German administration to support German settlers]. According to Farrell (1991: 275), the Costenobles changed their minds again and wished to move back to Saipan. However, their request was turned down by the German authorities. For a journalistic reply to these invectives, cf. Fritz (1906).

232

Thomas Stolz et al.

youngster‖) does not allow us to assume that the adult head of the family is referring to himself (as a grown man). Later on the same page, the author claims that s/he learned the language ―als Kind‖ i.e. ―as a child‖. Only the children Hermann, Hertha, Hilde and Herbert fit the description, as their ages at that time were 12–20 for Hermann, 10–18 for Hertha, 9–17 for Hilde and 7–15 for Herbert. In addition, the father always signed his publications at least with the initials of his three Christian names (either H.L.W. Costenoble or Hermann L.W. Costenoble), whereas the grammar makes do with one abbreviated first name. None of his children bore more than one baptismal name: for the Costenoble children, there were no middle initials or the like. In the above quote, the year 1913 is interesting. H. Costenoble mentions this year several times as the endpoint of his sojourn on Guam and explicitly talks about his ―Abfahrt von Guam im Jahre 1913‖ i.e. ―my departure from Guam in 1913‖ (Costenoble 1940a: 125 [and 259]). For all we know, Hermann L.W. Costenoble remained on Guam uninterruptedly until sometime during World War I. Since colonial history experts do not mention the other members of his family, we cannot be sure who left Guam and where s/he went and for what reason. What complicates the story a little bit is the fact that Hermann L.W. Costenoble was still alive when the manuscript of Die Chamoro Sprache was completed. The author signed the preface on 21 April, 1935 (i.e. five years prior to publication) in Del Carmen, Pampanga in the Philippines, where several of the Costenobles moved at the end of World War I.9 Hermann L.W. Costenoble died a year later in Leyte in the Philippines. There thus seems to be a very slim (but, in the end, unlikely) chance that he could have authored the grammar. His four children, however, are better candidates. The group of potential candidates for the author of Die Chamoro Sprache dwindles further because Hertha Costenoble died in Manila in 1918, at the age of 23. Thus, she could not have signed the preface of the grammar seventeen years later. In the preface, the author tells the reader that 9

The prelude to the Costenobles‘ taking residence in the Philippines could be turned into a Hollywood movie script. When World War I broke out, the father, who at that time worked for the Guam forest authority and had become its Chief Forester in 1910, was considered a potential German spy. A US army captain of German extraction, Hans Hornbostel, was ordered to keep watching on the suspect. Hornbostel fell in love with the Costenobles‘eldest daughter, Gertrude, whom he married almost straight away. Later, the head of the Costenoble family was fined for illegally trying to leave Guam (taking funds with him). In 1916, Hermann L.W. Costenoble was also temporarily declared insane after beating up his son Fritz. Released from prison, he was told to leave Guam for Manila. He must have returned from Manila relatively soon because in 1917, when the USA declared war on Germany, Captain Hornbostel had to arrest his father-in-law, who became an internee with other German citizens on Guam (Rogers 1995: 136, 140). After the war, the Hornbostels became actively involved with the preservation of Chamorro culture and did fieldwork on island archaeology in the Marianas for the Bishop Museum in Hawai‘i. Gertrude Hornbostel (also known as Trudis Aleman) published on various aspects of Chamorro ethnography and linguistics, e.g. on Chamorro nicknames (Hornbostel 1924).

Who was the author of Die Chamoro Sprache?

233

[m]it der Zu-Papier-Bringung der in diesem Werke enthaltenen Einzelheiten begann ich 1915, als mir die Sprache noch lebend in Erinnerung war; beendet wurde sie, und damit die erste Bearbeitung des Buches, im Jahre 1919 (Costenoble 1940a, v) [I began to write down the details contained in this work in 1915 when I still remembered the language vividly; I terminated this work and thus the first draft of the book in 1919]. This detail too excludes Hertha as the author of Die Chamoro Sprache, because the author was still working on the first draft of the grammar a year after Hertha‘s death. Moreover, since Hertha had married in 1917, according to the general practice of the times, she would have signed with her married surname, which was Lommans. One could argue that Solenberger (1953) and his local informants might have confused her erroneously with her younger sister Hilde. Unfortunately, Hilde is the Costenoble sibling we know least about. In 1919, she was reported to be living in Manila, where she received financial support from the Hornbostels (i.e. from her sister Gertrude and her husband Hans Hornbostel). We know nothing about her later whereabouts or the date of her death. However, she was married to a certain Erich Kohlhauser (of whom we know equally little) and thus one would expect her to use her married name from then onwards. Thus, the two female candidates can be ruled out with a relatively high degree of certainty. This leaves us with the brothers, Hermann and Herbert Costenoble. The younger of the two, Herbert (ý1986), married Marianne Lea Rabel in Shanghai on 26 October, 1935. This means that he must have been in China half a year after the preface of Die Chamoro Sprache was signed in the Philippines. This is admittedly a possible scenario: the six months between the completion of the manuscript and the marriage ceremony leave ample time for travelling from the Philippines to China. However, assuming that the future couple met in China, where both of them – for reasons unknown – were spending time, it seems relatively unlikely that Herbert was in the Philippines earlier in 1935. All these arguments, of course, rest entirely on conjecture. There is no hard proof of Herbert‘s absence from the Philippines in April 1935. Moreover, we lack information on most of Herbert‘s life. Going by the time-span mentioned in the preface, Herbert would have been 15 years old in 1913 and 17 when the author started writing the first draft. This would not be an impossible age for a gifted youngster to engage in grammarwriting, although the probability seems relatively low. What about the eldest of the Costenoble siblings, Hermann? Hermann was killed in 1942 when the Japanese attacked the Philippines. He died in Floridablanca, a municipality which happens to be part of Pampanga (Central Luzon) – exactly the same region of the Philippines where the preface of Die Chamoro Sprache was signed. (Del Carmen is a Barangay of Floridablanca.) The other Costenobles lived either in Manila (Luzon) or Leyte (Visayas), which are some distance from Pampanga. This makes Hermann our ―chief suspect‖. This view is supported by further, more substantial evidence. There are a number of other publications in the field of linguistics by H. Costenoble. The earliest

234

Thomas Stolz et al.

we know of is Costenoble (1934), whose date of publication does not help to identify the author. However, again in the preface of Die Chamorro Sprache (Costenoble 1940a, v), the author boasts of having learned a number of Philippine ―dialects‖. This is reflected by the list of further publications which are attributed to H. Costenoble: 1 2 3 4 5 6

Costenoble (1936), miscellanea on Philippine languages, Costenoble (1937a), the Philippine verb, Costenoble (1937b), monosyllabic roots, Costenoble (1937c), historical phonology, Costenoble (1940b), diachronic sound changes in Tagalog, Costenoble (1979), Proto-Philippine dictionary.

The last item was published posthumously by Ernesto Constantino and his collaborators. The compiler handed over the original manuscript to the renowned Philippine linguist, Cecilio Lopez, who had studied in Germany for some time and was thus able to translate the text into English. He died, however, before the dictionary was eventually published (Zamora & Olsen 1980). In the University of Philippines Integrated Library System (and elsewhere in the Philippines),10 the dictionary compiler is catalogued as Henry Costenoble, although no member of the family with this first name has been reported in the Costenobles‘ genealogy.11 Given the publication dates of the above items, Hermann L.W. Costenoble can be struck from our list of possible candidates for the author of Die Chamoro Sprache. All publications appeared after the father‘s death in 1935, and the manuscript of the Proto-Philippine dictionary was personally delivered to Cecilio Lopez by the compiler in 1942. It is hard to believe but even in the preface to the edition of this dictionary where Cecilio Lopez tells the reader about his personal experience with the author, Lopez always calls him H. Costenoble.12 Together, these pieces of evidence point to Hermann Costenoble as the only likely identity of H. Costenoble. Thus, in lieu of Hermann senior, his son Hermann junior authored Die Chamoro Sprache. The sources quoted in section 2.1 which dare to give the full first name(s) of H. Costenoble have not hit the mark: neither Hertha nor Hermann L.W. Costenoble is the correct answer to the question posed in the sub-title of this article. Discounting the few remaining doubts, the mysterious H. must be Hermann Costenoble. In the remainder of this article we thus spell out H. as Hermann.

10

11

12

At least in Germany, this dictionary is a rare bird as we were not able to locate it in any library, not even those which specialize in the Philippines. There were four ancestors of H. Costenoble with the first name Heinrich (the German equivalent of Henry), all of whom lived and died before the turn of the 19th century. Lopez also admits that he never asked H. Costenoble about his life experiences before he moved to the Philippines although H. Costenoble once let slip that on Guam, when there was a conversation at night time, people usually mistook him for a native Chamorro. Costenoble‘s English on the other hand is characterised as heavily Philippinised.

Who was the author of Die Chamoro Sprache?

235

3. Some facts about Die Chamoro Sprache 3.1. Hermann Costenoble on himself and on others Having narrowed down the list of suspects to one likely author, we now take a cursory look at the grammar itself. This sub-section focuses on the preface; sub-section 3.2 touches on various issues connected to the grammar‘s contents, style and terminology. In the preface, Hermann Costenoble provides a list of pertinent works that he had access to, namely, Ibáñez del Cármen (1865), Safford (1903, 1905), Fritz (1908a–b), Lopinot (1910), Kats (1917) and Von Preissig (1918). However, he finds fault with the extant literature on Chamorro and thus ―[i]ch selbst habe in meiner Behandlung der Grammatik dieser Sprache keines dieser Werke benutzt‖ [for my treatment of the grammar of this language, I have not made use of any of these works myself]. (Costenoble 1940a, v). He is evidently proud of his near native competence in Chamorro, observing that [i]ch glaube sagen zu können, dass ich die Sprache so gut sprach wie der Durchschnittseingeborene; dass ich nicht über den Wortschatz verfügte, wie z.B. ein sechzigjähriger Chamorro, ist dabei selbstverständlich; auch dass mir vielleicht diese oder jene Eigentümlichkeit der Sprache nicht zu Wissen gekommen ist. […] Ich habe die Sprache als Kind erlernt, so wie sie der Eingeborene selbst lernt, und ich glaube sie in allen ihren Einzelheiten so zu empfinden, wie sie der Eingeborene selbst empfindet [I feel entitled to state that I spoke the language as well as the average native; it is understood that I did not have the lexicon of, say a sixty-year-old Chamorro; similarly it is possible that this or that peculiarity of the language has not come to my attention. […] I learned the language as a child just like a native learns it, and I believe that I have a feeling for all its details just like a native]. (Costenoble 1940a: v) Safford (1903, 1905) and Lopinot (1910) are discounted because they are said to contain too many typographical errors and to be incomplete in many respects. For Fritz (1908), Hermann Costenoble reserves the following negative remark: Fritz hat die Chamorrosprache selbst nur gekauderwelscht, und beschreibt in seiner Grammatik dieses sein Kauderwelsch [Fritz spoke Chamorro only as a hotchpotch, and in his grammar he describes this hotchpotch]. (Costenoble 1940a: v) This severe judgement cannot really be based on personal experience, since Hermann Costenoble was hardly on Saipan long enough to become acquainted with the German governor‘s competence in Chamorro. Hermann Costenoble left Saipan for Guam after less than a year, at the age of 10 or 11. Most probably, his statement about Fritz‘s dis-

236

Thomas Stolz et al.

torted Chamorro reflects the Costenoble family‘s old rancour against the German colonial authorities for not giving them the support they expected to receive. In point of fact, Hermann Costenoble did not discard the extant literature in its entirety. For instance, he admits that he made use of Lopinot (1910) in order to reconstruct the pre-Hispanic numerals of Chamorro. Moreover, he made use of all the literature he mentioned when it came to producing the basic vocabulary list (Costenoble 1940a: v). In the descriptive grammar, the author reveals that he also consulted the standard Austronesianist literature of his time (Dempwolff 1920, Brandstetter 1915, so-called ―modern Dutch authors‖, etc., e.g. Costenoble [1940a: 101–102]). From the author‘s description, we infer that after the first draft was completed in 1919, he had no opportunity to do fieldwork in the Marianas. However, he mentions that he repeatedly discussed problematic issues with Chamorro native speakers who happened to be in Manila. Hermann Costenoble also concedes that his long absence from the Chamorro ambiance and his exposure to Philippine languages might have affected his ability to recall the Chamorro facts correctly. ―Ich glaube jedoch, dass die Wahrscheinlichkeit hierfür nicht gross ist.‖ [However, I believe that the probability is not high] (Costenoble 1940a: v). The author is obviously convinced of his practical language competence, but tries to hush potential professional critics by claiming that he is ―kein akademisch gebildeter Sprachwissenschaftler‖ [not academically trained linguist] (Costenoble 1940a: vi). He uses this excuse to defend his departures from the canonical handling of certain phenomena (cf. below). Despite his disclaimers, the grammar does not strike the reader as the work of a philological layman.

3.2. The grammar proper Space restrictions do not allow us to discuss the merits and flaws of Die Chamoro Sprache at length in this article. This is a task to be tackled in connection with our planned annotated re-edition of the text. Here we briefly mention a few aspects of the grammar which might be of interest to our readership. Overall, Die Chamoro Sprache is a very thorough in-depth study of the language, covering a wide range of areas. In this respect it surpasses all previous, more or less sketchy descriptions of Chamorro. Further, unlike many of his predecessors, Hermann Costenoble goes to the trouble of explaining his analyses and ideas and supporting them with arguments, most of which are linguistically sound. This is what makes Die Chamoro Sprache the very first truly linguistic description of Chamorro. The grammar is divided into four parts (―Hauptteile‖). Part I is dedicated to ―Laut und Silbe‖ [Sound and syllable] and covers almost 160 pages. Part II treats ―Die Wortklassen‖ [The word classes] and, at some 330 pages, is twice as long as the phonological part. There is also a syntax chapter, viz. Part III ―Der Satz‖ [The sentence], comprising 40 pages. The final 13 pages form Part IV ―Texte‖ [Texts]. These parts are subdivided into dozens of sec-

Who was the author of Die Chamoro Sprache?

237

tions and sub-sections (the table of contents is seven pages long). Like Kats (1917), Hermann Costenoble takes a pan-Austronesian comparative approach to Chamorro phonology, which accounts for three-quarters of the entire first ―Hauptteil‖ (pp. 13– 134). It is also remarkable to see that reduplication (―Verdoppelung‖) is treated under phonology (pp. 135–145). Diachronic and comparative observations are less prominent in the other parts of Die Chamoro Sprache. The coverage of phenomena in this grammar is so impressive that it can still compete with later work on Chamorro, such as Topping‘s (1973) reference grammar. We do not mean to imply that Hermann Costenoble never erred. However, for a variety of topics, he provides a much richer exemplification and explanation than do his successors. Before we turn to the grammar‘s idiosyncrasies, some of which Hermann Costenoble himself was aware of, let us address the issue of the Chamorro data presented in the grammar. There are literally hundreds of Chamorro sentences and phrases plus countless individual words used to illustrate the author‘s points. However, the author does not identify his sources properly. The only clear indication is given in the introduction to the sample texts (―Vorbemerkungen‖) (Costenoble 1940a: 522). Hermann Costenoble tells his readers that these sample texts are drawn from the anthology of texts collected by his sister and his brother-in-law, Gertrude and Hans Hornbostel, in the 1920s. More particularly, he specifies that, for the syntax chapter, he has used some of the texts published in Thompson (1932) which were originally collected by Gertrude Hornbostel. Since Costenoble employs a transcription of his own creation, the texts accompanying Die Chamorro Sprache do not closely resemble their counterparts in the Thompson (1932) collection.13 It remains to be seen whether this dissimilarity results only from differences in spelling. For the bulk of Costenoble‘s examples, a source is not specified. A number of sentences probably come from religious texts. Since the Spanish colonial orthography is replaced systematically by Costenoble‘s transcription (Costenoble 1940a: 8–9), it is hard to tell from which original source these sentences are drawn, although it cannot be doubted that the author also made use of texts written in the old orthography. It cannot be ruled out that a sizable number of examples are the product of Costenoble‘s (supposed) near-native competence and memory. This may explain the fact that Topping‘s (1973: 6) informants rejected many of Costenoble‘s examples as ungrammatical. Yet, it remains to be clarified how many and exactly which examples are rejected by contemporary speakers of the language, and why (which adds another task to the agenda of the re-editors of Die Chamoro Sprache). Some glimpses of Costenoble‘s treatment of synchronic Chamorro phonology can serve to illustrate his approach to analytical problems. At first glance, one is impressed by the inventory of symbols used to represent Chamorro speech sounds. With the help 13

In modernised spelling, two of Costenoble‘s sample texts have made it into the practical Chamorro language course by Topping (1980).

238

Thomas Stolz et al.

of superscript letters, Costenoble postulates a system of 13 monophthongs which covers the entire range of allophonic variation of simple vowels in Chamorro (Costenoble 1940a: 1–2). However, the author adds that [i]n dieser Schrift sind meistens nur die einfachen Zeichen a, o, u, e, i angewandt worden; nur wenn die genaue Klangfarbe hervorgehoben werden soll, sind die besonderen Lautzeichen verwendet worden [in this work, mostly the simple symbols a, o, u, e, i have been used; the special sound symbols have been used only to reproduce the exact sound shape]. (Costenoble 1940a: 2) In a way, this step back from the over-differentiating transcription system reflects the distinction between phonemic and sub-phonemic contrasts. As far as we can tell, Costenoble was not familiar with Praguean phonology and thus had to struggle with the problem of keeping relevant and irrelevant categories apart. Moreover, Costenoble‘s chart of the Chamorro consonants is interesting because it contains two columns whose slots are filled by identically transcribed elements, namely /b/, /d/, // and /g/. The two parallel columns bear the headings: (a) ―Orale Verschlusslaute stimmlos weich‖ and (b) ―Orale Verschlusslaute stimmhaft weich‖. That is, both contain non-nasal obstruents (more precisely, plosives) which are ―weich‖ (―soft‖) as opposed to the ―hart‖ (―hard‖) /p/, /t/, and /k/. However, set (a) of the so-called soft plosives also bears the feature [– voice], which is a property one would not automatically associate with a symbol like /b/. Costenoble (1940a: 3) acknowledges that the members of the two columns (a) and (b) would be better represented by distinct symbols. From his further comments, it becomes clear that the members of the (a)-set are lax and voiceless, whereas the members of the (b)-set are lax and voiced. The author is dissatisfied with his transcription convention because he recognizes that it would have been more in line with the phonology of Chamorro to represent the (a)-set by modified symbols of the tense and voiceless plosives. However, his original decision could not be revised because ―[d]ie restlichen Kapitel dieses Werkes sind bereits geschrieben‖ [the other chapters of this work have been written already] (Costenoble 1940a: 3). This remark suggests that he recognised the problem when he was putting the final touches on an almost completed manuscript and shied away from having to revise so extensively. In a later paragraph Costenoble observes that he left Guam more than twenty years ago and thus no longer remembers exactly which words contain syllables whose codas end in a (supposedly) glottalised plosive (―Verschlusslaut … mit gleichzeitigem Verschluss des Kehlkopfes‖). Thus, he refrains from indicating this phenomenon in the transcriptions (Costenoble 1940a: 3). Even more striking is his decision to classify the glottal fricative /h/ as a semivowel (―Halbvokal‖). He knows that this classification is at odds with the experts‘ commonly held view (Costenoble 1940a: vi). However, he defends his practice as follows. The ―Hauchlaut‖ /h/ occurs in intervocalic position to delete the hiatus and thus behaves functionally like the true semivowels /j/ and /w/. The reason for the unexpected classification of /h/ can be termed ―functional‖; i.e. the fact

Who was the author of Die Chamoro Sprache?

239

that /h/ patterns with the semivowels makes it a semivowel, too. However, this argumentation (which is interesting, though untenable) is based on Costenoble‘s interpretation of the reconstructed Proto-Indonesian (= Proto-Malayo-Polynesian) sound system (Costenoble 1940a: 13–7). Synchronically, the functional equivalence or similarity of the three phonemes cannot be upheld, as their distributional profiles diverge massively from one another. To Costenoble‘s credit, daring proposals of this kind are rare in Die Chamoro Sprache. Costenoble‘s treatment of Chamorro morphology recognizes some of the distinctive morphological properties of the language. For instance, he recognizes that word classes in Chamorro are not the same as in German, and suggests a classification for Chamorro words which differs from the German classification. Nonetheless, he seems sometimes to revert to the German classification. For instance, he observes that i (which in our view is a determiner) can serve as article, preposition, and relative pronoun; this leads him to propose a single word class for words fulfilling these functions (Costenoble 1940a: 11). Nonetheless, he lists several word classes that include i (article, preposition, conjunction, ‗Relativum‘), and within each word class, the German equivalent for i is given. It seems likely that Costenoble proceeded by translating Chamorro sentences into German and then listing the German equivalents for i; the word classes of these German equivalent were, in turn, taken as the basis for the description of i. Beyond the grammar‘s analyses, Costenoble‘s style and choice of words are at times idiosyncratic. On the one hand, his command of German is not always up to standard. Besides a noticeably high number of typos, there are also turns of phrase which run counter to idiomatic German usage, e.g. neuter gender with ―das Grad‖ in lieu of correct ―der Grad‖ (in the case of comparison of adjectives) (Costenoble 1940a: 428). This error is not a hapax but recurs throughout the text. A formulation like ―nach der Tonsilbe oder mehr-denn-eine Silbe vor der Tonsilbe‖ [after the stressed syllable or more than one syllable before the stressed syllable] (Costenoble 1940a: 8) looks very awkward with the multiply hyphenated sequence of words. Moreover, the correct case frames of German verbs seems to have escaped the author more than once: ―Chamoro bedient sich am Silbenende vielfach von Verschlusslauten‖ [In the syllable coda, Chamorro often makes use of plosives] (Costenoble 1940a: 3) where sich bedienen ‗to make use (of)‘ should govern the genitive on its complement NP (verschiedener Verschlusslaute ―of various plosives‖). The use of the PP governed by von ‗of‘ is absolutely ungrammatical even in non-standard varieties of German. Slips of the pen of this kind occur in various places in the book. Nevertheless, they are not so frequent as to render the text stylistically unacceptable. One notices that the writer occasionally was at a loss when the constructions became complicated. Perhaps some of his departures from good German style are caused by his exposure to English and to Philippine languages during his long years in the Philippines. The above remarks notwithstanding, Die Chamoro Sprache is not at all an assemblage of idiosyncrasies and unfounded speculations. On the contrary, Hermann Cos-

240

Thomas Stolz et al.

tenoble has achieved a great deal, and these achievements need to be made explicit. Again for reasons of space, we mention only a few of his insightful contributions to the linguistic understanding of Chamorro. A full-length appraisal of Costenoble‘s work must be postponed until the planned re-edition of Die Chamoro Sprache. First of all, Costenoble allots an entire ―Hauptteil‖ (Costenoble 1940a: 492–520) of his grammar to syntax, which his predecessors mostly did not deem to be necessary. In this chapter on syntax, he meticulously discusses the properties of constituents (arguments as well as adjuncts) and their syntactic functions (subject, object, predicate, attributive phrase, adverbial). Questions of word order are focussed upon repeatedly, including the word order of subject and predicate, differentiated according to main clause and subordinate clause, sentence type, etc. The interaction of pragmatics and word order is treated separately. The problem of the absent copula, comparative constructions and syntactically relevant verbal morphology are also highlighted. Most interestingly, Costenoble concludes his syntactic description with a sub-section devoted to prosodic aspects of utterances (―Satzakzent, Satzton und Satzrythmus‖).14 He is very particular as to the identification of stressed and unstressed syllables (in connection with primary and secondary stress), speech rhythm and style. None of his predecessors had ever cared about these matters. The same holds for the related area of word-based prosody, in which Costenoble (1940a: 9–13) excels. He has a very clear notion of the factors which interact in the realm of suprasegmental phonology and thus succeeds in discriminating them correctly. He demonstrates that quantity distinctions in vowels are dependent upon the interaction of syllable structure and stress.15 Without using the term, he employs a moraic model which assumes a trade-off between vowel length and an (un)filled coda; i.e. open syllables contain long vowels under stress, whereas closed syllables permit only short vowels. At the same time, Costenoble observes that stress and pitch correlate in interesting ways; for instance, in the Guam variety, the stressed syllable displays low pitch (and its unstressed neighbours have high pitch). He also shows that this is different on Rota. His acute observations in this area are at least as valuable as those provided by Topping (1973: 40–47) three decades later. To this remarkably modern treatment of traditionally neglected issues of prosody, one can easily add a variety of other areas in which Costenoble has helped Chamorro linguistics to make progress. Suffice it to mention his discussion of the tense-mood-aspect system (Costenoble 1940a: 293–300) and the intricacies of transitivity in Chamorro (Costenoble 1940a: 246–250 and 307–350). There are still treasures to be claimed.

14

15

―Satzrythmus‖ is another of Costenoble‘s spelling errors. According to the official German orthography of his times, the correct spelling would have been ―Satzrhythmus‖ with an etymologically justified additional . In this way, he solves the riddle posed by Fritz (1903) whose transcription system postulates a quantity correlation with vowels, cf. Stolz (this volume).

Who was the author of Die Chamoro Sprache?

241

4. Conclusion The identity of H. Costenoble can be considered established, although with some remaining fractions of doubt. Hermann Costenoble is the author, not his sister Hertha Costenoble nor his father Hermann L.W. Costenoble. The other H-initial siblings, Hilde Costenoble and Herbert Costenoble, can be ruled out as well. Further, the grammar holds much in store for modern scholars of Chamorro. To be sure, the book was written under rather peculiar circumstances, some of which were presumably detrimental, such as the lapse of two decades between the first draft and the final revision. Nevertheless, the richness of data, information and analyses provided by Hermann Costenoble call for a re-edition of Die Chamoro Sprache. This re-edition is a demanding task for the editors, who must not only translate the bulky text into English but also comment on the author‘s German grammar and style, clarify his references, check his Chamorro examples with competent native speakers, try to identify the sources from which the examples are drawn, and investigate his analyses more thoroughly. The outcome will be an annotated and translated re-edition of Die Chamoro Sprache, in a shape and format that both linguists and native speakers of Chamorro can make use of.

Bibliography Ballendorf, Dirk Anthony (2001): Die Deutschen und die Amerikaner in den Marianen 1899–1914, in: Hiery, Hermann Joseph (ed.), Die Deutsche Südsee 1884–1914. Paderborn: Schöningh, 748–753. Brandstetter, Renward (1915): Die Lauterscheinungen in den indonesischen Sprachen. Luzern: Haag. Burrus, E. J. (1954): Sanvitores‘ grammar and catechism in the Mariana (or Chamorro) language (1668), in: Anthropos 49, 934–960. Chung, Sandra (1998): The design of agreement. Evidence from Chamorro. Chicago, London: The University of Chicago Press. Cooreman, Ann M. (1987): Transitivity and discourse continuity in Chamorro narratives. Berlin, New York: Mouton de Gruyter. Costenoble, Hermann L.W. (1905): Die Marianen, in: Globus 88(1), 4–9; (2), 72–81; (4), 92–94. Costenoble, Hermann L.W. (1908): Einige Bemerkungen über Kulturmethoden und Betriebsformen der tropischen Landwirtschaft, in: Der Tropenpflanzer 12(8), 372–384. Costenoble, H[ermann] (1934): Chamorro language notes, Parts I–III, in: The Guam Recorder 12(143), 303 and 319; 13(1), 22–23. Costenoble, H[ermann] (1936): Philippine language notes, in: Philippine Magazine 33, 495–510. Costenoble, H[ermann] (1937a): The Philippine verb, in: Philippine Magazine 34(1). Costenoble, H[ermann] (1937b): Monosyllabic roots in Philippine languages, in: Philippine Magazine 34(2). Costenoble, H[ermann] (1937c): Philippine languages of today, tracing the original sound in the languages of today, in: Philippine Magazine 39(1). Costenoble, H[ermann] (1940a): Die Chamoro Sprache. ‗s-Gravenhage: Nijhoff. Costenoble, H[ermann] (1940b): Die Lautwandlungen des Tagalog, in: Bijdragen tot de Land-, Taalen Volkenkunde 99, 65–87.

242

Thomas Stolz et al.

Costenoble, H[ermann] (1979): Dictionary of Proto-Philippine. [Translated by Cecilio Lopez, edited by Ernesto Constantino et al.]. Quezon City: Archives of Philippine Languages and Dialects. The Philippine Linguistic Circle. Dempwolff, Otto (1920): Die Lautentsprechungen der indonesischen Lippenlaute in einigen anderen austronesischen Südseesprachen. Hamburg: Boysen. Farrell, Don A. (1991): History of the Northern Mariana Islands. Saipan: Public School System, Commonwealth of the Northern Mariana Islands. Fritz, Georg (1903): Chamorro Grammatik, in: Mitteilungen des Seminars für Orientalische Sprachen an der Friedrich-Wilhelms-Universität 6, 1–27. Fritz, Georg (1906): Von den Marianen, in: Globus 89, 287–289. Fritz, Georg (1908a): Chamorro-Wörterbuch: Deutsch-Chamorro. Berlin: Reimer. Fritz, Georg (1908b): Chamorro-Wörterbuch: Chamorro-Deutsch. Berlin: Reimer. Hardach, Gerd (1990): König Kopra. Die Marianen unter deutscher Herrschaft 1899–1914. Stuttgart: Steiner. Hiery, Hermann Joseph (2001): Zur Einführung: Die Deutschen in der Südsee, in: Hiery, Hermann Joseph (ed.), Die Deutsche Südsee 1884–1914. Paderborn: Schöningh, 1–24. Hiery, Hermann Joseph (ed.) (2001): Die Deutsche Südsee 1884–1914. Paderborn: Schöningh. Hornbostel, Gertrude C. (= Trudis Aleman) (1924): Chamorro names, in: The Guam Recorder 1(3), 16. Ibáñez del Cármen, P. Aniceto (1865): Gramática chamorra. Manila: Ramirez y Giraudier. Kats, J[acob] (1917): Het Tjamoro van Guam en Saipan vergeleken met eenige verwante talen (met opmerkingen van Prof. Mr. Dr. J.C.G. Jonker). ‘s-Gravenhage: Martinus Nijhoff. Lopinot, P. Callistus (1910): Chamorro Wörterbuch enthaltend I. Deutsch-Chamorro, II. ChamorroDeutsch nebst einer Chamorro-Grammatik und einigen Sprachübungen. Hongkong: Typis Societatis Missionum ad Exteros. Rodríguez-Ponga, Rafael (1995): El elemento español en la lengua chamorra (Islas Marianas). Unpublished Ph.D. thesis. Madrid: Facultad de Filología, Universidad Complutense. Rogers, Robert F. (1995): Destiny‘s Landfall. A history of Guam. Honolulu: University of Hawai‘i Press. Safford, William E. (1903): The Chamorro language of Guam – I–IV, in: American Anthropologist 5, 289–311 and 508–529. Safford, William E. (1905): The Chamorro language of Guam V, in: American Anthropologist 7, 305–319. Salazar, Marlies (2009): The enigma of H. Costenoble – the long search for the author of Die Chamoro Sprache, in: Archipel 77, 21–25. Solenberger, Robert R. (1953): Recent changes in Chamorro direction terminology, in: Oceania 24, 132–141. Spennemann, Dirk H. R. (2004): An annotated bibliography of German language sources on the Mariana Islands. Saipan: Division of Historic Preservation. Stolz, Thomas (2007): The Kurze Geschichte der Marianen by Georg Fritz. A commented re-edition, in: Schrader-Kniffki, Martina & Morgenthaler García, Laura (eds.), La Romania en interacción: entre istoria, contacto y política. Ensayos en homenaje a Klaus Zimmermann. Frankfurt: Vervuert, 307–349. Stolz, Thomas (this volume): The Gramática chamorra. Thompson, Laura Maud (1932): Archaeology of the Mariana Islands. Honolulu: Bernice P. Bishop Museum Library. Topping, Donald M. (with the assistance of Bernadita C. Dungca) (1973): Chamorro reference grammar. Honolulu: University of Hawaii Press. Topping, Donald M. (with the assistance of Pedro M. Ogo) (1980): Spoken Chamorro with grammatical notes and glossary. Honolulu: University of Hawaii Press. Von Preissig, Edward R. (1918): Dictionary and grammar of the Chamorro language of the Island of Guam. Washington: Government Printing Office. Zamora, Mario D. & Olsen, Kristin J. (1980): Obituary: Cecilio Lopez 1898–1979, in: American Anthropologist 82(3), 555–556.

List of contributors

SANDRA CHUNG Cowell Academic Services University of California, Santa Cruz Santa Cruz, CA 95064 USA [email protected] BARBARA DEWEIN Universität Bremen Fachbereich 10 Postfach 330440 28334 Bremen GERMANY [email protected] REBECA FERNANDEZ Universidad de Valladolid Av. Huertas del sacramento, 13, 4ºD 24400 Ponferrada (León) SPAIN [email protected] LAWRENCE A. REID 5-1-1-1701 Nyoidani Minoo-shi Osaka 562-0011 JAPAN [email protected] EMILIO RIDRUEJO Departamento de Lengua Española Facultad de Filosofía y Letras Universidad de Valladoloid Plaza del Campus, s.n. 47005 Valladolid SPAIN [email protected]

THOMAS STOLZ Universität Bremen Fachbereich 10 Postfach 330440 28334 Bremen GERMANY [email protected] CHRISTINA SCHNEEMANN Universität Bremen Fachbereich 10 Postfach 330440 28334 Bremen GERMANY [email protected] IWAR WERLEN Universität Bern Institut für Sprachwissenschaft Länggassstrasse 49 CH-3000 Bern 9 SWITZERLAND [email protected] PIERRE WINKLER Amsterdam Centre for Language and Communication (ACLC) University of Amsterdam Verspronckweg 249 2023 BH Haarlem THE NETHERLANDS [email protected]

244 JOHN U. WOLFF Professor Emeritus Linguistics and Asian Studies Cornell University Ithaca, New York 14853 USA [email protected] KLAUS ZIMMERMANN Universität Bremen Fachbereich 10 Postfach 330440 28334 Bremen GERMANY [email protected]

List of contributors OTTO ZWARTJES Universiteit van Amsterdam Romaanse Taalkunde Spuistraat 134 1012 VB Amsterdam THE NETHERLANDS [email protected]