From Latin to Portuguese: Historical Phonology and Morphology of the Portuguese Language [Reprint 2016 ed.] 9781512808957

Citation preview

FROM LATIN TO PORTUGUESE

ORTHOGRAPHIA DA

LINGOA

PORTVGVESA. Obra vtil, Se neceiïaria.afsi perabem fereueralingoa Hefjjanhol, como a Latioa, & quaefquer outra», que da Latina teem origem. f Utm bum trâHtÀt du

ht

cUufuUt.

Felo Licenciado Duarte Nunez do Liâo.

E M

L I S B O N ,

Per IoaodcBarreira impreflor dclRci N . S. M.D.LXXVL TITLE-PAGE OF FIRST TREATISE ON PORTUGUESE SPELLING 1576 By Duarte Nunes de

Ltao

FROM LATIN T O PORTUGUESE Historical

Phonology and

of the Portuguese

Morphology

Language

By E D W I N B. W I L L I A M S Profe.\.wr of University

Romance Languages of Pennsylvania

SECOND

EDITION

Philadelphia UNIVERSITY OF PENNSYLVANIA PRESS 1962

©

1 9 6 2 , by T h e T r u s t e e s of the U n i v e r s i t y of P e n n s y l v a n i a C o p y r i g h t 1 9 3 8 , by U n i v e r s i t y of P e n n s y l v a n i a Press P u b l i s h e d in G r e a t B r i t a i n , India, a n d P a k i s t a n by the O x f o r d U n i v e r s i t y Press L o n d o n , B o m b a y , and K a r a c h i

L i b r a r y of C o n g r e s s C a t a l o g C a r d N u m b e r :

61-18086

P r i n t e d in the U n i t e d States of A m e r i c a

To

LEONORE ROWE WILLIAMS

PREFACE the publication of Cornu's t r e a t m e n t of Portuguese phonology and morphology in Grober's Grundriss (2d ed., 1904—1906), much new material has become available through editions of medieval manuscripts, dialect studies, and investigations of individual problems. B u t no new synthesis of this material has been presented. Nunes and H u b e r have worked in this direction, b u t while t h e y have brought together the results of much research, they have left unsolved m a n y of the m a j o r problems. T h e present book is an effort to present in systematic form the phonological and morphological history of the Portuguese language in the light of the most recent scholarship in the field. T h e author's own previously published studies have been utilized freely, b u t generally in a form somewhat modified or elaborated, and m a n y new studies appear here for the first time. While most of the solutions arrived at apply only to the field of Portuguese, it is hoped t h a t some will prove significant in the broader fields of Hispanic and general Romance philology. I t is also hoped t h a t theories which are but roughly outlined here will inspire other searchers to f u r t h e r speculation and the collection of additional d a t a t h a t may confirm or disprove them and thus lead to definitive solutions. A word or two may be said concerning practical matters. T h e a u t h o r has striven to maintain a rigid line of demarcation between phonology and morphology and to base the latter rigorously on the former with the intervention of no other factor t h a n analogy. A knowledge of Modern Portuguese phonetics is presupposed. Phonetic symbols and diacritical marks are used where necessary to the exposition of phonological and analogical development. M o d e r n dialect forms are quoted simply to show the conSINCE

vii

viii

PREFACE

t i n u a n c e of tendencies found in Old Portuguese and the wider occurrence of changes found in s t a n d a r d M o d e r n Portuguese. T h e a u t h o r asks indulgence for the d o g m a t i c tone of m a n y assertions. R e s e r v a t i o n s are obvious. T o express t h e m a l w a y s would consume too much space. It is a pleasure to acknowledge m y obligations to numerous friends: to Professor J . P. W i c k e r s h a m Crawford for m y first training and sustained interest in R o m a n c e philology; to Professor R o l a n d G. K e n t for advice a n d g u i d a n c e in the solution of m a n y difficult problems; to M r . Alexandre de S e a b r a for information on special points in M o d e r n P o r t u g u e s e ; to Professor Albert C. B a u g h for constant e n c o u r a g e m e n t and helpful suggestions; and to m a n y g r a d u a t e s t u d e n t s over a period of more t h a n ten y e a r s , who b y cold logic or w a r m flashes of insight in classroom discussions h a v e step b y step helped me forward t o w a r d m y goal. T h e scholars on whose writings I h a v e d r a w n most freely are Leite de Vasconcellos, N u n e s , M e y e r - L i i b k e , and G r a n d g e n t , but a glance a t the bibliography will show to w h a t length this list m i g h t be extended. I a m especially indebted to the F a c u l t y R e search C o m m i t t e e of the U n i v e r s i t y of P e n n s y l v a n i a for several grants, and p a r t i c u l a r l y for one which m a d e possible a s u m m e r of s t u d y and investigation in P o r t u g a l . E. B. W .

PREFACE TO THE SECOND EDITION T h e new discoveries and the principles based on them that were set forth in the first edition of this book in 1938 (such as syncope in Portuguese, particularly as contrasted with Spanish—§ 51-§ 5 9 ; the fate of intervocalic n in all its ramifications—§ 7 8 ; assimilation, dissimilation, and contraction of vowels in hiatus—•§ 9 9 ; the final -ao—§ 157; radical-changing verbs—§ 1 7 4 ) have not been challenged by reviews or by the research that has been carried out since that time. All the reviews have been in general favorable. What they have criticized has been the restricted scope of the book to the study of historical phonology and morphology and heterogeneous minor matters, varying from one reviewer to another, that have no bearing on the fundamental findings and conclusions. In the present edition, accordingly, there are no basic changes, but some adjustments have been made in the light of the reviews, the bibliography has been brought up to date, and all available resources have been used to support and strengthen the principal arguments. It is hoped that the book will continue to serve as a guide for the study of the development of Latin into Portuguese. E. B. W .

ix

CONTENTS Page PREFACES

vii

PHONETIC SYMBOLS, DIACRITICAL

M A R K S , AND OTHER SIGNS

xiii

INTRODUCTION

Vulgar Latin Portuguese Portuguese Orthography

1 10 19

PHONOLOGY

T o n i c Vowels P r e t o n i c Vowels F i n a l Vowels Secondary Tonic Posttonic Penult I n t e r t o n i c Vowels Initial Short Consonants Initial Consonant G r o u p s Intervocalic Short Consonants Intervocalic Long Consonants Medial Consonant Groups Final Consonants General Phonological P h e n o m e n a Sporadic Changes

29 40 46 51 52 55 59 62 65 74 75 92 94 101

MORPHOLOGY

Nouns Adjectives Numerals Articles Pronouns G e n e r a l P h e n o m e n a of V e r b a l I n f l e x i o n Tense Endings Radical-changing Verbs Unclassifiable Verbs A B B R E V I A T I O N S AND L I S T OF B O O K S AND A R T I C L E S SUPPLEMENTARY

SELECTIVE

BIBLIOGRAPHY

INDEX

117 128 133 139 143 161 180 208 218 CITED. .

243 268 273

xi

PHONETIC SYMBOLS, DIACRITICAL MARKS, A N D OTHER SIGNS These equivalents are obviously only approximate. [ a (open a) English a in [m] English m father [ n ] English n in now [ * (close a) English a in [ i] ] (velarized n) English n about in bank (nasal a) close a nasal[ ji] (palatalized n) French [S ized gn in enseigner English b [ o ] (open o) French o in école [b [ o ] (close o) English o in note [ b Spanish b in cabo [Ô] (nasal o) close o nasalized English d [d [ p ] English p [d Spanish d in nada [ e (open e) English e in met [ r ] slightly trilled with tip of tongue [e (close f) French e (neutral e) French e in [ s ] English J tenir [ s ] ( c a c u m i n a l s) p r o nounced with tip of [ 5 (nasal e) close e nasalized tongue raised to roof of m o u t h [f English / English g in good [ t ] English t U (open i) English i in [ u ] (close u) English oo m food [! [ u ] (nasal u) close u nasalized (close i) English i in [ v ] English v [i machine [w] (consonantal u) English (nasal i) close i nasalw in well [ï ized [ z ] English z [ z ] ( c a c u m i n a l z) p r o [ j (yod) English y in yet English k nounced with tip of [k tongue raised to roof [1 English I in look (velarized /) English I of m o u t h [» in old [ S ] English sh in shall [X] (palatalized /) Italian gli [3 ] French j in paglia [ x ] G e r m a n ch. in ach xiii

PHONETIC SYMBOLS

XIV

p a I e j

same same same same same

as as as as as

[a] [tf] [ e] [e] [i]

g 0 u 1 u

same same same same same

as as as as as

[o] [o] [u] [j ] [w]

T h e long and short marks ~~ and w are used to indicate the quantity of Classical Latin vowels. T h e acute accent is used to indicate stress except on close a, e, and o in Portuguese words, where the circumflex accent is used according to the rules of the nova ortografia. T h e sign > means " b e c o m e s " and the sign < means " c o m e s from." Both are used to indicate analogical change as well as phonological change. Words not found in Harpers' Latin Dictionary are either marked with an asterisk or accompanied by an indication of their source in Vulgar Latin or Medieval Latin, even though they may be of late occurrence. In accordance with common practice, hypothetical etyma, i.e., words marked with an asterisk, are generally given in the form they would have had in Classical Latin.

FROM LATIN TO PORTUGUESE

I N T R O D U C T I O N

VULGAR LATIN 1.

KNOWLEDGE

OF V U L G A R

LATIN.

Latin as a living language was subject to constant change. While the language of the cultivated classes (Classical Latin) became more and more uniform under the stabilizing influence of culture and learning, the language of the people (Vulgar Latin) became more and more diversified as it spread with the expansion of the vast R o m a n E m pire. Classical Latin became a dead language while Vulgar Latin developed into the so-called Neo-Latin or Romance languages. N o great literature has been handed down to us to a t t e s t t h e existence of Vulgar Latin. Our knowledge of it is derived from the following sources: a) popular elements, of intentional or accidental origin, in Classical and Medieval L a t i n ; b) linguistic observations in Classical and Medieval L a t i n ; c) Latin elements in the languages of the peoples with whom the Romans came in contact; d) the Romance languages. Vulgar Latin is, therefore, a language reconstructed from heterogeneous fragments and largely on the basis of hypothesis. 2.

CLASSICAL

LATIN

ACCENT.

T h e place of the accent in Classical Latin is determined by q u a n t i t y , according to the following principles: a) In words of two syllables the penult is accented: homo. b) In words of three or more syllables the penult is accented if it is long, t h a t is, if it contains a long vowel: imperatorem, a diphthong: incautum, or a short vowel followed by two or more consonants: intendo (except a m u t e plus I or T: tenebras). 1

2

INTRODUCTION

[§ 2-5

c) In words of three or more syllables, the antepenult is accented if the penult is short: hominem. 3.

VULGAR

LATIN

ACCENT.

T h e accent falls on the same syllable in Vulgar Latin as in Classical Latin, with the following exceptions: a) In words of three or more syllables the penult is accented if it contains a vowel which was short in Classical Latin, followed by a mute plus / or r: tenebras. b) In compound verb forms of three syllables which would be accented on the antepenult in Classical Latin, the penult is accented; in other words, the accent falls on the radical vowel of the simple verb and not on the prefix: *recîpit, not récif it. Usually the vowel of the simple verb is restored: réficit > *rejâcit. c) Where an i or e, accented in Classical Latin, is in hiatus with a following short vowel, the accent shifts in Vulgar Latin to this following vowel, the i or e afterwards becoming a yod: mulierem > *mul\ere; jiliolum > *filiolu. 4. UNACCENTED

WORDS.

With the intensification of stress accent in Vulgar Latin, many unemphatic words lost their accent and became attached as proclitics and enclitics to other larger and more emphatic words. This was particularly true of personal pronouns used as direct and indirect objects of verbs and of verbs used as auxiliaries. The vowels of these words developed regularly as pretonic initial or unaccented final vowels. Dissyllabic words of this sort often became monosyllabic. 5.

PORTUGUESE

ACCENT.

T h e accent falls on the same syllable in Portuguese as in Vulgar Latin.

VULGAR

§6-7] 6.

VULGAR

LATIN

TONIC

LATIN

3

VOWELS.

I t is thought t h a t the v a r i a t i o n of a given vowel in Classical Latin was one of q u a n t i t y (cf. t h e difference in M o d e r n G e r m a n between t h e a of Strasze and the a of Gasse), while the corresponding variation in Vulgar L a t i n was one of quality, with the exception of the vowel a, in which no qualitative distinction was made. T h u s long o (o) and short o (o) of Classical L a t i n b e c a m e close o (o) and open o (p) respectively in V u l g a r L a t i n . T h e development of Classical Latin vowels into Vulgar L a t i n vowels is shown in this table. Classical L a t i n

3'}

Vulgar L a t i n

ae J e x oe i

?

i

ò

Q

Ü

u au

a

au

A. T h e frequent change of e to I in Medieval L a t i n d o c u m e n t s (cf. M T , 3 1 - 3 3 ) , which is c o n t r a r y t o the change indicated in the above table, is purely o r t h o g r a p h i c and probably arose through confusion of the two letters because of an awareness t h a t t h e y both stood for one a n d t h e same sound. T h e change of o to u arose in the same way. 7.

VULGAR

LATIN

ATONIC

VOWELS.

T h e Vulgar L a t i n vowels listed in § 6 were further simplified in the unaccented position: a, i, u, e, and o remained as in the accented position; £ b e c a m e e, g b e c a m e o, and

INTRODUCTION

4

[§7-9

initial au, followed b y a syllable c o n t a i n i n g a c c e n t e d u, b e c a m e a. A. Grandgent (GVL, § 228) judges from "subsequent developm e n t s " that initial unaccented g did not close to o. He may have in mind the situation in French where we find two different products of initial unaccented o, e.g., porter (from portàre) and journée (from *diurnàta). But Dauzat has attempted to overcome this difficulty by explaining that the o of such forms as porter is due to regression or the operation of analogy (Dauzat, § 100). But see § 99, 5 A. Cf. M - L , Gram, I, § 353. 8.

SYNCOPE

IN V U L G A R

LATIN.

1. T h e vowel of t h e p e n u l t of p r o p a r o x y t o n e s a n d t h e vowel of t h e i n t e r t o n i c syllable generally fell in V u l g a r L a t i n in t h e following cases: a) w h e n followed b y I or r ; b) w h e n preceded b y / or r a n d followed b y d, m, or p; c) w h e n preceded b y s a n d followed b y t; d) w h e n preceded b y a labial; e) in a few unclassifiable words like dïgïtum a n d frïgïdum. Cf. M - L , G r a m , I, § 29; G V L , §§ 2 3 1 - 2 3 8 . 2. I n t h e V u l g a r L a t i n of t h e P o r t u g u e s e t e r r i t o r y this vowel fell m u c h less f r e q u e n t l y t h a n elsewhere, p a r t i c u larly w h e n followed b y I or r, a n d w h e n preceded b y a l a b i a l ; it p r o b a b l y did n o t fall in dïgïtum a n d frïgïdum. A. The groups ab + consonant and av -F- consonant, arising in Vulgar Latin through the fall of the penult of proparoxytones or the vowel of the intertonic syllable (GVL, § 236), were rare in Portuguese territory. 9.

CLASSICAL

LATIN

AND

VULGAR

LATIN

YOD.

1. Initial u n a c c e n t e d i before a vowel a n d i n t e r v o c a l i c u n a c c e n t e d i were c o n s o n a n t a l in Classical L a t i n . Whether initial or n o t , u n a c c e n t e d i a n d e before a vowel g e n e r a l l y b e c a m e c o n s o n a n t a l in V u l g a r L a t i n . 2. V u l g a r L a t i n y o d preceded b y a c o n s o n a n t s o m e t i m e s fell.

§10] 1 0 . CONSONANTAL

VULGAR

LATIN

5

U.

1. One and the same character, v, stood for a c o n s o n a n t [w] and a vowel [u] in Classical L a t i n . 2. T h e consonant sound occurred in the following positions: a) initial before a vowel, b) between two vowels, and c) sometimes when preceded by a c o n s o n a n t and followed b y a vowel. E a r l y in the E m p i r e this sound b e c a m e a bilabial fricative, i.e., [b] in some regions and a dentilabial, i.e., [v] in others ( G V L , § 322). T h e sound [w] also occurred as t h e second element of a diphthong, where it generally remained unchanged in Vulgar L a t i n , except in initial au, followed b y a syllable containing accented u, where it fell. 3. T h e vowel sound occurred in the following positions: a) initial before a consonant, b) between two consonants, c) final after a consonant, and d) sometimes when preceded by a consonant and followed by a vowel. In the last of these positions it b e c a m e a c o n s o n a n t , i.e., [w] toward the end of the Vulgar L a t i n period ( G V L , § 3 2 6 ) . T h i s took place too late for the change to [b] or [v] (as in 2 c above), although isolated examples are found in several R o m a n c e languages, e.g., valuissel > O P t g . valvesse; paruit > I t . parve; januarium > F r . janvier. T h i s late [w] fell when followed by unaccented o or u, when preceded by a guttural and followed by accented o or u, and in a few unclassifiable cases ( G V L , § 2 2 6 ) . 4. As the letter v, preceded by a c o n s o n a n t and followed by a vowel, could stand for the vowel sound and the consonant sound in Classical Latin (2 c and 3 d above), it is possible to distinguish the two sounds in a given word only by determining by means of scansion the n u m b e r of syllables it contains. F o r example, if volvit is found to have two syllables, the second v (as well as the first) stands for [w]; if it is found to have three syllables, the second v

6

INTRODUCTION

[ § 10-13

stands for [u], and the word may be written voluit (perf. of velle), as is the practice in t h e modern o r t h o g r a p h y of Classical Latin. 11.

PROSTHETIC

e.

Because initial s before a consonant was found difficult to pronounce, an e was prefixed to it and came to form p a r t of the word. 12. VULGAR

LATIN

CONSONANTS.

1. C followed by e or i, by moving forward toward the teeth, went through the following stages: [k] > [kj] > [tj] > [ts], 2. G followed by e or i became [gj] and then a simple yod. 3. C + yod became [tsj], A century or two earlier, t -(- yod also became [tsj] but this sound had changed to [ts] by the time t h a t c + yod became [tsj]. See GVL, §§ 277-278. G + yod and d + yod became a simple yod. 4. Intervocalic surds and surds preceded by a vowel and followed by I or r became voiced. Intervocalic b and b preceded by a vowel and followed by / or r became [b]. 5. Final / and final d preceded by a vowel or a consonant fell. 6. T h e group rs became ss; ps became ss~, pt became tt; net became nt (except in Gaul); nf b e c a m e / ; ns became s; x became s before a consonant and in the prefix ex- sometimes before a vowel; and the final consonant of a prefix was usually assimilated to the initial consonant of the word to which it was a t t a c h e d . 7. Double consonants remained long as in Classical Latin. 8. T h e following consonants became silent: h] final m except in monosyllables, where it became n. 9. T h e sound of v changed from [w] to [v]. See § 10, 2. 1 3 . MORPHOLOGICAL

CHANGES

IN V U L G A R

LATIN.

1. Nouns, a) T h e five declensions of Classical Latin were reduced to three, nouns of the fourth declension

§13]

VULGAR LATIN

7

changing to the second and nouns of the fifth declension changing to the third (except a few which changed to the first, e.g., dies and rabies). b) T h e neuter gender disappeared: neuter singulars became masculine and neuter plurals became feminine singulars of the first declension. Feminine nouns of the second declension became masculine and feminine nouns of the fourth declension became masculine or changed to the first declension. c) All cases finally disappeared except the nominative and an oblique case formed by the fusion of the accusative and ablative (and in some nouns the dative). And in the first declension, the nominative was replaced by the accusative. Some nouns of the third declension, which had a shift of accent from the nominative to the oblique cases, formed a new nominative with the accent on the same syllable as in the oblique cases. 2. Adjectives, a) T h e declension of adjectives changed in accordance with the declension of nouns. However, the neuter singular was retained to express in the abstract the quality denoted by the adjective. b) There was a tendency of adjectives of the third declension to change to the type based on the first and second declensions. c) T h e endings of the comparative and superlative of adjectives and adverbs began to disappear and to be replaced by periphrastic expressions (§ 14, 3). 3. Pronouns, a) T h e nominative, dative, and accusative of most pronouns were retained. In a few pronouns the genitive was retained. The neuters of many pronouns were preserved; they were used to refer, not to single words, for neuter nouns had disappeared, but in an indeterminate way to whole previous statements or propositions. b) Some pronouns developed into two different forms according as they were accented or unaccented. 4. Verbs, a) The four conjugations of Classical Latin

8

[§ 13-14

INTRODUCTION

survived in Vulgar L a t i n b u t m a n y verbs shifted from one conjugation to another. b) T h e future indicative fell into disuse (for expressions which replaced it, see § 14, 5). T h e imperfect subjunctive was replaced by the pluperfect subjunctive; it fell into disuse except in the Portuguese territory, where it assumed a new function (§ 158, 2). T h e perfect subjunctive and the future perfect indicative fused into a new tense (§ 14, 6). T h e imperative lost most of its forms except the second singular and plural of the present. T h e endings of the passive voice were lost; deponent verbs, accordingly, took on active endings. T h e perfect infinitive, the supine, the future active participle and the gerundive disappeared. A new form in -tortus came t o be used as a gerundive. 14.

CHANGES

IN

SYNTAX

IN

VULGAR

LATIN.

1. I t is probable t h a t the phonological changes (chiefly the loss of final consonants and the loss and weakening of unaccented vowels), brought about by an increased stress accent (§ 16, 2), precipitated the breakdown of the morphological system of Classical Latin, which was thus rendered unfit for the needs of a highly synthetic syntax. 1 Inflection was replaced by periphrasis. Analysis took the place of synthesis. And word order came to assume an all-import a n t role in syntax. T h u s we find a much greater use of prepositions, auxiliary verbs and other periphrastic forms of expression in Vulgar Latin. 2. Ad with the accusative replaced the dative. De with the ablative replaced the genitive; and with the fall of final m the accusative and ablative became identical in form and function. 1

Vossler ( T h e Spirit

of Language

in Civilization,

London, 1932, Chapter I V )

argues, on the contrary, t h a t the synthetic syntax of Classical Latin

became

atrophied through disuse as a result of a new mode of thinking, a new

Weltan-

schauung.

§14]

VULGAR L A T I N

9

3. N e w periphrastic c o m p a r a t i v e s of a d j e c t i v e s a n d a d v e r b s were formed with magis a n d plus. 4. P r o n o u n s were used more c o m m o n l y t h a n in Classical L a t i n . Is and idem fell i n t o disuse a n d ille b e c a m e t h e regular p r o n o u n of the t h i r d person. A t t h e same t i m e ille began to f u n c t i o n as definite article while unus c a m e t o be used more widely as an indefinite article. 5. N e w periphrastic f u t u r e tenses were f o r m e d with t h e infinitive and t h e present tense of t h e following auxiliaries: debere, habere, ire, vadere, velle and venire. T h e tense f o r m e d w i t h the infinitive and t h e p r e s e n t tense of habere c a m e i n t o c o m m o n e s t use. A n o t h e r tense, f o r m e d with t h e infinitive and the i m p e r f e c t indicative of habere, a p p e a r e d along with t h e f u t u r e , first p r o b a b l y as a p a s t f u t u r e tense in indirect discourse (as t h e infinitive plus accusative cons t r u c t i o n was being avoided), a n d later with m a n y of t h e uses of the conditional in t h e R o m a n c e languages. 6. W h e n used with the p a s t participle, t h e p r e s e n t indicative and s u b j u n c t i v e of habere f o r m e d a new perfect indicative and a new perfect s u b j u n c t i v e respectively, while t h e old perfect indicative c o n t i n u e d in use. T h e old p e r f e c t s u b j u n c t i v e and the f u t u r e perfect indicative fused i n t o a single tense, which was used as a f u t u r e indicative or subj u n c t i v e (Tempuslehre, § § 8 - 1 1 ) . A new p l u p e r f e c t indicative was f o r m e d with t h e i m p e r f e c t indicative of habere plus the p a s t participle, while a new p l u p e r f e c t s u b j u n c t i v e was formed with t h e old p l u p e r f e c t s u b j u n c t i v e of habere plus the p a s t participle, t h e old p l u p e r f e c t s u b j u n c t i v e h a v i n g become t h e i m p e r f e c t s u b j u n c t i v e in Vulgar L a t i n in place of t h e Classical L a t i n i m p e r f e c t s u b j u n c t i v e . 7. T h e forms of the passive voice were replaced b y a new passive formed with esse plus t h e p a s t participle. A t t h e same time, the passive was avoided b y using t h e reflexive construction or homo as a n indefinite p r o n o u n . 8. T h e s u b j u n c t i v e had fewer f u n c t i o n s ; t h e y were m o r e like those of t h e R o m a n c e languages t h a n like those of

10

[§ 14-16

INTRODUCTION

Classical Latin. Some of the lost forms of the imperative were replaced by the subjunctive. T h e infinitive replaced the supine and the subjunctive clause introduced by ut. On t h e other hand, the infinitive plus accusative construction was avoided and replaced by a clause introduced by quia, 15.

quod, quoniam WORDS

AND

or ut.

WORD

FORMATION

IN

VULGAR

LATIN.

1. T h e r e was a large body of common words which were t h e same in Classical Latin and Vulgar L a t i n . M a n y words are found in Vulgar Latin which had developed from Classical Latin according to the regular laws of semasiology, b u t a surprisingly large n u m b e r of meanings seem to follow t h e law of specialization. M a n y Classical Latin words disappeared entirely. New words of uncertain origin app e a r e d ; some of these were probably Classical Latin words which had never been used in texts t h a t are now extant. And m a n y more new words were derived from Greek, Celtic and G e i m a n . N e w verbs were formed f r o m nouns, adjectives, present participles, and past participles while new nouns were formed from verbs. 2. Phonological attrition, due to the increased stress accent of Vulgar Latin, sometimes reduced words to a single syllable. These words and others were reinforced by t h e use of suffixes and prefixes of all kinds. T h u s augmentatives, diminutives and inchoatives were used w i t h o u t their special meaning and with no other effect t h a n to give more body to the original word. W h e r e this could not be done the word often disappeared to m a k e w a y for a more robust competitor. PORTUGUESE 16.

FROM

VULGAR

LATIN

TO

PORTUGUESE.

1. T h e differentiation of Vulgar L a t i n f r o m one region t o a n o t h e r , which finally resulted in its t r a n s f o r m a t i o n into

§16]

PORTUGUESE

11

t h e several R o m a n c e l a n g u a g e s , is t h o u g h t t o h a v e been d u e t o t h e following c a u s e s : a) t h e relative geographic isolation of one g r o u p f r o m a n o t h e r , b) t h e d e v e l o p m e n t of s e p a r a t e political u n i t s , c) t h e v a r i a t i o n of c u l t u r a l a n d e d u c a t i o n a l c i r c u m s t a n c e s , d) t h e period of r o m a n i z a t i o n , e) dialectal differences in t h e l a n g u a g e of t h e Italic colonists, f) t h e original linguistic s u b s t r a t a , a n d g) s u b s e q u e n t linguistic s u p e r i m p o s u r e s . P r o b a b l y t h e m o s t i m p o r t a n t cause of d i f f e r e n t i a t i o n was t h e intensified stress a c c e n t , s u p e r i m p o s e d , as it were, u p o n t h e V u l g a r L a t i n of I t a l y , G a u l a n d t h e I b e r i a n p e n i n s u l a in v a r y i n g degrees b y t h e i n v a d i n g G e r m a n i c races. 2. I t is believed t h a t Classical L a t i n always had a stress a c c e n t , a n d t h a t a p i t c h a c c e n t , which never a f f e c t e d p o p u l a r speech, w a s i n t r o d u c e d t o w a r d t h e middle of t h e second c e n t u r y B . C . a m o n g t h e highly e d u c a t e d classes b y G r e e k t e a c h e r s , p r o n o u n c i n g L a t i n in t h e i r own f a s h i o n ( K e n t , § 66, 1). F i v e or six c e n t u r i e s later t h e stress a c c e n t of p o p u l a r speech w a s g r e a t l y intensified in the m o u t h s of i n v a d i n g G o t h s , w h o a c c e n t e d L a t i n with t h e g r e a t e r stress a c c e n t c h a r a c t e r i s t i c of t h e i r own language. One of t h e results of this intensified stress a c c e n t was t h e increased s y n c o p e of t h e vowel of t h e p o s t t o n i c p e n u l t a n d t h e i n t e r tonic syllable b e t w e e n c e r t a i n pairs of c o n s o n a n t s , w h i c h t o o k place in V u l g a r L a t i n (§ 8). A. The stress accent of popular Latin was sufficiently strong to cause some syncope before the advent of the Goths (CPh, II, 454). 3. A d d i t i o n a l G e r m a n i c invasions ( F r a n k s , B u r g u n d i a n s , L o m b a r d s , etc.) b r o u g h t a b o u t f u r t h e r intensification of t h e stress a c c e n t , a n d w i t h t h e rise of t h e R o m a n c e languages, s y n c o p e of t h e vowel of t h e p o s t t o n i c p e n u l t a n d t h e i n t e r t o n i c syllable b e c a m e a general p h e n o m e n o n no longer limited t o t h e special positions in which it o c c u r r e d in V u l g a r L a t i n . B u t t h e s e a d d i t i o n a l G e r m a n i c i n v a s i o n s

12

INTRODUCTION

did not reach the territory where Portuguese was to develop. Aside from Visigoths and Suevi no Germanic tribes ever settled in this territory and Visigoths and Suevi left but slight traces of their s t a y (BF, II, 108-109). T h e linguistic result was t h a t there was less stress accent than in other Romance territory and accordingly, less syncope (§ 53 and § 58). The failure of ( and g to diphthongize and the slow formation of yod (§ 87, 2) and of Ptg. u ( § 9 3 , 1 a ) are further evidence of a weaker stress accent. 1 The Visigoths and Suevi left only about t w e n t y words in the Portuguese language ( R F E , X I X , 234—238). The relatively large number of Germanic place-names in northern Portugal and in Galicia is due to the settlement of the Visigoths in this territory after the year 700, t h a t is, after the transition to Romance and after their complete amalgamation with the earlier Roman and Celtic inhabitants, when they were fleeing from the south and centre of Spain from the Moors and had to protect themselves in fortified towns, contrary to their previous custom elsewhere (RG, I, 361). T h e separate Romance which developed in the south among the Mozarabs ( R L , X I , 354) was entirely free of Germanic influence; hence the especial fondness of the people of the south even today for proparoxytones (BHi, VII, 194) and the impression they give of singing when they speak (Esquisse, p. 154, n. 1). Thus while certain characteristics of Old Portuguese, such as the fall of intervocalic I and n, arose in the north, the resistance to syncope, a far more distinctive characteristic, was stronger in the south (cf. Ent, 278). A. The remark of Gamillscheg: "No puede hablarse, en lo que se refiere al desarrollo fonètico, de una influencia del gòtico sobre el ibero-románico" (RFE, XIX, 260) and his reference to Brüch's study (RLiR, II, 66 ss.) have no bearing on the question of 1 For a similar h y p o t h e s i s applied to N e a p o l i t a n and other I t a l i a n dialects, see V a u g h a n , 11-14. See also Pope, § 223.

PORTUGUESE

§ 16-18]

13

accent as Briich does not discuss it a n d as f a r as p h o n o l o g y is concerned deals only with G e r m a n i c h a n d w. 17.

OLD

AND

MODERN

PORTUGUESE.

Vulgar L a t i n along the western c o a s t of the Iberian peninsula, freer from G e r m a n i c stress accent than anywhere else, freer especially t h a n in the rest of the peninsula, grew into P o r t u g u e s e . N o one knows e x a c t l y when it ceased to be Vulgar L a t i n a n d began to be Portuguese. T h e earliest documents in P o r t u g u e s e a p p e a r e d a t the end of the twelfth century and m a r k the historical beginning of Old P o r t u g u e s e . F o r four centuries the language underwent m a n y changes. T h e m o s t i m p o r t a n t of these was the m a r k e d intensification of stress accent which occurred in the sixteenth century. T h i s is shown by the increased syncope found in verse (§ 54 and § 59) and b y the tendency toward greater word individuation (§ 118). T o w a r d the end of the sixteenth century nearly all of the distinctive characteristics of Old P o r t u g u e s e had d i s a p p e a r e d ; the l a n g u a g e had become in all essentials the s a m e as the l a n g u a g e of t o d a y . A. I n c r e a s e d stress accent and increased word individuation developed in P o r t u g u e s e in p r a c t i c a l l y the s a m e period as decreased stress accent a n d decreased word individuation (i.e., liaison) in French. Cf. P o p e , § 170 and § 223. B. T h e history of the P o r t u g u e s e l a n g u a g e has been divided b y some into three p e r i o d s : a) the Old or N a t i o n a l P e r i o d — f r o m the twelfth to the sixteenth century, b) the M i d d l e or Classical P e r i o d — f r o m the sixteenth to the eighteenth century, a n d c) the F r e n c h or A r c a d i a n P e r i o d — f r o m the eighteenth century to the present d a y . See L P , I, 192-193. 18.

L E A R N E D W O R D S , D O U B L E T S AND R E G R E S S I V E

WORDS.

1. F r o m the earliest times new L a t i n words have entered into P o r t u g u e s e , first through the church and the law, later through the work of scholars and men of letters, a n d

14

INTRODUCTION

[§18

still l a t e r t h r o u g h s c i e n c e . T h e s e l e a r n e d o r s e m i - l e a r n e d w o r d s h a v e n o t u n d e r g o n e all t h e c h a n g e s w h i c h p o p u l a r w o r d s h a v e u n d e r g o n e , first, b e c a u s e t h e y h a v e o f t e n been t a k e n i n t o t h e l a n g u a g e a f t e r c e r t a i n c h a n g e s h a d ceased t o t a k e p l a c e , a n d s e c o n d , b e c a u s e of a c o n s c i o u s e f f o r t to p r e s e r v e t h e i r L a t i n f o r m . See M e i l l e t , 3 1 9 A. Late learned words with popular changes in imitation of popular words are sometimes found in dialects, e.g., politigo for politico, inorante for ignorante, the suffix -airo for -ario (Esquisse, § 60 d; RL, X I , 141 and 278; RL, X I I , 307). B. Literary Portuguese probably contains fewer learned words t h a n literary Spanish. 2. S o m e t i m e s a l e a r n e d o r s e m i - l e a r n e d w o r d a l r e a d y e x i s t e d as a p o p u l a r w o r d ; t h e t w o f o r m s a r e t h e n called d o u b l e t s , e.g., pelago a n d pego; artigo a n d artelho. Borrowings f r o m d i a l e c t s a n d o t h e r l a n g u a g e s m a y a l s o b e c o m e d o u b l e t s , e.g., arena ( f r o m Sp.) a n d areia. 3. S o m e t i m e s a p o p u l a r w o r d w a s m o d i f i e d o r r e p l a c e d b y t h e L a t i n w o r d f r o m w h i c h it o r i g i n a l l y c a m e ; t h e n e w w o r d is called a r e g r e s s i v e w o r d . T h u s O P t g . seenqo w a s r e p l a c e d b y silencio, vesso b y verso, t h e suffix -do b y -ano, e.g., romao b y romano. T h i s is a f o r m of c o n t a m i n a t i o n . A. By a conscious or unconscious striving for discrimination, regression often restored a difference between two words which regular phonological development had destroyed, e.g., both venam and vela regularly became vea in Old Portuguese; the regression to vela has restored the distinction in form. I n some cases regression did not t a k e place, e.g., both *fidare (for jidere) and filare became fiar, which survives with the meanings of b o t h originals. In other cases the resultant form has lost one of its meanings, i.e., one of the two original words has disappeared, e.g., *adcalescere and *adcadescere both became aquecer in Old Portuguese, which survives in M o d e r n Portuguese with only t h e meaning of *adcalescere. Cf. BSC, X, 814. 4. C o n t r a r y t o t h e a d v i c e of K i n g E d w a r d ( D o m D u a r t e ) in his " L e a l C o n s e l h e i r o " e a r l y in t h e fifteenth c e n t u r y

PORTUGUESE

§ 18-19]

15

t h a t the writer " n o m ponha pallavras latinadas nem d'outra linguagem," he himself introduced m a n y Latin words, spellings, and constructions in this very work, and the practice became general a c e n t u r y later, not only in translations f r o m Latin b u t in original works. Devotion to Latin led some poets to write exclusively in t h a t language. 19.

SPANISH

INFLUENCE.

In the first half of the fifteenth century, Spanish poets such as t h e M a r q u é s de Santillana gave to the Spanish language the high position in p o e t r y t h a t Alfonso el Sabio had won for it in prose. I t was natural, then, t h a t Portuguese poets, who had a b a n d o n e d the tradition of their early lyric, should t u r n to Spanish for t h a t contact with t h e Renaissance which had thus far been denied t h e m . Accordingly, we find them in the second half of the century writing much of their verse in Spanish (in the " Cancioneiro G e r a l " ) , a n d such distinguished poets as Gil Vicente, Sa de M i r a n d a , a n d Camôes continued t o do so in the sixteenth century. M a n y violent a t t a c k s against this practice were m a d e b y g r a m m a r i a n s such as Fernâo de Oliveira and poets such as Antonio Ferreira. In his Elegy on the death of Ferreira, Diogo Bernardes was able to say, Pois dando à patria tantos versos raros, um sô nunca lhe deu em lingua alheia. In the defense of Portuguese against Spanish, efforts were made to prove t h a t Portuguese was superior t o all t h e other R o m a n c e languages because it was more like Latin t h a n a n y of them, or as some writers p u t it, because it represented a s t a t e of Latin less corrupt t h a n any of the others. T h i s was the a r g u m e n t of the " D i a l o g o em louvor da nossa l i n g u a g e m " of J o â o de Barros, published in 1540, and of t h e " D i a l o g o em defensam da lingua p o r t u g u e s a " of Pedro de Magalhâes de G a n d a v o , which appeared in

16

INTRODUCTION

[§ 19-20

1574, four y e a r s before Henri Estienne's " D i a l o g u e " against the italianisants. T h e argument was continued and elaborated in the following century in m a n y books and treatises, the most notable of which was the " B r e v e s louvores da lingua portuguesa: com notaveis exemplos da muita semelhança que tem com a lingua l a t i n a " of Alvaro Ferreira de Vera, published in 1631. T h a t these works were not without effect is revealed in the efforts of the Portuguese to m a k e their language as unlike Spanish as possible, a tendency a l r e a d y noted b y D u a r t e Nunes de Leáo in his "Origem da Lingoa P o r t u g u e s a " (N-L, Origem, 126), published in 1606. 20.

FRENCH

INFLUENCE.

T h e Romance language which had the greatest influence on Portuguese from the earliest times was French (Rom, II, 293). T h e infiltration of French manners and customs and the introduction of French coins and commercial products into Portugal had begun in the tenth and eleventh centuries ( B S C , VII, 188-190). It was Afonso Henriques, the son of Count H e n r y of B u r g u n d y , who by wresting the title of king from Alfonso V I I of Castile and León in 1143 and by taking Lisbon from the Moors in 1147, established the House of B u r g u n d y , the first d y n a s t y of Portuguese monarchs, and laid the foundations of a new country in the lower v a l l e y of the T a g u s . Frenchmen came as pilgrims to the shrine of Santiago in Galicia, as soldiers of fortune to help fight the Moors, and as monks from the Benedictine A b b e y of Cluny. After the Sixty Years of C a p t i v i t y ( 1580— 1640), begun under the ominous figure of Philip II of Spain, France became a friend and a l l y of Portugal in the wars of John IV against Spain. And now again Frenchmen came as courtiers, statesmen, scholars, and soldiers. To the spread of the doctrine of Boileau was added in the second half of the eighteenth century the influence of the Encyclopedists and of French scientific thought.

§ 20-22]

PORTUGUESE

17

T h e influence of F r e n c h in words and expressions has continued steadily until it has become a cause of alarm, and m a n y modern Ferreiras h a v e risen to defend t h e p u r i t y and integrity of their m o t h e r tongue. B u t this influence has been an i m p o r t a n t force in a c c e n t u a t i n g the difference between Portuguese and Spanish. 21.

PORTUGUESE.

T o the often asked question, " W h y did a separate language develop on the western e x t r e m i t y of the Iberian p e n i n s u l a ? " , the answer would seem to b e : a) because of geographic isolation b y high plateaus and waste land, b) because of less G e r m a n i c influence in the critical f o r m a t i v e period, c) because of political independence as early as the middle of the twelfth c e n t u r y under a leader of e x t r a ordinary prowess and s t a t e s m a n s h i p , d) because of the ultimate triumph, in spite of literary and political opposition, of the spirit of those s i x t e e n t h - c e n t u r y critics who believed t h a t their language was w o r t h y of preservation as a separate tongue, and e) because of the steadily increasing influence of F r e n c h . A. Celtic influence has been suggested as the cause of the development of P o r t u g u e s e as a separate language (Wechssler,

459).

22. THE

SPREAD OF

PORTUGUESE.

As a result of discovery and colonization, Portuguese has been carried to all parts of the world. I t is spoken in Brazil, the Azores, the M a d e i r a Islands, the Cape Verde Islands, in parts of W e s t Africa, E a s t Africa, India, and the M a l a y Peninsula. T h e n u m b e r of people speaking it is about fifty-seven million, seven million of whom are in continental Portugal. T h a t all of P o r t u g a l ' s discoveries were made to the south and east is explained b y the f a c t t h a t t h e y were inspired

18

INTRODUCTION

[ § 22-23

by the project of finding an eastern sea route to India. T h e plan of Columbus t o reach I n d i a by sailing west was rejected by J o h n II, who later challenged t h e right of Spain t o the lands t h a t Columbus discovered. T h e question of determining the respective rights of Portugal and Spain soon became acute. Pope Alexander VI intervened in the m a t t e r , which was finally settled in 1494 by the T r e a t y of Tordesillas between J o h n II and Ferdinand and Isabella. I t was agreed t h a t Portugal should receive all lands east of a line running from pole to pole three hundred and seventy leagues west of the Cape Verde Islands while Spain should receive all lands west of the line. Brazil, the only exception in the eastward trend of Portuguese discovery, whether discovered by accident or not, was at first looked upon by the Portuguese as merely a way-station on the route to India. B u t M a n u e l I's claim to possession led to fresh disputes with Spain as to the exact location of the new country with respect to the line of demarcation, disputes which continued for almost three h u n d r e d years. 23.

DIALECTS.

S t a n d a r d Portuguese is t h a t spoken in t h e region between Lisbon and Coimbra and imitated by educated people t h r o u g h o u t the country ( V P N , § 39). T h e continental dialects of Portuguese are found either in Portugal or in neighboring sections of Spain. T h e chief dialects in Portugal are, in the northwest, I n t e r a m n e n s e with the subdialects of the provinces of D o u r o and M i n h o ; in the northeast, T r a s m o n t a n o ; in t h e centre, Beiráo; and in the south, Meridional with the subdialects Estremenho, Alentejano and Algarvio (Esquisse, § 7). Some of these dialects extend into Spain: T r a s m o n t a n o at Ermisende in León (RL, V I I , 139-145); BeirSo a t San M a r t í n de T r e v e j o in E x t r e m a d u r a (RL, X X V I , 247-259; Onis, 63-69); and Alentejano at Olivenza in E x t r e m a d u r a (RL, I I , 347-349).

§ 23-25]

PORTUGUESE

ORTHOGRAPHY

19

T h e chief c o n t i n e n t a l d i a l e c t o u t s i d e of P o r t u g a l is G a l i c i a n . S o m e h a v e classified G a l i c i a n a n d P o r t u g u e s e as c o - d i a l e c t s ( R L , I I , 345). M i r a n d e s , w h i c h is s p o k e n in a v e r y small region in t h e e x t r e m e e a s t of T r a s - o s - M o n t e s , is c o n s i d e r e d b y s o m e t o be a b r a n c h of L e o n e s e ( H a n s s e n , § 3, 15; M - L , I n t r o , p . 53, n o t e 1 of C a s t r o ) b u t is r e a l l y a s e p a r a t e l a n g u a g e i n t e r mediate between Galician-Portuguese and the AsturianL e o n e s e d i a l e c t of S p a n i s h ( P h M , I I , 76). T h e c o n t i n e n t a l dialects never became sufficiently diff e r e n t i a t e d f r o m t h e l i t e r a r y l a n g u a g e t o be a b l e t o e x e r t a g r e a t i n f l u e n c e o n it ( R o m , I I , 290), while t h e d i a l e c t s of o t h e r p a r t s of t h e w o r l d w e r e t o o r e m o t e t o d o so. O n t h e c o m p l e x p r o b l e m of t h e r e g i o n a l e l e m e n t s i n v o l v e d in t h e rise of P o r t u g u e s e , see Z R P h , L V I I , 632. PORTUGUESE 24.

PERIODS

OF

ORTHOGRAPHY

PORTUGUESE

ORTHOGRAPHY.

T h e h i s t o r y of P o r t u g u e s e o r t h o g r a p h y is d i v i d e d i n t o t h r e e p e r i o d s : a) t h e p h o n e t i c p e r i o d , w h i c h c o i n c i d e d w i t h t h e p e r i o d of Old P o r t u g u e s e ; b) t h e e t y m o l o g i c a l p e r i o d , which lasted f r o m t h e Renaissance to the t w e n t i e t h cent u r y ; a n d c) t h e r e f o r m e d p e r i o d , w h i c h b e g a n w i t h t h e a d o p t i o n of t h e P o r t u g u e s e G o v e r n m e n t ' s nova ortografia in 1916. 25.

THE

PHONETIC

PERIOD.

I n t h e p h o n e t i c p e r i o d scribes t r i e d t o r e p r e s e n t p h o n e t i c a l l y t h e s o u n d s of t h e w o r d s t h e y w r o t e . As t h e r e were m a n y n e w s o u n d s w h i c h did n o t exist in L a t i n a n d for w h i c h n o t r a d i t i o n h a d d e v e l o p e d , t h e y were o b l i g e d t o a d a p t old g r a p h s a n d i n v e n t n e w ones. A n d m a n y i n c o n sistencies d e v e l o p e d . Scribes w o u l d o f t e n r e p r e s e n t t h e s a m e s o u n d in d i f f e r e n t w a y s a n d d i f f e r e n t s o u n d s in t h e s a m e w a y ( R L , I X , 261). F o r e x a m p l e , a w a r e t h a t g w a s

20

INTRODUCTION

[§ 25-26

h a r d in some words, t h e y a s s u m e d t h a t it could be hard in all words, hence the spelling gisa for guisa; disregarding this conclusion and a w a r e also t h a t g was soft in some words, t h e y assumed t h a t it could be soft in all words, hence the spelling f u g o for f u j o . T h u s in gisa and guisa, t w o graphs, g and gu represent the same sound, while in gisa and fugo, t w o sounds, [g] and [3] are represented b y the same letter. 26.

CONFUSION

OF

GRAPHS.

1. Use of qu for c: cinquo

for cinco;

nunqua

for

nunca.

A. Because of this confusion, quo was sometimes used to represent the sound [kw], e.g., quoall (RL, IV, 207, A.D. 1453). Cf. section 2 A below. B. The use of c for qu is found but is rare, e.g., pecena for pequena and esceeqidas for esqueecidas (Abraham, § 11). 2. Confusion of g and gu: amiga; algem for alguem; gerra

alguo for algo; for guerra.

amigua

for

A. Because of the use of gu for g [g], scribes began early in the fifteenth century to use guo for gu [gw], e.g., linguoa (Castelo, folio 41 v o ) ; d a g u o a (RL, IV, 202, A.D. 1456), i.e., de aguoa. By the end of the fifteenth century guo was replaced by go: agoa for agua: goardar for guardar. The use of 0 for u [ w ] was then adopted in other positions, e.g., continoar, Manoel. Cf. section 1 A above. 3. Confusion of g, gi, and j: agia for haja; mangar for manjar; sega lor seja. 4. Confusion of i, y, and j: ay a for haja; iulgar for julgar; oye for hoje; Ijuro for livro; mujto for muito; ydade for idade. 5. Confusion of u and v: auer for haver; vsar for usar; ovuir for ouvir. 6. Use of * for is: rex for reis; ex for eis. 7. Confusion of m, n, and t i l : ano and ano for anno, camiho for caminho; cimco for cinco; grade for grande; hoe ( R L , VIII, 2 5 6 ) for home or homem; huanal ( R L , I, 3 3 6 ) for humanal; huildade ( R L , XVI, 103) for humildade; menesmo (RL, VIII, 36; RL, XI, 8 8 ) for meesmo ( o l d ) ; poho

§ 26-27]

PORTUGUESE

ORTHOGRAPHY

21

for ponho; poner (Cd'A, I, 157, n. 1) for poer ( o l d ) ; saom for sao ( f r o m sanu-); senpre for sempre; tepo for tempo; emader ( F M , II, Glossary) for eader. A. This confusion obviously did not occur in the initial position. B. T h e til was used in some cases for intervocalic m to save space in an effort to keep the line f r o m running into the margin (CA, I, xvii, n. 2 ) . c. It seems t h a t there was a predilection for n before p and b (cf. BF, I, 45; Rad, 51). This consistent use by m a n y scribes may indicate that the Old Portuguese sound was [n] or a combined [ n ] and [ m ] ; or perhaps it may simply indicate nasalization of the preceding vowel (cf. Manual, § 35, 1 c ) . As late as 1672 Bcnto Pereira found it necessary in his "Ars Grammaticae" (Pereira, 288) to urge the use of m before p. D. T w o acute accents on two similar vowels were sometimes used for the til: huu for huu (old); homees for homees (old). E. T h e use of final m to indicate the nasalization of the final vowel arose perhaps in monosyllables in legal prose in imitation of Latin spelling, e.g., com, quem, rem (old), tarn (RL, X X V I I I , 28). T h e change f r o m n to m took place in the course of the t h i r t e e n t h century ( A H P , IV, 198).

27.

USE

OF

h.

1. T h e letter h was used t o s h o w hiatus b e t w e e n unlike v o w e l s or b e t w e e n v o w e l s of unlike q u a l i t y : poher for poer (old); tehudo for teudo (old); maho ( A H P , III, 21, A . D . 1331) for mao, the spelling maho representing the word with t w o syllables before the d i p h t h o n g ao d e v e l o p e d ; veher for ve^r (modern vier), whereas veer (modern ver) was never written w i t h h because b o t h i's were of the s a m e q u a l i t y . Cf. S P V , § 75. A. This h survived in some words until it was eliminated by t h e nova ortografia; one can, for example, distinguish the new or remodeled theatres in Lisbon by the sign saida, which in the older theatres is spelled sahida. 2. T h e letter h w a s used before initial v o w e l s , perhaps

22

INTRODUCTION

[§27

at first t o show hiatus with the final vowel of a preceding w o r d ; later, this purpose being f o r g o t t e n , it came t o be considered as p a r t of t h e regular spelling of t h e w o r d : ha for a (article); hi for i or y (old); hidade for idade; hir for ir; hordenar for ordenar; honde for onde; hu for u (old); hum for urn; husar for usar. A. In he (for e) the A was probably used to distinguish the verb from the conjunction e. The spelling e, instead of he, is referred to as new in 1574 by Magalhaes de Gandavo in his "Regras que ensinam a maneira de screver a orthographia da lingua portuguesa." The spelling he for the conjunction is rare, e.g., Rom, XI, 374, line 30; RL, VII, 61. B. The h in alghuu was used in imitation of huu, which scribes realized was a component of algum. 3. The letter h was used by false regression: themor (FM, II, Glossary) for temor; theudo for teudo (old). 4. Initial h was often omitted in words in which it existed in Classical Latin: ouve for houve; ome for homem. 5. T h e letter h was used after a consonant for consonantal i: sabha for sabia (old); servho for servio (old); termho for termio (old). This use of h has survived only in the graphs Ih and nh. 6. In early Portuguese ni, n, and nn were used to represent the sound [ji], and l i , I, and 11 were used to represent the sound [X]. T h e graphs nn and 11 came from Spain via Galicia and are especially common in the "Cancioneiro da Ajuda." Pedro A. d'Azevedo states (RL, I X , 263) that the first known dated occurrence of Ik was in a document of the year 1269 and of nh in a document of the year 1273. However, these dates can be moved back slightly, for in the "Livro de D. Joao Portel," which Azevedo published subsequently, Ih (in lha) appears in a document of 1265 ( A H P , IV, 306) and nh (in gaanhar) in a document of

§ 27]

PORTUGUESE

23

ORTHOGRAPHY

1267 ( A H P , V I , 68). 1 T h e theory generally accepted for the origin of these graphs is that they were borrowed f r o m Provençal (Grund, I, 922, n. 3; Huber, § 6 1 ) . T h e borrowing m a y have been brought about in several w a y s : a) through the reading of T r o u b a d o u r poetry ( R L , X I , 84), b) through the work of the secretaries of French prelates, who occupied m a n y of the principal sees in Portugal in the twelfth and thirteenth centuries ( R L , X X V I I I , 25), and c) through the work of reorganization by French scribes of the chancery of Afonso III or Dinis between the years 1270 and 1280 ( C A , I, xv, n. 4). T h e theory of Provençal borrowing is in a measure confirmed by a document dated 1281 in which [X] is represented in a Provençal proper noun by the graph Ih, viz., Vidalhac, but in all other words by the graph //, e.g., fillado ( R L , V I I , 73-74). See also R L , VI, 263. A. T h e s o u n d [P] d i d n o t e x i s t in all w o r d s in O l d in w h i c h it e x i s t s t o d a y .

raina

[reie] a n d

dieiros

Portuguese

T h i s is s h o w n b y e a r l y s p e l l i n g s , e . g . , in t h e s a m e t e x t w i t h

senior

1214).

L a t . -inu

in w h i c h h w a s u s e d s i m p l y t o

and -ina),

T h e s p e l l i n g s -iho

ienio

and

( R L , V I I I , 82, A . D .

and -iha

(from

indicate

h i a t u s , a t first r e p r e s e n t e d t h e p r o n u n c i a t i o n s [ïu] a n d [ie]. the combination

nh (or ~h) in t h e w o r d nenhum

(or

And

nehum)

did

n o t a t first r e p r e s e n t t h e s o u n d [p] b u t c a m e a b o u t t h r o u g h u n i o n in a n e w c o m p o u n d of t h e w o r d s nen

the

a n d hum.

It

t h u s h a p p e n e d t h a t w h e n [ïu] a n d [ÏBJ b e c a m e [ijiu] a n d

[ipu]

and when

[nëù] b e c a m e

(or ne)

[nijiu], t h e c o m b i n a t i o n

nh

w h i c h h a d p r e v i o u s l y b e e n a d o p t e d f o r [jij in o t h e r

(and positions,

w a s a l r e a d y in t h e p r o p e r p l a c e . 1 However, any

supposition

that

this

p r i o r o c c u r r e n c e of

( a n d ~h) in p o s i t i o n s w h e r e t h e s o u n d [ji] l a t e r d e v e l o p e d ,

nh was

1 Both Ik and nh appear in a document dated 1262 of the "Livro de D. J o i o Fortel" (AHP, VII, 474) but the repeated use of «A in the word dinhtiro, in which a palatalized nasal had not developed at .the time, indicate* that thii manuscript is a copy made considerably later than the year 1262.

* O b v i o u s l y t h e c o m b i n a t i o n ~h w a s used in m a n y p o s i t i o n s where the s o u n d [ p ] d i d not l a t e r d e v e l o p , e . g . , •pot).

mêhos

for

mios

(modern

menos), poker for pôer

(modern

INTRODUCTION

24

[§ 2 7 - 2 9

the origin of the g r a p h nh, and later b y imitation, of the graph Ih, c a n n o t be s u b s t a n t i a t e d , in view of the f a c t t h a t nh and Ih (as well as mh, bh, vk, etc.) appeared before [iu] and [IE] became [ipu] and [ipu]. This is p r o v e d b y spellings f o u n d in e a r l y dated documents, e.g., vinha (from vineam) and uelha in the same text with dineyros ( R L , I X , 2 6 8 - 2 6 9 , A . D . 1 2 7 3 ) ; quihentos (from OSp. quinentos) and Iky in the same t e x t with muino and mu\o

(from molinum)

(RL, VIII, 45, A.D. 1293); and termhos

(RL,

I X , 270, A . D . 1 2 7 3 ) . 28.

INTRUSIVE

ORTHOGRAPHIC

p.

1. T h e practice in Vulgar Latin and M e d i e v a l Latin of inserting a p between m and n for the purpose of preserving the sound of both nasal consonants (GVL, § 307) was continued in Old Portuguese merely by orthographic tradition, as the group was probably pronounced [n] with nasalization of the preceding vowel; e.g., dampno; solepnemente. T h a t the p had no phonetic value is shown by the fact t h a t it was sometimes entirely misplaced, e.g., compdenar ( F M , II, Glossary) for condenar. Cf. F M , I, xxviii. 2. A p was likewise inserted in forms of the verb escrever, e.g., escrepver, escprito, perhaps through reminiscence of the Latin forms scripsi and scriptus. This p sometimes took the place of c, e.g., esprever, esprito. Cf. F M , I, xxix. 29.

DOUBLE

VOWELS.

1. Double vowels first developed through the fall of an intervocalic consonant, e.g., cree (from credit); poboo (from populum). 2. T h e y continued to be used b y tradition long after t h e y had contracted in pronunciation. 3. Later t h e y were used in place of single vowels to indicate nasalization or perhaps the lengthening of the nasalized vowel, e.g., coontar (from computare); seentir (from sentire). In m a n y of these cases, the nasalizing consonant

§ 29-30]

PORTUGUESE

ORTHOGRAPHY

25

(intervocalic n) had fallen, e.g., liodes (from leones)] maao (from manum). Cf. Ent, 289. 4. They were also used in place of single tonic vowels after the fall of other intervocalic consonants, e.g., ceeos (from caelos); mandaae (from mandate); quaaes (from quales). 5. It is thought that they were also used to indicate stress. This use probably originated through the fact that in their original occurrence (i.e., after the fall of an intervocalic consonant) one of the two vowels was usually stressed (FM, I, xxi). 6. They were used by a sort of orthographic contamination, e.g., deestro (from dextrum) in imitation of seestro (from sinistrum). 7. Early in the sixteenth century they came to be used to indicate the open sound of the vowel (a, et and o), the closed sound being indicated by a single vowel. This use is mentioned in 1536 by Fernao de Oliveira in his "Grammatica da Lingoagem Portuguesa" (Oliv, 28). 30.

DOUBLE

CONSONANTS.

1. All double consonants which existed in Classical Latin may be found in Old Portuguese, where they do not, however, represent long sounds. Intervocalic rr and intervocalic ss are the only double intervocalic letters which represent sounds different from the single intervocalic letter. The use of intervocalic ff for /, e.g., defender for defender, may have been adopted in order to indicate unmistakably the sound of /, inasmuch as Latin intervocalic short / had become v in Portuguese and was probably pronounced v in the Latin of the time (RF, XXV, 649). Where the scribe used intervocalic ss for s, e.g., ussar for usar, intervocalic s for ss> e.g., dise for disse, and intervocalic r for rr, e.g., corer for correr, he simply failed to recognize the difference in sound. Such spellings may have resulted from imitation of the indiscriminate use of other single and double intervocalic consonants where there was

26

INTRODUCTION

[§ 30-31

no difference in pronunciation, e.g., pallavra for palavra; cavalo for cavallo. However, aside from the intervocalic position, most new cases of doubling seem to have had a phonetic purpose. A. In Old Galician, 11 and nn were used instead of Ih and nk. 2. These are the cases in which double consonants occurred in other t h a n the intervocalic position: a) Initial ff and R a f t e r a consonant: ffe for fe; conffirmar. This use m a y have developed in imitation of the use of ff in t h e intervocalic position. b) Initial 11: Uaa for la. This use m a y indicate a longer sound than modern initial /; it is possible t h a t this long sound saved initial I f r o m falling when the word in which it stood was joined in close syntactical union with a preceding word ending in a vowel. c) Final 11 and 11 before a consonant: mortall for mortal; malldade for maldade. This use indicates t h e velar sound [1], which I still has in these positions t o d a y . d) Initial rr, rr before and after I and n, and rr after s: rreter for reter; Carrlos for Carlos; honrra for honra. This use indicates the more vibrant sound which r still has in these positions today. e) Initial ss and ss after a consonant: ssempre for sempre; consselho for conselho; converssar for conversar. T h i s use was probably adopted to indicate voiceless s because of an awareness t h a t in the intervocalic position ss was voiceless while s was voiced.

31. T H E

ETYMOLOGICAL

PERIOD.

In the etymological period (sometimes called the pseudoetymological period), Latin and Greek spellings were introduced with u t t e r disregard for pronunciation. T h i s practice had begun long before the sixteenth c e n t u r y in a small group of words, especially in works translated from Latin

§31]

PORTUGUESE ORTHOGRAPHY

27

(see R L , X I X , 64), e.g., escripto, feicto, node, reigno, sancto,1 b u t it was t h e w r i t e r s a n d p a r t i c u l a r l y t h e p r i n t e r s of t h e R e n a i s s a n c e w h o g a v e it a v o g u e t h a t was t o last d o w n t o our times. W e a c c o r d i n g l y find ch, ph, rh, th, a n d y in words of G r e e k or s u p p o s e d G r e e k origin, e.g., chrystallino, eschola, phrase, rhetorico, theatro, estylo, nympha; a n d ct, gm, gn, mrt, mpt, a n d double c o n s o n a n t s in words of L a t i n origin, e.g., aucthor, fructo, augmento, digno, magno, damno, somno, prompto, bocca, peccar, cabello, setta. A t t h e same t i m e , o r t h o g r a p h i c false regressions a b o u n d e d , e.g., th in thesoura a n d in ethymologia; y in pkylosophia; d o u b l e c in occeano. A m o n g these belongs t h e c h a n g e of final s to z in mez, portuguez, poz, etc., which arose t h r o u g h i m i t a t i o n of words like simplez, vez,fez, etc. O p i n i o n in f a v o r of t h e new spelling was f a r f r o m u n a n i m o u s . D u a r t e N u n e s de Leao c o n d e m n e d it in his " O r t h o g r a f i a da lingoa p o r t u g u e s a " (1576) as did A l v a r o F e r r e i r a de V e r a in his " O r t o g r a p h i a ou a r t e p a r a escrever c e r t o na lingua p o r t u g u e s a " (1633). C o n s t r u c t i v e criticism was n o t l a c k i n g b u t it was generally disregarded. J o a o F r a n c o B a r r e t o in his " O r t o g r a f i a da lingua port u g u e z a " (1671) first p r o p o s e d t h e m o d e r n use of accent m a r k s t o i n d i c a t e o p e n a n d close vowels a n d t h e m o d e r n use of t h e final g r a p h s -ao a n d -am t o distinguish a c c e n t e d a n d u n a c c e n t e d t h i r d p l u r a l s a n d B e n t o Pereira in his " A r s G r a m m a t i c a e " (1672) r e c o m m e n d e d t h e m o d e r n use of i a n d / , a n d u a n d v, a n d c o n d e m n e d t h e use of double vowels a n d initial d o u b l e c o n s o n a n t s (Pereira, 287-318). T h e e i g h t e e n t h c e n t u r y was a period of b i t t e r o r t h o g r a p h i c polemic; it e n d e d with etymological spelling r a m p a n t . O n e b o o k in p a r t i c u l a r , t h e " O r t h o g r a p h i a ou a r t e de escrever e p r o n u n c i a r com a c e r t o a lingua p o r t u g u e z a " of J o a o de M o r a e s M a d u r e y r a F e y j o , which was first 1

T h e l e t t e r i n s e r t e d in i m i t a t i o n of t h e L a t i n spelling was s o m e t i m e s m i s p l a c e d ,

e.g., scatifica

( A b r a h a m , § 19, 1) for sactifica;

maglino

for

maligno.

28

INTRODUCTION

[§31-32

printed in 1734, had tremendous influence in the cause of etymological spelling for more than a century and a half. The

author not only urged the L a t i n spelling of

borrowed

words

to this spelling

but

also

(Moraes,

their pronunciation 5-7).

The

newly

according

restoration

of

sound [g] as well as the letter g in words like digno pkleugma

the and

resulted f r o m efforts of this kind on the part of

grammarians. W h i l e the Spanish A c a d e m y was reforming o r t h o g r a p h y in the eighteenth century on a rational phonetic basis, the A c a d e m y of Lisbon was consolidating the tradition of t w o centuries b y setting up e t y m o l o g y as the supreme principle of orthography.

Perhaps it was influenced in this policy

b y French o r t h o g r a p h y and b y a pedantic desire t o devise a new mark of distinction for men of letters and the small reading public ( R H i , I, 4). 32. T H E

REFORMED

T h e nova ortografia

PERIOD.

represents a return t o phonetic spell-

ing but it differs f r o m the phonetic spelling of the M i d d l e Ages in that it has been made uniform b y convention.

It

was formulated b y a commission appointed b y the P o r t u guese g o v e r n m e n t in 1911 and was officially adopted Portugal

in

1916.

Its

original

form

has

been

altered b y subsequent amendments and b y the

by

slightly "Acordo

Ortografico Luso-Brasileiro," on the basis of which it was finally adopted b y Brazil in 1931. R o m a o ; V O ; R - M , 322-323-

Cf. R L , X I V , 200-226;

PHONOLOGY TONIC VOWELS 3 3 . V . L . TONIC a ( C L . L . a AND a ) .

1. V. L. tonic a > Ptg. a: pratum > prado; bondade; caput > cabo; mare > mar, A. Cl. L. aldcrcm 1) > alignr.

> V. L. *dlecrem

bonitatem

>

> * alter em (GVL, § 195,

2. V. L. tonic a + j > ai or ei: andamium (Du Cange) > andaimo; -arium > -airo > -eiro, e.g., contrarium > contrairo (old and popular), primdrium > primeiro; basium > beijo; *basslum > baixo; caveam > gaiva; *rabiam (for rabiem) > raiva; sapiam > sabia > saiba; laicutn > leigo. A. The consonants with which attraction took place in these examples are b, p, v, m, r, s and ss. A consonant affected by yod (except s and ss) prevented attraction: allium > alho; facto > fa(o; araneam > aranha. See § 90. B. Attraction took place relatively late in words in which ai did not become amai > amei; laicum > leigo. But if hiatus with e developed through the fall in Portuguese of an intervocalic consonant, the change to ei did not take place: amatis > amades > amais; tales > taes > tais. For the development of reis (from regales), see § 99, 3 D. 3. V. L. tonic a + { from the palatal consonant of a following group (ct, sc, x) > ei: fascem > feixe; fraxinum > 29

C§33

PHONOLOGY

30 freixo; lacte > leite; > seixo.

placitum

> *plactum

> preito;

saxum

A. T h e vowel was not affected if the group was gn or cl: agnum > anho; novdculam > *novacla > navalha.

4. V. L. tonic a + u and therefore the diphthong au > ou [o]: amavit > amaut > amou; soube; taurum > touro.

causam

> cousa;

sapuit

>

A. T h e intermediate stage was pronounced [ow], the afterglide of which began to disappear probably in the sixteenth c e n t u r y a n d now survives only in the n o r t h of Portugal (Esquisse, § 56 e). B. It was the consonantal quality of the second element of the d i p h t h o n g au or of its successor ou t h a t sometimes prevented the voicing of a following medial consonant, e.g., paucum > pouco; raucum > rouco; cautum > couto. c. Examples of the change to ou are found as early as the first half of the t e n t h century ( B F , II, 186). D. Cl. L. pauperem > V. L. *poperem ( M a n u a l , § 47, 3 a) > pgbre. E. If the attraction took place late, au remained u n c h a n g e d : aquam > auga (old and dialectal); tabulam > tabua > tauba (Esquisse, § 43 c). F. T h e fall of intervocalic d took place early enough for t h e development of a + o to ou: vado > vou.

5. V. L. tonic a + I followed by a consonant > ou [o] or al: alterum > outro; jalcem > fouce; saltum > souto; and altum > alto; jahum > falso. See § 9 4 , 1 A a n d B. 6. V . L. t o n i c a f o l l o w e d b y a n a s a l > [£] o r [E] : agnum > anho; amat > ama; ambos > ambos; araneam > aranha; manum > mao; pannum > pano; ranam > ra; tantum > tanto. A. T h e intermediate stage was [5], which has survived before consonant groups beginning with a nasal (cf. § 95, 1), in t h e diphthong ao (§ 78, 3), and a f t e r contraction (§ 78, 2). For t h e change of final [I] to [ew], see § 157, 2. B. This [E] closed one more step to [O] in the word fome ( f r o m famem), a change which was b r o u g h t a b o u t t h r o u g h t h e influence

TONIC

§ 33-34]

VOWELS

31

of come in such expressions as Come quern tern fome ( N a s c e n t e s ) . A l t h o u g h spelled fame, it is f o u n d in rimes with come a n d home in t h e early Cancioneiros ( C D , note to line 2741). See L P , I, 251—254 for p r o v e r b s a n d o t h e r expressions c o n t a i n i n g rimes of fome, come, a n d home, e.g., Quem tem fome, come um home (Beira A l t a ) . T h e / in F r . soif arose t h r o u g h similar influence of beif in such expressions as beif se as seif ( N y r o p , I, § 503, 3). F o r o t h e r e x p l a n a t i o n s of fome, see N a s c e n t e s . 34.

V . L. T O N I C ¿ (CL. L. e AND

ae).

1. V . L . t o n i c e > P t g . decern > petram > p(dra; quaerit > qu^re; levem >

pedem

> p'e;

l¡ve.

A. C. L. stellam > V. L. * stellam ( G V L , § 163) > estrela. B. T h e e of vespa ( f r o m vespam) has been explained as d u e t o t h e influence of besta ( G r u n d , I, 926). 2 . V . L . t o n i c ( + i > e o r ei: cere slam ( V o k , I , 192) > cereja; ingenium > engenho; materiam > madeira; nervium ( A i n s w o r t h ) > nervo; sedeam > seja; superbiam > soberba; tertium > ter(o; venio > venho. A. T h e d e v e l o p m e n t to ei is found only where the ¿ and t h e y o d were s e p a r a t e d b y a short r. It occurred with o t h e r cons o n a n t s in Old P o r t u g u e s e , e.g., cereija for cereja a n d in dialects, e.g., beinho (Opúsculos, I I , 204) for venho. B. Final i h a d t h e s a m e effect as y o d : heri > eire (old). c. T o w a r d t h e beginning of t h e n i n e t e e n t h c e n t u r y , e followed b y ch, j, Ih, nh, or x, b e c a m e [E] in t h e p r o n u n c i a t i o n of Lisbon ( V P N , 92; D u n n , 10-11).

3. V. L. tonic f in proparoxytones with e or i in the penult developed as follows: a) it became i if the penult w a s i ( f r o m C l . L . i ) : decimum

> dizimo;

perticam

>

pirtiga

(Grund, I, 927); b) it became e if the penult was e (from C l . L . i ) : levitum

>

néspera;

>

pessego;

c) i t r e m a i n e d £ if t h e p e n u l t w a s e ( f r o m C l . L . e):

nepetam

perditam

> *perdeda

> n 'eveda; vesperam

( R E W ) > lévedo; > perda >

i'espera.

mespiiam

(§ 1 1 4 ) ; persicum

PHONOLOGY

32

A. T h e f o r m tibio (which is c o m m o n to P o r t u g u e s e and Spanish) has been explained as coming f r o m a V. L. *tipedo (for tepidum), in which t h e close t arose t h r o u g h t h e influence of the semantically related frigidum. T h i s explanation is all the more convincing because this close i is f o u n d only in those regions ( P o r t u g a l and Spain) where the close i of frigidum survived (Language, X I I I , 145-146). 4. V . L. t o n i c ( + i f r o m t h e p a l a t a l c o n s o n a n t of a f o l l o w i n g g r o u p (ct, x, gr) > ex: despectum > despeito; integrum (§ 3 a) > inteiro; ledum > leito; sex > seis. A. T h i s change did not t a k e place if the group was cl: veclum ( A p P r ) > v(lho. T h e e of velho is sometimes pronounced [c] in Lisbon t h r o u g h t h e influence of words in -elho in which the e was originally close. T h e f o r m speclum ( A p P r ) (for speculum) proba b l y h a d e t h r o u g h t h e influence of the c o m m o n ending -iculum (S-G, § 29) and t h u s became espelho. 5. V . L. t o n i c g + u > eu: equam > euga ( p o p u l a r ) . 6. V . L. t o n i c i in h i a t u s w i t h a f o l l o w i n g a > i: judaeam > judia; meam > *mea ( M a n u a l , § 66, 1) > mia > minha. A. T h i s change did not t a k e place in later borrowings, e.g., Team > ree >

re.

7. V. L. t o n i c f in h i a t u s w i t h a f o l l o w i n g o > e: deus > deus; ego > eo ( G V L , § 385) > eu; judaeum > judeu; meum > meu. A. These words rime in the early Cancioneiros with each other b u t not with t h e third singular ending -eu of weak preterits; it is, therefore, likely t h a t the e had not closed at t h a t time. B. If the hiatus arose relatively late, t h a t is, in a late borrowing or t h r o u g h t h e fall in P o r t u g u e s e of an intervocalic consonant, this change did not t a k e place: reum > reu; caelum > ceu. 8. V . L. t o n i c f f o l l o w e d b y a s y l l a b l e e n d i n g in o > e: catellum > cadelo; metum > medo. 9. V. L. t o n i c ( f o l l o w e d b y a c o n s o n a n t g r o u p b e g i n n i n g w i t h w o r n > [e]: dentem > dente [denta]; tempus > tempos [tempu$].

TONIC

§ 34-35]

VOWELS

33

A. I n some dialects t h e vowel did not close (Esquisse, § 40 b ; R L , I I , 111; R L , X I , 270). 10. V . L. t o n i c ( + a final n a s a l c o n s o n a n t , | a n n t h a t b e c a m e final in P o r t u g u e s e , or + nes > [e], t h e n [ej] a n d in t h e r e g i o n b e t w e e n L i s b o n a n d C o i m b r a [ej]: quem > quem [ k i j ] ; bene > bem [ b i j ] ; venit > vem [v§j]; tenes > tens [t£j$]. 35.

V.

L. TONIC e (CL. L.

e,

i, AND

oe).

1. V . L. t o n i c e > P t g . e: aceturn sitim

> sede; viridem

>

> azedo;

cito >

cedo;

verde.

A. Cl. L. I b e c a m e i in a few learned a n d s e m i - l e a r n e d w o r d s : arliculum > artigoo > artigo; firmum > firme; llbrum > livro; missam > missa; pium > pio. I t is possible t h a t t h e t of L a t i n benignum, malignum a n d dignum was long (cf. C R , X V , 311—314), a l t h o u g h t h e d e v e l o p m e n t of gn in O P t g . benino, malino a n d dino shows these w o r d s t o be s e m i - l e a r n e d . Circum > cerco a n d also circo. Likewise cippum > cepo a n d expo unless expo c a m e f r o m *cxppum, which m a y h a v e arisen t h r o u g h t h e influence of cxpum on cxppum ( G V L , § 163). T h e i of siso ( f r o m seso < sensum) a n d t h e i a n d t h e accent of juxz ( f r o m judlcem) w e r e p r o b a b l y d u e t o t h e influence of juizo (cf. R L , X X I I I , 8 6 ) ; t h a t t h i s a c c e n t was an e a r l y d e v e l o p m e n t is s h o w n b y t h e r i m e juyz : fiz ( C V , N o . 1023). Christum > crista a n d escam ( L E W ) > Isea t h r o u g h t h e closing effect of / plus a s t o p in V u l g a r L a t i n ( S M , I, 614). B. T h e r e was a t e n d e n c y t o p r o n o u n c e e o p e n in l e a r n e d words, e.g., completum > compl(to; fetidum > fetido; sedem > sede. Cf. G I t , § 21 a n d M - L , G r a m , I, § 15. T h i s t e n d e n c y was p r o b a b l y intensified b y a following r: verum > v(ro. Cl. L. nivem > V. L. *n£ve (cf. Sp. nieve), w h e n c e P t g . n(ve. T h e e of s'e ( f r o m sedem) m a y h a v e d e v e l o p e d in i m i t a t i o n of t h e Old P o r t u g u e s e t h i r d s i n g u l a r p r e s e n t i n d i c a t i v e s{ (§ 198, 3). T h e r e is n o a p p a r e n t reason f o r t h e o p e n i n g of e in fe ( f r o m fidem) a n d veu ( f r o m velum). 2. V . L. t o n i c e + { > i or ei: ferlam > misteiro

( o l d ) ; sepiam

> siba;

> feira;

vindemlam

>

mysterlum vindima.

34

PHONOLOGY

[§35

A. T h e consonants with which the change took place in these examples are m, p, and r, t h e change to ei appearing only with r. Cirio (from cereum) is probably a borrowing from Spanish. B. In late learned and semi-learned words the yod had no effect whatever and the vowel opened to ( as in other learned words where there was no yod (§35, 1 B): cereum > céreo; devium > d'evio; ebrium > ebrio; mystérium > misterio; nitidum > nidio. c. A consonant affected by yod prevented the e from closing: cilia > celha; cerevisiam > cerveja; consilium > conselho; justitiam > justeza; stamineam > estamenha; video > vejo; vitium > vezo. For the sound of e in some of these words in the pronunciation of Lisbon, see § 34, 2 c. T h e word tainha is not an exception, as it comes, not from tagenia (Facciolati) b u t f r o m *tagenia with the accent of t h e Greek word rayoviav, t h u s : *tagenia > taeia > taiia (§99, 3) > tainha. Nor is the word juizo an exception as it comes, not f r o m judicium but from *judicium, which arose through the influence of the suffix -icium. T h e change to ei is found in some of these words in Old Portuguese, e.g., cerveija for cerveja and in dialects, e.g., beijo (Opúsculos, II, 204) for vejo. D. In early learned and semi-learned words CI. L. t followed by a consonant affected by yod became i: erviliam > ervilha; justitiam > justifa; milium > miiho; mirabilia > maravilha; tineam > tinha; vitium > vi(o. There is other evidence t h a t some of these words are semi-learned, viz., the development of t + j in justi(a, vi(o, as compared with justeza, vezo (cf. § 89, 4 A). In purely learned words the yod did not affect either t h e vowel or the consonant: cilium > cilio; exilium > exilio; familiam > familia; mmum> minio; although it partially affected a /: iit'iunj > vicio. E. If the yod developed through hiatus after the fall in P o r t u guese of an intervocalic consonant, the result was ei: habetis > havedes > haveis. 3 . V. L. tonic e + i from the palatal c o n s o n a n t of a following group (ct, sc, gr) > ei• benedictum > béeiío ( o l d ) ; piscem > peixe; strictum > estreito; jilictum > jeito ( o l d ) ; riigram > neira (RL, XXI, 2 7 1 ) .

TONIC VOWELS

§35]

35

A. T h e vowel was not affected if t h e g r o u p was cl, gl, o r apiculam lignum

> abelha; > lenho;

articulum

tegulam

> artelho;

*impignus

>

gn:

empenhos;

F o r t h e sound of e in these

> telha.

w o r d s in the p r o n u n c i a t i o n of L i s b o n , see § 34, 2 c . B. F o r m s such as sortilha

a n d lentilha

(from and

( G V L , § 4 2 , 2).

-iculam 4.

( f r o m sorticulam)

resulted f r o m confusion of t h e suffixes -iculam

lenticulam)

V. L. tonic e followed b y

( 2 d sg. p e r f . i n d . )

> -iste

final

i >

i: feci

>

viinte

( o l d ) ; viginti

>

fi.z;

>

-isti

vinte.

5. V. L. tonic e in proparoxytones with e or i in the penult developed as follows: a) it b e c a m e i if the penult w a s i ( f r o m C l . L . i ) : debitam >

liidimo

>

lidimo;

>

sericum

divida; >

legitimum

*sirigo

>

>

sirgo

leidimo

(§ 5 2 ,

1);

b ) it remained e if the penult was e (from Cl. L. /): bibitum

>

Cf. M - L , G r a m , I, § 80.

bebedo.

6. V. L . tonic e in hiatus with a following a > i: ( f o r diem)

>

dia;

viam

>

*diam

via.

7. V . L . tonic e in hiatus with a following a or o through t h e fall in Portuguese of an intervocalic consonant > ei: alienum feio;

>

alheo

sinum

>

>

alheio;

seio;

telam

credo >

teia;

>

creo

venam

>

creio;

>

veia.

foedum

>

A. T h e c h a n g e t o ei did not t a k e place until t h e beginning of the sixteenth century.

8. V . L . tonic e followed by a syllable ending in a > apothecam

>

>

Tegulam

mo(da;

bodega;

ilia >

>

rggra;

(la; telam

metam >

t(la;

>

m(da;

vela

>

monetam vgla.

A. S o m e of these words a r e learned o r s e m i - l e a r n e d .

9. V . L . tonic e followed b y a syllable ending in o > i: ipsum

>

isso.

10. V. L . tonic e followed by [qk] or [qg] > i: >

domingo;

pinquum

linguam >

provinco

>

lingua;

>

fi{l.

>

seringa;

pro-

(Eluc).

11. V. L . tonic e followed b y / > fidelem

syringam

dominicum

Ptg.

crudelem

>

cru{l;

36

PHONOLOGY

C§ 36-37

A. T h e following / m a y have caused the e to open in véu tion 1 B a b o v e ) and in tela and vela (section 8 a b o v e ) . 36.

V.

L . TONIC i ( C L .

L.

(sec-

t).

1. V. L. t o n i c i > P t g . i: audire > ouvir;frigidum > /rio; litem > lide. A. The development of pêga (from picam) and esteva (from stivam) is not clear. 2. V. L. tonic i + J > !•' dixi frictum > /rito. A. T h e form coelho

of the suffixes -iculum 37.

> disse;

(from cuniculum)

and -ïculum

filium

>

filho;

resulted from confusion

(GVL, § 42, 2).

V . L . TONIC q ( C L . L . O ).

1. V. L. tonic ç > P t g . ç: fôrtem > f g r t e ; hospitem > hóspede; pôrtam > pçrta; rôtam > rçda. 2. V. L. tonic ç + {' > o or oi: côrium > coito; jolia > folha; fôrtiam (Du C a n g e ) > força; hodie > hoje; *môrîo (for môrior) > moiro (old); Saxôniam > Sansonha (old); sòmnium > sonho. A. T h e c h a n g e to oi is found only where the p and the y o d were s e p a r a t e d by a short r. It is also found under the s a m e conditions in learned words in modern popular speech: historiam

> história

> histoira

(popular); memoriam

> memòria

>

memoira

( p o p u l a r ) . T h e change to oi is found likewise where the ç a n d the y o d were s e p a r a t e d by a single m or « in words in which t h e action of the yod was c o m p a r a t i v e l y late, t h a t is, words w i t h

Romance yod or semi-learned words: comedo > corneo > coimo (old); daemônium > demònio > demoino ( F M , II, Glossary; and popular).

3. V. L. tonic ç + i from the p a l a t a l consonant of a following group > oi: nòctem > noite; octo > oito. A. T h e T of oi w a s absorbed in M o d e r n Portuguese b y a follow-

ing [ $ ] : côxum (LEW, s.v. coxim) ( R L , X X V I I , 21) > coxa.

> coixo > coxo; côxam > coixa

B. If the group was cl, the vowel s i m p l y closed to o:

> oclum

> ôlho; rotulam

> *ròclam

>

rôlha.

oculum

§ 37-38]

>

T O N I C VOWELS

4. V. L . t o n i c p + u > o: pbtuit pos.

> poude

37 > pode;

posuit

A. T h e intermediate stage was ou, which had become O before the time of the change of au to ou. 5. V. L . t o n i c p f o l l o w e d b y a s y l l a b l e e n d i n g in o > focum > fggo; populum. > povo; positum > pgsto.

o:

A. T h e same change took place if the g was in hiatus with final o: teriiolum (Du Cange) > terfoo > ter^o. 6. V. L . t o n i c p f o l l o w e d b y a n a s a l c o n s o n a n t > [o] o r [o]: dominam > dona; bonam > boa > boa; pontem > ponte [ p o n t a ] ; longe > longe; sonum > soo > som; bonum > boo > bgm. A. In some dialects the vowel did not close (Esquisse, § 4 0 b ; R L , II, 111; RL, X I , 270). B. Even a f t e r the nasal resonance disappeared, the vowel remained close, e.g., boa. 38.

V.

L . T O N I C o ( C L . L . O AND «).

1. V . L. t o n i c g > P t g . g: ambrem > amgr; autumnum > outono; buccam > bgca; cursum > corso; fundum > fgndo ( o l d ) ; lumbum > Igmbo; lutum > lodo; nomen > nome; punctum > ponto; truncum > tronco; turdum > tordo; turpem > torpe; turrim > torre; undam > gnda; unde > gnde; vota > bgda. A. Cl. L. o became Ptg. P in a group of words most of which are learned or semi-learned: atrocem > alrgz; ferocem > fergz; remotum > remgto; sacerdotem > sacerdgte; solem > sgl; sonorum > songro; Urrae motum > terremgto; velocem > velgz; vocem > vgz; votum > vgto. B. Cl. L. u became Ptg. u in a group of words most of which are learned or semi-learned: crucem > cruz; culparn > culpa; curtum > curto; curvum > curvo; fundum > fundo; jurtum > jurto; mundum > mundo; secundum > segundo; sulcum > stdco; surdum > surdo; ursum > usso (old) and UTIO. I t is possible t h a t this u was in some words the result of the influence of an

38

PHONOLOGY

[§38

initial p a l a t a l : j ü g u m > jugo; jüstum > justo; plumbum > chumbo. c. T h e p of jgvem (from jüvenem) has been explained as due to the dissimilating effect of the v (Rad, 14). CI. L. nòbilem > semi-learned ngbre, in which the p was due to the influence of pobre. CI. L. nùcem > V. L. *nocem (cf. Sp. nuez) > tifi. CI. L. nirarn (for nürum), through the influence of socram or jorór, > V. L. *noram (cf. Sp. nuera, It. nuora) > Ptg. npra (see REW). For CI. L. octobrem there was apparently a V. L. octübrem (Carnoy, 64), whence outubro. The ff of tgsse (from tùssim) may have developed in imitation of the third singular present indicative tgsse (5 176, 3). There is no apparent reason for the opening of o in nó (from nòdum). 2. V . L. tonic o + i > o, u, oi, or ut: clcònlam > ceggnha; püteum > po(o; *risoneum > risonho; verecündxam > vergpnha; cüneum >cunho; testimonium > testemunho; querimòniam > caramunha; gürgülio > gorgulho; nasturtium > mastruqo; augurium > agoiro; calümniam > coima; cüphiam ( D u Cange) > coifa; güviam ( D u C a n g e ) > goiva; rüsseum > roixo (Opúsculos, II, 34) > roxo; sal müriam > salmoira; tonsoria > tesoira; plúviam > chuiva > chuva; rübeum > ruivo; introitum ( L E W , s.v. intra) > entruido > entrudo. A. The consonants with which attraction took place in these examples are b, / , vf m, r, and ss. It also took place in learned words in modern popular speech where the o and the yod were separated by a single r or n: glòriam > gioirà (popular); Antonium > Anioino (Estremenho). The attracted vowel disappeared in roxo and chuva. B. It is not clear why some forms have o or oi while others have u or ui. The theory has been advanced that u came directly from Q . L. « before the change to p (RHi, L X X V I I , 8 - 9 ; M P , X I , 349). c. In learned words the yod had no effect whatever and the vowel opened to p as in other learned words where there was no yod (§ 38, 1 A): cerimóniam > cerimònia; glòriam > glòria; testimonium > testemónio. 3. V. L. tonic o in p r o p a r o x y t o n e s w i t h an t in t h e

5 38-39]

TONIC VOWELS

39

penult > u: *dûbitam > duvida. 4. V. L. o + i from the palatal consonant of a following group > u: lûctam > luita > lut a; tructam > truita > truta. A. Aside f r o m certain dialects, where the intermediate stage ui has s u r v i v e d (Esquisse, § 56 i), the general tendency is for ui, w h a t e v e r its origin, to c o n t r a c t to u, e.g., communes > comuues

(BF, II, 215 and 220) > comuns;

plûvïam

> chuiva

> chuva.

B. If the group w a s cl or gn (or n g l ) , the vowel simply closed to u: acûculam ( D u Cange) > acuclam > agulha; pùgnum >

punho;

ungulam

>

unha.

5. V. L. tonic o -f- It > ut: ausculto > ascuitd multum > muito; vulturem. > abuitre > abutre.

>

A. T h e i n t e r m e d i a t e stage ui has been preserved in t h r o u g h the influence of the apocopated form mui.

escuto; muito

6. V. L. tonic o followed b y final i > u: *fûgï (for juge) (2d sg. i m p v . ) > f u g e (old); potûï > *podi (§ 37, 4) > pude; fui > fui. 7. V. L. tonic o followed b y a syllable ending in a > p.formant ( L E W ) > fçrma; hôram > hgra; -osam > -gsa, e.g., formôsam > formçsa. 8. V. L. tonic o followed b y a syllable ending in o > u: totum (neuter) > todo > tudo. 9. V. L. tonic o in hiatus with a > u: duas > duas; suam > sua. 10. V. L. tonic o and final o in hiatus > ou: duos > dous; suum > sou (old). 11. V. L. tonic o + / > o: dûlcem > doce; insulsum > ensosso. 12. V. L. tonic o followed b y [rjk] or [rjg] became u: jûncum > junco; numquam > nunca. 39.

V . L . TONIC u ( C L . L .

U).

1. V. L. tonic u > Ptg. u: acutum > agudo; mutum mudo; scutum > escudo. 2. V. L. tonic u + i remained u n c h a n g e d : adduxit

> >

PHONOLOGY

40 ( o l d ) ; fructum

adusse luto;

pecullum

>

> fruito

> fruto;

[§40 lucturn

> luito

>

pegulho.

A. For the contraction of ui to v, see § 38, 4 A. PRETONIC

VOWELS

4 0 . V . L . PRETONIC a (CL. L . a AND a). 1. V . L . p r e t o n i c a > P t g . a [b]: catenam > cadeia; sapdnem > sabao.

apertum

>

aberto;

2. V . L . p r e t o n i c a -f- j > ai or ei: areolam > eiro ( R L , X V , 3 4 2 ) ; bdsiare > beijar; *bassiare > baixar; disvariare ( D u C a n g e ) > desvairar; *gdviottam > gaivota; sdpiamos > saibamos. A. T h e consonants with which attraction took place in these examples are p, v, r, s and ss. T h e same change took place with a Romance yod (the consonant being m): *exfaminare > esfamear > esfaimar. A consonant affected by yod (except s and ss) prevented attraction: alienum > alheio; *maneanam > manha; rationem > razom > razao. T h e e of such forms as menha and rezao, which arose in the late fifteenth or early sixteenth century, developed by dissimilation. B. This change took place also if the a and the yod were in contact in L a t i n : majorinum > meirinko; and if the yod developed through the fall in Portuguese of an intervocalic consonant: *adradicare > arreigar; traditorem > treidor (old). 3. V . L . p r e t o n i c a -(- i from t h e p a l a t a l c o n s o n a n t of a following group (ct, x, gr) > ei: factionem > feiqdo; fragrdre > fiagrare ( G V L , § 2 9 2 ) > cheirar; jactare > jeitar (Eluc); lactucam > leituga; laxare > leixar (old). A. In learned words, a followed by ct (in which c did not become a yod) remained unchanged, viz., [a]: actionem > acfao; activum > activo. If the group was pt, the a remained [a] whether the p fell in pronunciation or not: aptitudinem > aptiddo [aptidiw]; baptizare > baptizar [batizar]. In the earlier learned borrowings praticar (from practicare) and tratar (from tractdre), the a has become [B]. 4. V .

L.

p r e t o n i c a + u and t h e r e f o r e t h e

diphthong

§40] au > ou [o]: audire (for sapueram) >

PRETONIC VOWELS > ouvir; soubera.

laudare

> louvar;

41 *sapueram

A. In some words au had become o in Vulgar Latin, e.g., oriclam (GVL, § 229, 7) > orelha. B. V. L. au followed by a syllable containing accented u > a: augurium > *agurium (GVL, § 228) > agoiro; auscultat > asculta (GVL, § 228) > ascuita (old). For an analysis of this change, see RHi, LXXVII, 155. c. If the diphthong au arose in Old Portuguese through the fall of an intervocalic consonant, no further change took place: salutare > saudar; *a + lo > au; although in Old Portuguese these words are occasionally found with au changed to ou: soudar, ou. 5. In learned words the prefix ab followed by s > au: absentem > ausente; abstinentiam > austinencia ( B F , I, 126); absolutum > ausoluto (old). A. This change was condemned in the eighteenth century (Moraes, 158). B. In the popular development of this prefix, the b fell, e.g., abstinentiam > asteeqa (Rom, XI, 363). 6. V. L. pretonic a + / followed by a consonant > ou [o] or al [ a i ] : *aliqu'unum (GVL, § 7 1 ) > algum; altarium > outeiro; -pal-pare > poupar and palpar. See § 94, 1 A and B. 7. V. L. pretonic a preceded or followed by a labial > o: quadraginta > quarenta > corenta (old and popular); quarn magnum > quamanho > comanho (old); Germ, wardon ( R E W ) > guarir > gorir (old); cf. Rom, X, 341-342; M - L , Gram, I, § 363. 8. V. L. pretonic a and tonic a in hiatus contracted to a [a]: palatium > paa$o > paqo; panatam (Du Cange) > paada > pada. 9. V. L. pretonic a and intertonic a in hiatus contracted to a [a]: *panatarium > paadeiro > padeiro; sanativum (Du Cange) > saadio > sadio. 10. V. L. pretonic a and tonic ( or e in hiatus > g: anellum > aello > ¿lo; sagittam > saeta > S£ta.

42

[§ 40-41

PHONOLOGY

A. If t h e c o n t r a c t e d vowel was followed by a g r o u p beginning with a nasal, it closed, according to § 34, 9: calentem > caente > queente > quente. 11. V . L. p r e t o n i c a a n d t o n i c o in h i a t u s > p: majorem maor > mggr > mor.

>

A. If t h e c o n t r a c t e d vowel was followed b y a nasal, it closed, according to § 37, 6: palumbum > paombo > poombo > pombo. 12. V . L. p r e t o n i c a in h i a t u s w i t h t o n i c i satire > sair; caninum > cainho.

remained:

13. V . L. p r e t o n i c a in h i a t u s w i t h t o n i c u r e m a i n e d : padules ( D u C a n g e ; for paludes) > pauis; salutem > saude. 41.

V . L. PRETONIC e (CL. L .

e, i, oe,

e AND

ae).

1. V . L. p r e t o n i c e > P t g . e [a]: caepullam > cebola; foetdrem > fedor; picare > pegar; securum > seguro; sermonem > sermorn > sermao. A. Some words in which t h i s c h a n g e took place were l a t e r replaced b y regressive w o r d s : figurant > figure (old), replaced b y figura; liccntiam > lecenqa (old), replaced by licenfa; virtutem > vertude (old), replaced b y virtude. 2. W h e n p r e t o n i c e w a s a t t h e s a m e t i m e t h e initial l e t t e r of t h e w o r d , it r e m a i n e d u n c h a n g e d in s p e l l i n g b u t c a m e t o b e p r o n o u n c e d [i] in M o d e r n P o r t u g u e s e : aeternum > eterno; erviliam > ervilha. A. T h e » of igreja (from ecleslam) and of idade ( f r o m aetatem), as is shown b y t h e spelling, developed much earlier. B u t t h e c h a n g e of t h e cl of ecleslam to gr was a semi-learned d e v e l o p m e n t (§ 86, 1 d). For an a t t e m p t t o explain t h e i of these words, see R L , X I I I , 433-434. Cf. section 4 A below. T h e e i g h t e e n t h c e n t u r y spellings egreja and edade were used in i m i t a t i o n of t h e L a t i n spellings and of the spelling of o t h e r words with initial e, now come t o be p r o n o u n c e d [i], 3. V . L. p r e t o n i c e + { > i or ei: laesidnem > lijom ( o l d ) ; prensionem > *presionem > prijom ( o l d ) ; serviamus > sirtamos; vindemiare > vindimar; feriamus > feiramos (old); feriari > feirar.

PRETONIC VOWELS

§41]

43

A. T h e c h a n g e t o ei t o o k p l a c e if t h e i n t e r v e n i n g c o n s o n a n t w a s a s h o r t r. T h e c h a n g e t o i t o o k p l a c e if t h e i n t e r v e n i n g c o n s o n a n t w a s a s h o r t s a n d in all o t h e r c a s e s e x c e p t w h e r e t h e y o d a f f e c t e d t h e c o n s o n a n t p r e c e d i n g it. I n this case t h e e r e m a i n e d u n a f f e c t e d : meliorem > melhor; *metidmus ( f o r metiamur) > megamos; pretidre > prezar; seniorem > senhor; videdmus > vejamos. T h e p a l a t a l i z e d c o n s o n a n t s [3], [ji], a n d [X], p r o d u c e d b y t h e y o d in s o m e of t h e s e w o r d s , l a t e r c l o s e d t h e e t o [1], a c h a n g e t h a t is o n l y o c c a s i o n a l l y s h o w n o r t h o g r a p h i c a l l y : meliorem > milhor ( o l d ) ; tlneolam > tinho. T h e c h a n g e t o [1] took place whether t h e palatalized consonant was produced by y o d o r n o t , e.g., regendum > regendo > [ r j 3 e n d u ] ; it a l s o t o o k p l a c e b e f o r e p a l a t a l i z e d s ( j + c o n s o n a n t ) , e.g., vestlre > vestir [vi$tir], a n d a f t e r i n i t i a l ch, e.g., plicdre > chegar [Sigar]. I n historia ( f o r hesloria < historlam) a n d mistura (for mestura < misturam), the c h a n g e to i was a regression to the Latin f o r m a n d n o t a r e s u l t of t h e l a t e r n o r m a l d e v e l o p m e n t of e b e f o r e palatalized T h e w o r d humilhar

( f r o m humilidre)

is s e m i - l e a r n e d .

4. V. L. p r e t o n i c e + i f r o m t h e p a l a t a l c o n s o n a n t of a f o l l o w i n g g r o u p ( c t , sc, x) >

leitura;

miscere

> meixer

> ei: ficticium

> jextiqo;

lecturam

(old).

A. I n i t i a l ei b e c a m e i: exemptum > eisento > isento; *cxitum ( f o r exltum) > eixido > ixido ( o l d ) . L e a r n e d spellings conceal t h i s c h a n g e in m a n y w o r d s b e g i n n i n g in ex-, e . g . : exactum > exacto [ e j z a t u ] o r [ i z a t u ] . T h i s w a s t h e r e g u l a r d e v e l o p m e n t of O P t g . ei, n o m a t t e r w h a t i t s o r i g i n : eclesiam > eigreja > igreja; inimicum > eimigo > imigo ( o l d ) . F o r t h e p r e f i x enx-, see § 111, 3. B. If t h e g r o u p w a s ct p l u s y o d , t h e d i p h t h o n g w a s f o r m e d b y t h e y o d w h i c h c a m e f r o m [k], w h i l e t h e a c t i o n of t h e f o l l o w i n g y o d w a s l i m i t e d t o i t s e f f e c t o n t h e t: correctibnem > correifao; directionem > direi(om ( o l d ) . T h e d e v e l o p m e n t of lifao ( f r o m lectionem) is n o t c l e a r . I n l e a r n e d w o r d s , t h e [k] d i d n o t b e c o m e a y o d b u t fell in p r o n u n c i a t i o n , a n d t h e e h a s b e c o m e [e]: directionem > direcqao.

5. V. L. p r e t o n i c e followed b y a labial > 0: imaginem imagem

> omagem

( R o m , X , 3 3 6 ) ; sepulturam

>

>

sepultura

PHONOLOGY

44 >

sopultura

[§41

( R L , X V I , 1 1 ) ; *similiare

> semelhar

>

some-

Ihar (old). А. This change is common in certain dialects, e.g., Ptg. beber > bober; Ptg. semana > somana (RL, X X V I I , 103-104). In some d i a l e c t s t h e o c l o s e s t o u, e . g . , P t g . beber > buber;

P t g . semana

>

sumana (Esquisse, § 69 c; RL, V, 145; RL, X I , 274; RL, X I I , 305). б. V. L. p r e t o n i c e followed b y u > i: aequalem *minuare

( f o r minuere)

>

>

igual;

minguar.

7. V. L. p r e t o n i c e followed b y / + a c o n s o n a n t > [e]: delicatum

>

delgado;

*bellitatem

>

beldade.

8. V. L. p r e t o n i c e p r e c e d e d or followed b y r s o m e t i m e s f e l l : periculum

> perigo

( G V L , § 229) > u l a r ) ; berillum

gritar; >

>

prigo

( o l d ) ; quiritare

therlacam

S p . brillo

>

>

theriaga

>

*c,ritare

> triaga

(pop-

brilho.

A. This change took place in Vulgar Latin and in Portuguese. 9 . V . L . p r e t o n i c e f o l l o w e d b y r > a: aeramen serrate >

( D u C a n g e , s . v . serare)

caramunha;

*verrere

>

farrar

( f o r verrere)

>

(old);

>

arame;

querimoniam

varrer.

10. V. L. p r e t o n i c e followed b y a c o n s o n a n t g r o u p beginning with m or n > >

[ e ] : mendicum

>

mendigo;

memorare

lembrar.

A. When pretonic initial e, followed by a consonant group beginning with m or «, was at the same time the initial letter of t h e word, it generally b e c a m e [I]: intrare

> entrar;

implicdre

>

11. V. L. p r e t o n i c e followed b y [qg] > i: vindicate

>

empregar.

vingar;

*nec-unum

>

nengum

>

ningum

(old).

12. V. L. p r e t o n i c e a n d a following t o n i c e in h i a t u s > e: videre

> veer

>

ver; sigillum

> seello

>

selo.

13. V. L. p r e t o n i c e a n d a following i n t e r t o n i c e in h i a t u s > >

creditdrem mfzinha;

>

creedor

praedicare

>

>

crgdor;

preegar

>

medicinam

>

pregar.

14. V. L. p r e t o n i c e a n d tonic i in h i a t u s > i: >

ceinza

crido;

>

venire

ciinza > veir

> cinza; > viir

*creditum >

meezinha

>

creido

>

*clnltia criido

>

vir.

15. V. L. p r e t o n i c e in h i a t u s with t o n i c a, o, u or £ > i

§ 42-43]

45

PRETONIC VOWELS

(sometimes e pronounced [j]): creatum > criado; leom > ledo; minutum > miudo; veneria > veeira

leonem > > vieira.

4 2 . V . L . PRETONIC i (CL. L . T).

V. L. pretonic i > P t g . i: dicendum > primeiro. 43.

V . L . PRETONIC o ( C l . L . o , u,

> dizendo;

primarium

AND o ) .

1. V. L. pretonic p > P t g . o [u]: cupiditiam (Du C a n g e ) > cobi^a; dormire > dormir; formicam > formiga; monstrdre > mostrar; plorare > chorar; potere (GVL, § 403, 1) > poder; superdre > sobrar; superbiam > soberba. A. In some dialects this o became e, e.g., Ptg. procurar > prewar (RL, XI, 271); Ptg. fortuna > fertuna (RL, XXVII, 107). 2. V. L. pretonic o + i > w or oi: dormidmus > mos; *moriatis (for moridmini) > moirades (old).

durma-

A. The change to oi took place if the consonant separating the o and the yod was a short r. The change to u took place in all other cases except where the yod affected the consonant preceding it. In this case the o remained, although it too came to be pronounced [u], in accordance with §43, 1: mulierem > *mulierem. > molher (old); *poneamus (for pondmus) > ponhamos. Mother [muXer] came to be spelled mulker in the course of the sixteenth century in imitation of its Latin etymon and of Spanish mujer. 3. V. L. pretonic p + 1 from the p a l a t a l consonant of a following group > oi: octdvum > oitavo. See § 92, 7 c. A. If the group was gn, the vowel became u or o [u]: > cunhado; cognosco > conhefo.

cognatum

4. V. L. pretonic o (from Cl. L. u) + It > ut: auscultare > ascuitar > escutar; cultellum > cuitello > cutelo. 5. V. L. pretonic p + u > o: potuissem > podesse (old); posuissem > posesse (old). 6. V. L . pretonic p followed b y a consonant group beginning with m o r n > [o]: rumpendum > rompendo; computare > contar.

PHONOLOGY

46

[§ 43-46

A. Words in which [Q] appears may be learned or semi-learned: umbilicum > umbigo; unguentum > ungüento. Popular forms of these words are embigo and enguento. For cumprir (from complire for complete), see § 176, 8 B. 7. V. L. pretonic o followed by / -+- a consonant > o [o]: *voltare ( R E W ) > voltar; volvendum > volvendo. 8. V. L. pretonic o and tonic o in hiatus > o: colorem > COOT > COT. 9. V. L. pretonic o and intertonic o in hiatus > g: coldrátum > cgrado. 10. V. L. pretonic o in hiatus with tonic a, e, or i > o [u] or [w]: solanum > soao; volare > coar; dolere > doer; monetam > moeda; polire > poir; molinum > moinho. 44.

V . L . PRETONIC u ( C L . L .

Ü).

V. L. pretonic u > Ptg. u: durare muralha. FINAL 45.

> durar;

múralia

>

VOWELS

V . L . F I N A L a ( C L . L . á AND

a).

1. V. L. final a > Ptg. a: causam > cousa; mens am > mesa; clamat > chama; hebdómada > domaa > doma (old). 2. V. L. final a and tonic g in hiatus > g: molam > moo > mó; *telariolam > *telariolam > *teleirola > teeiroo > teiró; *avulam > *avolam > avoo > avó. A. The vowel change and shift of accent in *av6lam were due to the influence of V. L. *fili6lam; cf. Sp. abuela and hijuela. B. Nasalization closed tonic P to P and kept it close (§ 37, 6 B); hence, assimilation did not take place when hiatus was produced by the fall of intervocalic n: bonam > boa > boa; sónat > soa. 3. V. L. final a followed by a nasal > [%]: amant

>

amam.

A. For the change from [B] to [iw], see § 157, 2. 46.

V . L . F I N A L e ( C L . L . ?, i, e AND

1. V. L. final e > Ptg. e: Ule > ele;

ae).

veritátem

>

verdade;

§46]

F I N A L VOWELS

famem > jome; carnem dente; sltim > sede.

> carne;

dixit > disse;

47 dentem

>

2. If e w a s t h e final l e t t e r of t h e word in Vulgar L a t i n a n d was preceded b y a s h o r t /, n, r, s or c, or t h e g r o u p t + J preceded b y a vowel, it fell: male > mal; solem > sol; homlnem > ome (old); canem > cam (old); rationem > razom ( o l d ) ; vériít > vem; jinem > fim; commünem > comum; amórem > amor; *habuerit (for habúérit) > houver; quaerit > quer; mensem > mes; vicem > vez; pósüít > * poseí > pos; jacit > faz; fecit > / - f z . A. If the consonant was long, the e did not generally fall: Ule > ele; türrim > torre; vallern > vale. B. In proparoxytones, e preceded by r generally fell in Old Portuguese, but it has been restored in Modern Portuguese: arborem > arvor (TA, Glossary) and árvore; carcerem > carcer (FM, II, Glossary) and cárcere. In learned proparoxytones, e preceded by n did not fall: canonem > cánone. c. In certain dialects a paragogic e is found after / and r: mare > mar > mare; solem > sol > sole. See § 116, 2. D. The assibilation of c and of t plus yod took place before the fall of final e. E. Final e preceded by c plus yod did not fall, e.g., faclem > face, except where c plus yod became confused with t plus yod: aciem > az (old); faclem > faz (old). Cf. § 89, 2 B; § 98, 3. 3. If e was n o t t h e final l e t t e r , it did n o t fall, a l t h o u g h it was p r e c e d e d b y /, n, r, s or c, or t h e g r o u p t + j ; this is s h o w n b y t h e plurals of s o m e of t h e nouns a n d v e r b s listed in section 2 a b o v e : soles > sois; canes > caes; rationes > razdes; homines > omees (old); amores > amores; *habuerint (for habúerint) > houverem; *quaerent (for quaerunt) > querem; menses > meses; vices > vezes; *facent (for faciunt) > fazem; -itles > -ezes. A. In the change from omees to homens final e disappeared by contraction. 4. V. L. final e a n d a p r e c e d i n g e in h i a t u s resulting f r o m

48

PHONOLOGY

[546

the fall of intervocalic d > e: fidem > fee > fé; mercedem > mercee > mercé; sédem > see > sé; crédit > ere. A. This change took place much earlier than the fall of P o r t u guese d (from Latin t) in second plural endings. B. For the e oi fé and sé, see § 35, 1 B . 5. V. L. final e and a preceding e in hiatus resulting from the fall of intervocalic n > é: ténes > ties > tens; *benes > bées > bens; homines > homées > homens. A. For the sound of final -ens, see § 34, 10. 6. V. L. final e in hiatus with a preceding e w h e r e the hiatus resulted from the fall of intervocalic t (2d pi.) > P t g . i: movete > movede > movei; habétis > havedes > haveis. A. Contraction of the two vowels is sometimes found in Old Portuguese, e.g., stétls > estedes > estes (CG, facsimile, fólha 51 vo, col. 3). Cf. Theoria, 33. T h e commoner change to -eis took place probably because the final vowel had already partially closed before the fall of d (first quarter of fifteenth century). However, Hanssen referring to the same forms in Spanish, argues t h a t the contracted ending -es is regular and t h a t -eis developed in imitation of -ais (Hanssen, § 26, 3). 7. V. L. final e in hiatus with a preceding £ where the hiatus resulted from the fall of intervocalic I > P t g . i: *mélés > méis. A. T h e development was the same if the preceding vowel was e, because tonic and intertonic e followed by / had become crudìles > crucis; fideles > fiéis; -ibiles > -iveis (§35, 11 and § 60, 7). 8. V. L. final e and a preceding [j] (CI. L. g) > i: gregern > grei; légem > lei; régem > rei ( R F , X X , 563). See § 73, 4. 9. V. L. final e and tonic i in hiatus > i: civiles > civis; auditis > ouvides > ouvis; finés > fits > fins. 10. V. L. final e in hiatus with a, o or u > i: vadit > vai; canes > edes [kij$]; sòlés > sóis; padulés ( D u C a n g e ; for paludés) > pauis; *amabatis (for amabatis) > amàveis. 11. V. L. final e + a following nasal c o n s o n a n t > [è],

§ 47-48]

FINAL VOWELS

49

t h e n [ëj] a n d in t h e r e g i o n b e t w e e n L i s b o n a n d [Êj]: dêbent > devem. 47.

Coimbra

V. L . FINAL J (CL. L . i).

1. V . L . final i > e: habuï > houve; servi (2d sg. i m p v . ) > serve; illi > Ihe.

amasti

>

amaste;

A. This change was not completed until the beginning of the f o u r t e e n t h century (Grund, I, 953). 2. V . L . final i p r e c e d e d b y a single c, s or / ie\\: feci > fiz; pôsui > *posi > pus; sali (2d sg. i m p v . ) > sal (old). 3. V. L. final i in hiatus w i t h a tonic vowel > i: amàvi amai > amei, fui > fui.

>

A. If the tonic vowel was e and the hiatus arose t h r o u g h the fall of n, t h e j of the resultant ei fell: veni (2d sg. impv.) > *vii > vem. Cf. the change of beito to bento (§ 92, 7 A). At a later d a t e final em came to be pronounced again with j: vem > [vij] (§ 34, 10). 4. V. L. final i a n d t o n i c i in h i a t u s > i: vidi > vii > vi. A. If the hiatus arose through the fall of n, the result was i: veni > vii > vii (§ 35, 4) > vim (§ 78, 2). 48.

V . L . F I N A L o ( C L . L . o, u, AND O).

1. V . L . final o > P t g . o [u]: amo > amo; casiim cito > cedo; sumus > somos.

>

caso;

A. This o became [u] as early as the twelfth c e n t u r y (Behr, 15). Cf. Theoria, 23; R F , X X I I , 405-406. 2. V. L . final o in h i a t u s w i t h a o r { > u [w]: caelum céu; malum > mau; deus > deus; meum > meu.

>

A. T h e letter u appeared first in words in which the hiatus existed in L a t i n ; it appeared much later in words in which the hiatus arose through the fall in Portuguese of an intervocalic consonant. Accordingly, in medieval documents we commonly find the spellings deus, meu, etc. alongside the spellings ceo, mao, etc. And in the meter of the early Cancioneiros, eu in deus, meu, etc. is a d i p h t h o n g while the vowels of ceo, mao, etc. count

50

PHONOLOGY

[§48

as separate syllables. Neither the sound [w] nor the letter u developed in words in which the tonic vowel in hiatus was e in Vulgar Latin: credo > creo > creio [kr^ju]; plenum > cheo > cheio [ / e j u ] , The 3d sg. pret. ending -eu [ew] is an Old Ptg. development in which the [ w ] arose in imitation of the endings -ou [ow] and -iu [iw]. There were four combinations of tonic e and final o (or u) in early Portuguese: [ew], e.g., meu (§ 34, 7 A); [eu], e.g., ceo (§ 34, 7 B); [eu], e.g., creo (§ 35, 7); and [ew], e.g., defendeu (§ 166, 3). The sound [ew] must have become [ew] before [eu] became [ew], else meu and ceu would both have the same diphthong today. And the sound [eu] must have remained dissyllabic, for if it had not, it would be the same today as the third plural preterit ending -eu [ew] and the diphthong ei would not have developed. In Modern Portuguese these four combinations of tonic e and final o have become respectively [ew], [ew], [eju] and [ew], B. In hiatus with tonic / final o, although still spelled o, became [ w ] , and in some regions [ u ] with an intercalated [ j ] before it: jllum > jio [fiw] and [ f i j u ] ; rlvum > no [riw] and [riju], 3. V. L . final p and tonic o in hiatus > o: *avulum *avblum (cf. § 45, 2 A) > avoo (§ 37, 5 A) > avo.

>

A. If the hiatus arose through the fall of n, the result was o (§ 78, 2): donum > doo > dom; bonum > boo > bom. B. If the hiatus existed in Latin, the result was ou: duds > dous (old); suum > sou (old). 4. V. L . final p and posttonic o in hiatus > o [u]: > oragoo > orago; -populum > foboo > povo.

oraculum

5. V. L. final p and tonic u in hiatus > « : crudtim culum > cu; nudum > nu.

>

cru;

A. If the hiatus arose through the fall of n, the result was u: unum > uu > um. Ptg. um is not the result of apocope like Sp. un but is the phonological equivalent of Sp. uno. See § 78, 2 B. 6. V. L. final o sometimes b e c a m e e: tosium > toste ( o l d ) ; quomodo > come (old a n d dialectal); magistrum > mestre; miraculum > milagre; liberum > livre; contentum > contente; saeculum > segre (old).

§ 48-50]

SECONDARY TONIC

51

A. This change has been a t t r i b u t e d to the influence of other adverbs such as onde (from -unde) and tarde (from tarde) particularly in alliterative expressions such as tarde ou toste and to the influence of French or Provençal words. B. In some words in M o d e r n Galician o and e alternate, e.g., milagro and milagrt (L-F, 24, footnote). 7. V. L. final o s o m e t i m e s f e l l : anellum > anel; linteolum > lençol; lusciniolum > rouxinol; vascellum > baixel. A. Perhaps this O first became e, which then fell according to § 46, 2. It is more likely, however, t h a t these words are borrowings from French or Provençal. A similar change is found in Vulgar L a t i n : barbarus non barbar ( A p P r ) ; figulus non figel (ApPr); masculus non mascel (ApPr). 49.

C L . L . F I N A L Û.

T h e r e a r e n o cases w h e r e C l a s s i c a l L a t i n final u n a c c e n t e d u c o m e s i n t o P o r t u g u e s e . T h e a c c u s a t i v e s i n g u l a r of n e u t e r n o u n s a n d t h e a c c u s a t i v e p l u r a l of m a s c u l i n e n o u n s of t h e f o u r t h d e c l e n s i o n h a v e n o d e r i v a t i v e s in P o r t u g u e s e , as n o u n s of t h i s d e c l e n s i o n j o i n e d t h e s e c o n d d e c l e n s i o n in V u l g a r L a t i n (§ 13, 1). A. M e y e r - L u b k e (M-L, Intro, § 50) mention the O l t . plural mano o but G r a n d g e n t elsewhere ( G i t , declension and cites the form manos SECONDARY 50.

THE

§ 130) and G r a n d g e n t ( G i t , to prove t h a t Cl. L. u became § 156) points to the shift in in Vulgar Latin (GVL, § 355). TONIC

S E C O N D A R Y T O N I C A N D ITS

DISAPPEARANCE.

1. I n e a r l y P o r t u g u e s e , w h e r e t h e t o n i c s y l l a b l e w a s a p p a r e n t l y less s t r e s s e d t h a n in M o d e r n P o r t u g u e s e , al! a t o n i c syllables h a d m o r e v i t a l i t y t h a n t h e y h a v e t o d a y . A seco n d a r y a c c e n t fell on t h e s e c o n d s y l l a b l e b e f o r e t h e t o n i c , as in V u l g a r L a t i n ( G V L , § 153). T h i s is r e v e a l e d b y t h e f a c t t h a t t h e i n t e r t o n i c in h i a t u s w i t h t h e v o w e l of t h i s

52

PHONOLOGY

[§ 50-52

syllable was sometimes assimilated to it as t h e final was t o t h e tonic (§ 60, 5). I t is also revealed b y t h e fact t h a t a p a l a t a l i z e d c o n s o n a n t did not h a v e t h e s a m e effect on a n e in this syllable t h a t it had on pretonic e (cf. § 41, 3 A) : degenerar, jejuar, mexedor, mexicano ( A S t N S , C X X V , 374); in all of these words this e is p r o n o u n c e d [a], A. In s o m e words the secondary accent fell irregularly on a syllable which bore the primary accent in a related or root word.

2. T h e s e c o n d a r y tonic b e c a m e a p r e t o n i c initial vowel in words in which syncope of t h e i n t e r t o n i c t o o k place. A n d t h e vowel resulting f r o m c o n t r a c t i o n or f r o m assimilation a n d contraction of t h e s e c o n d a r y tonic a n d t h e intertonic likewise became a pretonic initial. T h u s t h e n u m b e r of words bearing a secondary a c c e n t was g r e a t l y reduced in early P o r t u g u e s e . W i t h t h e increased stress on the tonic vowel t h a t developed in t h e sixteenth c e n t u r y , t h e seconda r y a c c e n t continued to disappear, with t h e result t h a t all words (except adverbs in -mente a n d a u g m e n t a t i v e s a n d d i m i n u t i v e s with an i n t e r c a l a t e d z) now h a v e b u t one accent. POSTTONIC 51.

PENULT

S Y N C O P E OF T H E P O S T T O N I C P E N U L T I N V U L G A R

LATIN.

T h e posttonic penult (a, e, i, o or u) fell w h e n s t a n d i n g b e t w e e n certain pairs of c o n s o n a n t s in the Vulgar L a t i n p e r i o d : alterum > outro; dominum > dono; leporem > lebre; oculum > olho; positum > posto; viridem > verde. Cf. § 8. 52.

S Y N C O P E OF T H E P O S T T O N I C IN

PORTUGUESE.

1. If t h e posttonic penult was e or i (CI. L. e or i), preceded b y I, m, n or r, or preceded b y c a n d followed by t, it fell in t h e late Vulgar L a t i n or early P o r t u g u e s e p e r i o d : aliquod > algo; *pullcam (for pulicem) > pulga; gallicum > galgo; amites > andes > andas (with shift in declension); domitum > dondo ( R L , X X V I I , 30); llmites> lindes; antmam > alma; manicam > manga; erigo > ergo; sericum >

POSTTONIC PENULT

§ 52-53] sirgo;

placitum

> prazo.

S3

Cf. Behr, 46-47.

A. T h i s c h a n g e t o o k place before t h e t i m e of t h e fall of i n t e r vocalic I and n b u t a f t e r t h e t i m e of t h e voicing of i n t e r v o c a l i c [k] a n d t. B. T h e p o s t t o n i c b e a r s t h e s a m e relation to t h e t o n i c as t h e final in p a r o x y t o n e s does t o t h e tonic. C o m p a r e t h e fall of final e a f t e r /, n and r (§ 46, 2). c. If m e t a t h e s i s h a d n o t t a k e n place in *retinam, syncope would p r o b a b l y not h a v e o c c u r r e d . T h e d e v e l o p m e n t was acc o r d i n g l y : *retinam > *renida > renda. In Spanish t h e o r d e r was t h e reverse; m e t a t h e s i s t o o k place a f t e r s y n c o p e : *retinam > riedna > rienda. D. G r a n d g e n t ( G i t , § 68) finds t h a t syncope took place u n d e r similar conditions in e a r l y I t a l i a n .

2. If the e was preceded by m but followed by a short n, it d i d homines

n o t f a l l : feminam > jemea; > homees > homens.

geminum

>

g'emeo;

A. T h e nasalization of t h e p e n u l t b y t h e a d j a c e n t nasal cons o n a n t s m a y h a v e increased its resistance t o syncope. 53.

S U R V I V A L OF T H E P O S T T O N I C

PENULT.

All posttonic penults which did not fall in Vulgar Latin and in which the conditions of § 52, 1 did not obtain, remained in P o r t u g u e s e : angelum > angeo > anjo; biferam > bebera; cubitum > covedo ( o l d ) ; debitam > divida; decimum > dizimo; *dubitam > duvida; duracinum > durazio; fraxinum > freixeo > freixo; juvenes > jovees > jovens; legitimum > lidimo; persicum > pessego; *retinam > redea; -abilem > -avel; -ibilem > -ivel; capitulum > cabidoo > cabido; maculam > -magna; nebulam > nevoa; periculum > perigoo > perigo; populum > povoo > povo; spatulam > espadua; tabulam > tabua. C f . G r u n d , I , 9 5 7 , § 107. A. T h e p e n u l t d i s a p p e a r e d in some f o r m s t h r o u g h a s s i m i l a t i o n and c o n t r a c t i o n , b u t n o t b e c a u s e of syncope. B. Freixeo b e c a m e freixo t h r o u g h a b s o r p t i o n of t h e e of t h e p e n u l t b y [$] a n d angeo b e c a m e anjo t h r o u g h a b s o r p t i o n of t h e e of t h e p e n u l t b y [3].

54

[ § 53-55

PHONOLOGY

c. In most of these words in Spanish the posttonic penult fell: breva, codo, deuda, diezmo, duda, durazno, fresno, rienda, -able, -ible, cabildo, niebla, peligro, pueblo, espalda, tabla. For syncope of the posttonic penult in Spanish, see Manual, §§ 25-26. D. On the preservation of posttonic -ul-, see M-L, Intro, § 133; Biblos, VII, 520. 54.

L A T E R S Y N C O P E OF T H E P O S T T O N I C

PENULT.

W i t h t h e g r e a t l y intensified stress a c c e n t which developed in t h e . s i x t e e n t h c e n t u r y , t h e p e n u l t of p r o p a r o x y t o n e s b e g a n t o fall q u i t e g e n e r a l l y in colloquial a n d dialectal P o r t u g u e s e , w h e t h e r t h e vowel was a, e or o. T h i s c h a n g e is n o t c o m m o n l y r e v e a l e d in spelling, which h a d become f a i r l y fixed b y p r i n t i n g a n d was b e i n g m o r e a n d more i n f l u e n c e d b y L a t i n . H e r e are s o m e e x a m p l e s of this c h a n g e : P t g . arvore > arvre; P t g . aspero > aspro; Ptg. dialogo > diaglo; P t g . divida > divda; P t g . hospede > ospde; P t g . pessego > pesgo; P t g . sabado > sabdo; P t g . temporas > tempras; P t g . comodo > comdo. Cf. G r u n d , I, 957; R L , X I , 272; N S , X X X I V , 4 5 7 ; O p u s c u l o s , I I , 190 a n d 285. 55.

T H E V O W E L OF T H E P O S T T O N I C

PENULT.

1. V. L. p o s t t o n i c a > P t g . [b]: stomachum anatern > adem; hebdomada > doma (old).

>

estomago;

A. Hiatus with another a was followed by contraction. other a was tonic the contracted vowel became [a].

If the

2. P o s t t o n i c a a n d final e in h i a t u s t h r o u g h t h e fall in P o r t u g u e s e of a n i n t e r v o c a l i c c o n s o n a n t b e c a m e ei: -abatis > -avades (§ 154, 1) > -aveis. 3. V . L. p o s t t o n i c a a n d tonic f in h i a t u s > pelagum > peago > peego > p(go. 4. V. L. p o s t t o n i c a a n d tonic q in h i a t u s > g: monachum. > moago > moogo > mggo > mogo. A. The tonic vowel of mogo was closed by metaphony (§ 100, 7). 5. V. L. p o s t t o n i c a a n d final o in h i a t u s a f t e r t h e fall of i n t e r v o c a l i c n (which h a d nasalized t h e a) > ao [£w]: raphanum > rabao.

§ 55-57]

I N T E R T O N I C VOWELS

55

6. V. L . posttonic e (Cl. L . i) > P t g . e [a] or i: > dizimo; *dubïtam > dûvida; blbltum > bêbedo; > -pes se go. 7. V. L . posttonic vesperam > véspera. 8. V. L . posttonic ar; aspërum > asparo 9. V. L . posttonic > tâbua; nebûlam >

e (Cl. L . e) > e [a]: nëpëtam

>

n'eveda;

e followed by r > a: aërem > aar > ( o l d ) ; passèrent > pâssaro. o in hiatus with final a > [w]: tabulant névoa.

INTERTONIC 56.

decimum persîcum

VOWELS

S Y N C O P E OF T H E I N T E R T O N I C IN V U L G A R

LATIN.

T h e intertonic vowel fell in the V u l g a r L a t i n period under the same conditions as the posttonic penult, although not so e x t e n s i v e l y : calïdarium > caldeiro; computàre > contar; *consûtûram > costura; honôràre > honrar; labôràre > lavrar; liberate > livrar. Cf. § 8. A. While Classical Latin long vowels cannot be posttonic penults, they can be intertonic. 57.

S Y N C O P E OF T H E I N T E R T O N I C IN

PORTUGUESE.

1. If the intertonic vowel was e (CI. L . ë, ë or t) or i (Cl. L . i ) , preceded b y /, m or r, or preceded by c and followed b y t, it fell in the late Vulgar L a t i n or early Portuguese period: *bellïtàtem > beldade; dellcatum > delgado; jollïcàre > folgar; melimèlum > marmelo (see Nascentes) ; molinarium > molneiro > moleiro; olivaria > olveira ( o l d ) ; *salicarium > salgueiro; comitàtum > condado; seniïtarium > semedeiro > sendeiro; arênarium > arneiro; verêcundïam > vergonha; vêritàtem > verdade; *amîcitàtem (for amicitiam) > amizade; rècitàre > rezar. Cf. K J , I V (Part I ) , 3 4 6 . A. With few exceptions this change took place before the time of the fall of intervocalic I but after the time of the voicing of intervocalic [k] and t. In solteiro (from solitarïum) it was obviously very early while in soidade (old and dialectal) (from

56

PHONOLOGY

solïtâtem) t h e intertonic remained exceptionally late and was t h e r e b y preserved. B. T h e intertonic vowel bore the same relation to t h e secondary tonic as the final did t o the tonic; it was sometimes assimilated to t h e secondary tonic (§ 50, 1) a n d fell under conditions somew h a t similar to those under which final e and i fell (cf. § 46, 2 and § 47, 2). c. If metathesis had not t a k e n place in sïbïlâre, syncope would probably not have occurred. T h e development was accordingly: sïbïlâre > *silibare > silvar. T h e intertonic was reintroduced in modern olxveira through t h e influence of the word oliva. In its regular development in Portuguese, judïcâre became juigar ( R L , X X I I I , 27, s.v. desjuigado), which was replaced at an early d a t e by judgar, a borrowing f r o m Old Spanish, which later became juigar in Portuguese. D. In some words syncope of the intertonic was avoided because of the unpronounceable combination of c o n s o n a n t s t h a t would have resulted f r o m it, e.g., aprïcàre > abrigar; neglegentïam > negrigença (old and popular). Words in which such syncope did occur, with the consequent loss of one of the consonants, are extremely exceptional and are probably borrowings f r o m Spanish, French, or P r o v e n ç a l : aestïmâre > esmar; mastïcàre > mascar; vindicate > vingar. 2. If i n t e r t o n i c e w a s p r e c e d e d b y m b u t f o l l o w e d b y a s h o r t n , it d i d n o t f a l l : nominate > nomear; rûmïnâre > rumiar; seminâre > semear. 3. If i n t e r t o n i c e w a s p r e c e d e d b y n, it d i d n o t fall a s t h e p o s t t o n i c p e n u l t d i d : *adrepoenïtêre > arrepèedeT > arrepender; generàlem > geeral > gérai; ïnïmicum > eimigo (old) ; monimentum. > moimento; vanitdtem > vaidade. A. T h i s seems to indicate either t h a t syncope of the i n t e r t o n i c took place later t h a n syncope of the posttonic penult or t h a t intervocalic n fell earlier before t h e intertonic t h a n before t h e posttonic p e n u l t . B. T w o like vowels contracted a f t e r the fall of c o n s o n a n t a l N, e.g., arrepender; compare Sp. arrepentir, in which the i n t e r t o n i c fell before the time of the voicing of intervocalic t, a l t h o u g h

§ 57-58]

INTERTONIC VOWELS

57

M e n e n d e z Pidal ( M a n u a l , § 54, 1) a t t r i b u t e s the t to learned influence. c. T h e development of bondade (from bonitatem) was probably influenced by the a d j e c t i v e bom.

4. Where there were two vowels between the initial and the tonic vowel, the first was a secondary tonic, the second a n i n t e r t o n i c : caballicare ( D u C a n g e ) > cavalgar; *humilitosum > humildoso; legalitatem > lealdade; *adluminari (for illuminare) > alumiar. A. In heredltare, intervocalic d fell and intervocalic t became d; t h e n the vowel resulting f r o m contraction of the two f's fell, because it was an intertonic vowel preceded by r: heredltare > *hereedar > *heredar > herdar. Comungar (from communicdre) developed by analogy with forms in which the e was not intertonic b u t posttonic, e.g., comungo (from communlco). T h e form irmandade (from germanltatem), which replaced an older and regular irmdidade, developed t h r o u g h the influence of irmd; a similar d e v e l o p m e n t took place in orfandade through the influence of ¿rfa. 58.

SURVIVAL

OF T H E

INTERTONIC.

All intertonic vowels which did not fall in Vulgar Latin and in which t h e conditions of § 57 did not obtain, remained in P o r t u g u e s e : capitdlem > cabedal; clvitatem > ciidade > ctdade; cupiditiam ( D u C a n g e ) > *cobeifa > cobiifa > cobiqa; medicinam > meezinha > mezinha; mirdbilia > maravilha; nuditatem > nuydade ( D C , 3 6 2 ) ; praecdnari > pregoar; salutare > saudar; suspirare > suspirar; traditorem > trcidor ( o l d ) ; voluntatem > voontade > vontade. A. In some of these words the intertonic vowel finally disappeared t h r o u g h assimilation and contraction, b u t not because of syncope. B. T h e intertonic did not fall in Portuguese and Spanish in learned or semi-learned words, e.g., caritdtem > P t g . carxdade and Sp. caridad. I t generally fell in Spanish in popular words

58

PHONOLOGY

[§ 59-60

unless it was a (Manual, §§ 23-24): capìtàlem > cabdal > caudal; cïvitàtem > abdad > ciudad; cup'id'ifiam (Du Cange) > cobdicia > codicia. Ptg. caudal was borrowed from Spanish while Ptg. coudel ( f r o m capïtellum) was probably borrowed from Provençal. 59.

LATER

SYNCOPE

OF T H E

INTERTONIC.

W i t h the greatly intensified stress accent which developed in t h e s i x t e e n t h c e n t u r y , t h e i n t e r t o n i c , as well as t h e p r e t o n i c initial a n d t h e vowel w h i c h h a d p r e v i o u s l y b e e n t h e s e c o n d a r y t o n i c (i.e., all p r e t o n i c vowels) b e g a n t o fall in l e a r n e d as well as p o p u l a r w o r d s w h e n s t a n d i n g b e f o r e or a f t e r a l i q u i d : P t g . cerimònia > cermónia; P t g . parecer > farcer; P t g . esperar > esprar; P t g . poderoso > podroso; P t g . coronel > cornel P t g . superior > suprior; P t g . oferecer > ofrecer; P t g . experimentar > exprimentar; Ptg. corregidor > corgidor; P t g . averiguar > avriguar; P t g . esterilidade > estrelidade; P t g . perigo > prigo; P t g . paraiso > praiso; P t g . verâo > vrâo; P t g . corôa > crôa; P t g . pelo > pio; P t g . para > pra; P t g . querer > qu'rer. T h i s c h a n g e is n o t g e n e r a l l y r e v e a l e d in spelling, w h i c h h a d b e c o m e f a i r l y fixed b y printing a n d was being m o r e a n d more influenced by L a t i n ; it is r e v e a l e d b y t h e r h y t h m of t h e v e r s e of Sa de M i r a n d a a n d Gil V i c e n t e , a l t h o u g h it is u n c o m m o n in t h e l a t e r , m o r e polished v e r s e of C a m ô e s ( S a M , p p . cxxii-cxxiii, e r r o n e o u s l y n u m b e r e d cxx a n d cxxi; Z R P h , I V , 6 0 0 ; R L , X X V I I , 126). F o r s p o r a d i c e x a m p l e s of this c h a n g e b e f o r e t h e s i x t e e n t h c e n t u r y , see C D , n o t e t o line 299; F M , I, xxiii. 60.

THE

VOWEL

OF T H E

INTERTONIC.

1. V. L. i n t e r t o n i c a a n d s e c o n d a r y t o n i c a in h i a t u s > a [a] : *panatarium > paadeiro > padeiro; sanatlvum (Du C a n g e ) > saadio > sadio. 2. V. L. i n t e r t o n i c e a n d s e c o n d a r y t o n i c e in h i a t u s > ç: crêditôrem > creedor > crçdor; medïcïnam > meezinha > mçzinha; praedicàre > preegar > pregar. 3. V. L. i n t e r t o n i c o a n d s e c o n d a r y t o n i c o in h i a t u s > p; cblóràtum > coorado > egrado.

§60-61]

I N I T I A L S H O R T CONSONANTS

59

4. T h e i n t e r t o n i c vowel was assimilated to t h e tonic as the p r e t o n i c initial w a s : *adcalescit > acaece > aqueece > aquece; cupiditlam ( D u Cange) > *cobei(a > cobii^a > cobifa. 5. T h e i n t e r t o n i c was assimilated t o t h e s e c o n d a r y tonic as t h e final was to the t o n i c : *adminaciare > ameagar > ame(ar ( R L , V I , 3 3 9 ; R L , V I I , 1 0 7 ) ; civltdtem > *ciedade > ciidade > cidade; considerate > *consierar > consiirar > consirar ( o l d ) ; divinitatem > *diviedade > diviidade > divindade. 6. V . L . querimoniam

i n t e r t o n i c e preceded or followed b y r > a: > caramunha; sacerdotem > safardote (old). INITIAL

61.

SHORT

CONSONANTS

LABIALS.

1. CI. L. initial b > P t g . b: bene

> bem;

buccam

>

boca.

A. There was some confusion between initial b and v; see Comp, 90 and section 4 A below. 2. CI. L. initial / > P t g . f : ferrum > ferro; filium > filho. 3. CI. L . initial p > P t g . p: pacem > paz; personam > pessoa. A. I n i t i a l p s o m e t i m e s b c c a m e b, p e r h a p s b y s a n d h i : polire

buir.

>

See Comp, 89-90.

4. CI. L. initial v [w] > P t g . r : vallem ver; vlnum > vinho; vitem > vide.

> vale;

videre

>

A. Initial v became b in a few words: vaginam > bainha; vessicam > bexiga; viperam > bibera (Fab, Glossary); votum > bodo. B. T h e change of initial v to g has been attributed to Germanic influence (M-L, Gram, I, § 4 1 6 ; Comp, 9 3 ) : vastare > gastar; vUiculam > guedelha; vomitare > gomitar (popular); voracem > goraz. c. Initial v became / in OPtg. femen^a (from vehementiam). The change of initial v to / is not altogether unknown in the Romance languages, e.g., Fr. fois (from vicem). Cf. Huber, § 177. The change to / in Jerrolho (from veruculum) was due to contamination by ferro.

60 62.

PHONOLOGY PALATALS

AND

[ § 62

VELARS.

1. C . L. initial c followed b y a, o, or u > P t g . c [k]: cantare > cantar; color em > for; culum > cw. A. T h e change of initial velar c to j has been a t t r i b u t e d to sandhi ( G i t , § 87): camellam > gamela; cattum > gato; caveam > gaiva. For the same change in dialectal Portuguese, see RL, X I , 140. B. T h e change to ch is found in borrowings from French: chapeu, chefe, charrua. 2. CI. L. initial c followed b y e or i (V. L . [ts]) > P t g . c [s]: centum > cento; circa > cerca; civitatem > cidade. A. Ptg. chinche (from cimlcem) was probably borrowed from It. cimice either directly or through Spanish. B. T h e assimilation of pretonic a to e took place too late to affect the development of initial c, e.g., calentem > caente > queente > quenle. 3. C . L. initial g followed b y a, o, or u > P t g . g: gallinam > galinha; gubernare > governar. F o r g followed b y u, see § 69, 4. 4. CI. L. initial g followed b y e or i (V. L. [j]) > P t g . g [3]: genuculum > geolho > joelho; gingivam > gengiva; generum > genro; gentem > gente. A. T h e changes through which this sound passed were the following: [g] > [gj] > [j] > [d 3 ] > [3]. B. T h e form germanum seems to have developed regularly at first, e.g., germao (AHP, VII, 477, A.D. 1282) and germaho (Eluc, A.D. 1288). T h e loss of the g in irmao has been a t t r i b u t e d to its intervocalic position in the frequent occurrence of germao and germaa with possessive adjectives (Grund, I, 991). 5. CI. L. initial j [j] > P t g . / [3]: januarium jurare > jurar; julium > julho.

>

janeiro;

A. Beginning with Vulgar Latin this sound and the sound coming from Classical Latin initial g followed by e or : (section 4 above) have been identical.

§ 63-65] 63.

INITIAL SHORT

61

CONSONANTS

DENTALS.

1. CI. L. initial d > P t g . d: dare dolorem

>

2. CI. L. initial t > P t g . t: taurum turrim 64.

>

> dar;

decern

>

dez;

> toiro;

tenere

>

ter;

llnguam

>

dor. torre.

LIQUIDS.

1. CI. lingua;

L.

initial

lupum

>

/ > Ptg.

I: latus

> lados;

lobo.

A. T h a t this I was longer in Old P o r t u g u e s e t h a n it is t o d a y is i n d i c a t e d b y t h e fact t h a t it was s o m e t i m e s w r i t t e n 11 and t h a t it did not fall when it b e c a m e intervocalic b y sandhi (except in u n a c c e n t e d monosyllables). Cf. § 30, 2 b. 2. CI. L. initial r > P t g . r: rivum rubeurn 65.

>

> rio; rosam

>

rosa;

ruivo.

SIBILANTS.

1. CI. L. initial s > P t g . s: sal > sal; sonum

>

sigillum

>

selo;

som.

A. T h e c h a n g e of initial s t o x, e.g., syringam > seringa > xeringa ( R L , X I , 278; R L , X I I , 307), has been a t t r i b u t e d to A r a b i c influence ( R A , V I I , 117), b u t as it occurred also in regions t h a t were free of A r a b i c influence, it is m o r e likely t h a t it was d u e t o t h e close similarity of dialectal c a c u m i n a l s to [5] ( E n t , 302). T h i s similarity a c c o u n t s for t h e c o m m o n use of A r a b i c shin in P o r t u g u e s e a l j a m i a d o texts t o represent P o r t u guese initial s a n d intervocalic ss while sin is used t o represent f. On t h e d i s a p p e a r a n c e of c a c u m i n a l s in literary P o r t u g u e s e , see K J , IV, P a r t I, 327; on its survival in dialectal P o r t u g u e s e , see P h M , I, 34, n. 1 and 189; Esquisse, § 59 e. Cf. G - D , C, § 3 2 , 4 , n. 6. B. T h e c of cerrar ( f r o m serare) developed b y c o n t a m i n a t i o n with some o t h e r word, p e r h a p s cercar (cf. N a s c e n t e s ) . T h e z of zoar ( f r o m sonare) is p r o b a b l y onomatopoeic.

62

PHONOLOGY

[§ 65-67

2. CI. L. z (Greek f ) > P t g . z or c [s]: zelum > zeo (old) and zelo; zephyrum > zéfiro; zelum > cio (cf. N a s c e n t e s ) ; fanóv > fumo. A. I t is t h o u g h t t h a t in Vulgar Latin, z sometimes became confused with d plus yod or merely yod (GVL, § 339). This m a y have occurred in gengibre (from zingiber) or the z may have been assimilated to the g of the following syllable. 66.

NASALS.

1. CI. L. initial m > Ptg. m: male > mal; metum monetam > moeda.

>

medo;

A. T h e change of initial m to « in néspera (from mespilam) goes back to Vulgar Latin as most R o m a n c e languages have forms with n ( R E W ) . B. Initial m sometimes nasalized the following vowel: matrem > mat; meant > mia > minha; mihi > mt > mim; multum > muito [müjntu]. This phenomenon is very common in popular and dialectal Portuguese: mensem > mes > mes (RL, V, 59); mensam > mesa > menza (RL, V, 58); missam > missa > minsa (Opúsculos, II, 313); milium > muu > mü (RL, X X V I , 249); forms of the verb medir (RL, X X V I , 254, s.v. mil). Cf. Esquisse, § 40 a. 2. CI. L. initial n > Ptg. n: natáre novo.

> nadar;

novum, >

A. T h e change of initial n to m in mastruqo (from nasturtium) goes back to Vulgar Latin as several Romance languages have forms with m ( R E W ) . B. Initial n sometimes nasalized the following vowel: nec > ne > nem; *ne-quem > neguem > nenguem > ninguém; nidum > nio > ninho; nudum > nüu (BF, I, 46). INITIAL 67.

CONSONANT

+

CONSONANT

GROUPS

I.

1. CI. L. initial bl > P t g . br: blandum > brando; > bredo; blasphemare > brasjamar (Fab). A. T h e form blasfemar is learned.

blitum

§67]

INITIAL CONSONANT GROUPS

2. CI. L. initial cl > P t g . ch or clamare > chamar; clavum > cravo.

cr:

clavem

63 >

chave;

A. T h e development of cl was similar to t h a t of pi (see sections 5 A and 5 B below). B. In learned words cl has remained unchanged: clima > clima; clarum > claro; classem > classe. 3. Cl. L. i n i t i a l ft > P t g . ch or jr: flagrdre ( G V L , § 292) > cheirar; flammam > chama; flaccum > fraco; phlegma > freima. A. T h e development of fl was similar to t h a t of pi (see sections 5 A and 5 B below). B. In learned words fl has remained unchanged or has been restored: flavum > flavo; florem > flor; fluidum > fluido. 4. Cl. L. i n i t i a l gl > P t g . I or gr: glandem > lande; glaream > leira; globellum ( D u C a n g e ) > lovelo > novelo; gluten > grude; glossam > grossa (old). See section 5A below. A. In learned words gl has remained unchanged: glob-um > globo; gloriam > gloria. 5. Cl. L . initial pi > P t g . ch o r pr: plagam > chaga; plenum > cheio; plus > chus ( o l d ) ; plagam > praia; placere > prazer. A. Words in which pi, cl, gl, and / / became pr, cr, gr, and fr respectively correspond in general to Spanish words in which no change took place (Hanssen, § 19, 3 ) ; they are less popular and later, and were perhaps used by a different class of people (RL, XXVIII, 23). B. T h e development of pi to ch seems to have been as follows: pi > pll (lengthening of /) > [pA] (whence Castilian [X]) > pi > ptch > tch > ch (Ent, 287-288; RC, X I , 1120-1130; Z R P h , X L I I , 228-229; M-L, G r a m , I, § 422). T h e last change, from [t$] to [$], took place about 1700 (Grund, I, 972, n. 2) b u t [tS] is still heard in certain dialects (RL, X I X , 173; RL, X X V I I , 99). c. In learned words pi remained unchanged or was restored: plumam > pluma; plantdre > plantar. D. T h e word lhano (from planum) was borrowed from Spanish.

64 68.

PHONOLOGY CONSONANT +

[5 68-69

r.

1. CI. L. initial br > Ptg. br: bracas > bragas; bracchium braqo. 2. CI. L. initial cr > P t g . cr: credo > creio; creatum > criado. >

A. In some words initial cr became gr: crassum > graxo; cratem > grade; cretam > greda; quiritare > *c'ritare (GVL, § 229) > gritar. 3. CI. L. 4. CI. L. > fruto. 5. CI. L. > grdo. 6. CI. L. > pre(o. 7. CI. L. tredecim > 69.

VELAR

initial dr > P t g . dr: dracdnem > dragao. initial fr > P t g . fr: fraxinum > freixo; fructum initial gr > P t g . gr: graecum > grego;

granum

initial pr > P t g . pr: pratum

pretium

initial tr > P t g . tr: traditorem treze.

> prado; > treidor

(old);

+

1. CI. L. initial qu followed b y tonic a > P t g . qu [kw]: quattuor > quatro; quando > quando; quasi > quasi. A. There is a tendency for u in this position to fall in dialectal Portuguese (Esquisse, § 48 d). 2. CI. L. initial qu followed by prctonic a > Ptg. c [ k ] : quattuordecim > calorie; quaternum > caderno; quantitatem > cantidade (old); qualitatem > calidade (old); quam magnum > camanho (old). A. The form quamanho developed through the influence of tonic quao; MPtg. quantidade and qualidade are regressive words. 3. CI. L. initial qu followed b y e or i > P t g . qu [k]: quaerendum > querendo; quid > que; quintum > quinto. 4. Initial gu [gw] occurs only in words of G e r m a n i c origin. W h e r e the following vowel is e or i, t h e [w] h a s d i s a p p e a r e d : wardon > guardar; *werra (REVV) > guerra; wisa > guisa.

§ 70-72] 70.

INTERVOCALIC SHORT CONSONANTS

DENTAL +

65

J.

CI. L . i n i t i a l d + i > P t g . g [3] or j: deorsum > juso ( o l d ) .

diaria

>

geira;

A. See § 12, 3. The development is the same as that of initial ; (§62, 5). B. The word jornal could be derived directly from diurnalem but is a borrowing from French. 71.

PROSTHETIC

e.

1. T h e e w h i c h w a s p r e f i x e d in V u l g a r L a t i n t o i n i t i a l c o n s o n a n t g r o u p s c o n s i s t i n g of s plus o n e or t w o c o n s o n a n t s (§11), b e c a m e e [|] in P o r t u g u e s e : scutum > escudo; stüdium > estudo; strictum > estreito; speráre > esperar; sphaeram > espera (old) a n d esfera; scriptum > escrito; smaragdum > esmeralda ( w i t h s h i f t of d e c l e n s i o n ) . A. This e has disappeared in many dialects particularly in the north (Grund, 1,955; Esquisse, § 50 a; RL, II, 19 and 40; RL, IV, 34-35; Opúsculos, II, 189). 2. L a t . i n i t i a l scl > P t g . escr: sclavum > escravo. 3. T h e u of i n i t i a l squ f o l l o w e d b y p r e t o n i c a, f e l l : squamare > escamar; f o l l o w e d b y t o n i c a, it r e m a i n e d : squalum > esqualo. 4. P r o s t h e t i c e is f o u n d w i t h i n i t i a l sc f o l l o w e d b y e o r i: scindo > escindo. B u t it did n o t d e v e l o p in l e a r n e d w o r d s w i t h i n i t i a l sc f o l l o w e d b y e o r i: scenam > cena; sceptrum > cetro; scientiam > ciencia; scillam > cila; scintillam > ' S p . centella > centelha. INTERVOCALIC SHORT 72.

CONSONANTS

LABIALS.

1. CI. L . i n t e r v o c a l i c b > P t g . v: debet > deve; fabam Java; habere > haver; nübem > nuvem.

>

A. Some words have v in Old Portuguese but b in Modern Portuguese by regression: bibere > bever (old) and beber; sébum >

66

PHONOLOGY

sevo (old) and sebo; tabulam > tâvoa (old) and tabua. T h e intervocalic b of beber m a y have developed by assimilation to the initial b. I n t e r v o c a l i c b has remained in learned and semilearned words: diabolum > diabo; scabellum > escabelo. 2. Cl. L . i n t e r v o c a l i c v [w] > P t g . v: avinam lavâre > lavar; novem > nove.

>

aveia;

A. Intervocalic v followed by u n a c c e n t e d u regularly fell in Vulgar L a t i n , sometimes t o be restored b y analogy with a form of the word in which the v was not followed by u ( G V L , § 324). Such restoration did not t a k e place in P o r t u g u e s e in t h e derivative of t h e ending -ïvum ( C o m p , 108): aestïvum > estio; fugitïvum > fugidio; rivum > rio; sanativum ( D u Cange) > sadio. In adjectives ending in -ïvum, t h e v of the feminine fell b y analogy with the masculine. In vivo (from vxvum) t h e v was preserved or restored t h r o u g h t h e influence of the verb viver. Cl. L. bovem became V. L. boe (Vok, II, 479), whence P t g . boi; p e r h a p s the v fell by dissimilation ( M P , X I , 349). In cidade (from cïvïtàtem) the v p r o b a b l y fell because it stood between two i's (cf. K e n t , § 181, V I I ) . 3. Cl. L . i n t e r v o c a l i c / > P t g . v: aurïjïcem > defensam > devesa; profectum > proveito; Stephânum têvâo.

ourives; > Es-

A. T h e b of bêbera (from biferam) developed probably by assimilation to t h e initial b. T h e form defesa (from defensam) is semi-learned, h a v i n g been influenced by the learned defensa or the French défense, with b o t h of which it has meanings in common. Leite finds t h a t as place-names Defesa and Devesa occur in s e p a r a t e regions ( R L , X I , 354). T h e b of rdbâo (from raphânum) m a y h a v e arisen t h r o u g h t h e Latin pronunciation p of Greek ph (GVL, § 332). 4. Cl. L . i n t e r v o c a l i c p ( V . L . b) > P t g . b: apîciïlam abelha; lupum > Idbo; ripam > riba; sapit > sabe.

>

A. In some words this p developed f u r t h e r to v: -nepetam > n'eveda; pbpulum > poboo > povo; praesaepem > preseve (old) b u t modern pesebre; propinquum > provinco (Elue); scopam > escôva.

§ 72-73]

INTERVOCALIC SHORT

CONSONANTS

67

T h i s a d d i t i o n a l s t e p m a y h a v e been occasioned b y dissimilation in some of these w o r d s (cf. R L , X X X I V , 306). 73.

PALATALS

AND

VELARS.

1. CI. L. intervocalic c f o l l o w e d by a, o o r u ( V . L . g) P t g . g: amicum > amigo; caecum > cego; dico > digo. A. In pouco ( f r o m paucum) a n d rouco ( f r o m raucum) p r e v e n t e d f r o m voicing b y t h e u of t h e d i p h t h o n g .

>

the c was

2. CI. L. i n t e r v o c a l i c c f o l l o w e d b y e o r i ( V . L . [ts]) P t g . z: dicis > dizes; placere > prazer; vices > vezes.

>

A. T h e changes t h r o u g h which this sound passed in Portuguese were the f o l l o w i n g : [ t s ] > [ d z ] > [ z ] . See Hanssen, § 17, 10. B. T h e c of receber d e v e l o p e d as initial c b e c a u s e t h e word was felt t o be a c o m p o u n d . T h e c of racimo ( f r o m racemum) is regressive as t h e f o r m razimo is f o u n d in Old P o r t u g u e s e . The c of recente ( f r o m recentem) is l e a r n e d . 3. CI. L. i n t e r v o c a l i c g f o l l o w e d b y a, o o r u > P t g . g: legumen > legume; negare > negar; plagam > chaga. A. I n some words this sound fell in V u l g a r L a t i n : ego > eu; legalem > lial; lltlgdre > lidar; strlgam > estria. See G V L , § 263. For t h e t h e o r y t h a t this g r e m a i n e d a f t e r t h e a c c e n t a n d fell before it, see R H i , L X X V I I , 146. T h e word rua ( f r o m rugam) belongs to t h i s list unless it is a borrowing from French. 4. CI. L. i n t e r v o c a l i c g f o l l o w e d b y e o r i ( V . L . [j]) f u s e d w i t h t h e f o l l o w i n g e o r i: magistrum reginam > vinte; frio;

> rainha;

sigillum

cogitare

> cuidar;

gregem

> grei;

legem

> seello

> maestre > selo; vlginti

digltum > lei; regem

> dedo; >

>

mestre; >

frigidum

vnnte >

rei.

A. T h i s c h a n g e t o o k place in t h e V u l g a r L a t i n period in most of these words ( G V L , § 259). B. I n m a n y of t h e s e w o r d s t h e r e was a f u r t h e r fusion with a preceding e or i. c. I n learned and semi-learned words V u l g a r L a t i n intervocalic [j] b e c a m e [3]: mugilem > mugem; -uglnem > -ugem; -aglnem >

68

PHONOLOGY

[ § 73-75

-agem; fugere > fugir ( w i t h s h i f t of c o n j u g a t i o n ) ; mugire > mugir; regendum > regendo; vigilare > vigiar. C o r n u e x p l a i n s s o m e of t h e s e f o r m s as p o p u l a r b y t h e r u l e t h a t [j] p r e c e d e d b y u b e c a m e [3] ( G r u n d , I, 991). T h i s w o u l d e x p l a i n mugem, if it is a p o p u l a r w o r d as B e h r c o n t e n d s ( B e h r , 3 2 - 3 3 ) . D. T h e v e r b velar ( f r o m vigilare) is a s e m i - l e a r n e d w o r d ( b o r r o w e d f r o m S p a n i s h ) in w h i c h t h e r e g u l a r c h a n g e o c c u r r e d , t h a t is, [j] a n d t h e f o l l o w i n g v o w e l c o n t r a c t e d a n d t h e r e s u l t a n t v o w e l again contracted with the preceding vowel. P t g . j [3]: *ajat

5. CI. L. i n t e r v o c a l i c j > ha]a;

cujum

> cujo;

jejunare

>

( f o r habeat)

>

jejuar.

A. T h i s s o u n d f u s e d w i t h a p r e c e d i n g u n a c c e n t e d e: dejectare > deitar; pejorem > peor > pior ( § 9 9 , 6). T h e j of majdrem was p r o b a b l y lost in i m i t a t i o n of peor: majorem > maor > moor > mor. T h e f o r m s maior a n d peior (old) a r e n o t p o p u l a r ( B e h r , 4 2 ) . In t h e d e v e l o p m e n t of majdrinum t o meirinho, j c e a s e d t o b e i n t e r v o c a l i c a f t e r t h e fall of i n t e r t o n i c o in V u l g a r L a t i n . T h e d e v e l o p m e n t of Maio ( f r o m Majum) is n o t c l e a r . 74.

DENTALS.

1. CI. lampada dum

L.

intervocalic

> lampaa

> nuu

> ntl;

d (V.L.

> lampa;

[d])

nidum

*peducam

>

fell > nlo

in

Portuguese:

> ninho;

nu-

peuga.

2. CI. L. i n t e r v o c a l i c t ( V . L . d) > P t g . d [d]: catenam cadeia;

natam

> nada;

ventatem

>

>

verdade.

A. T h i s d fell in t h e s e c o n d p l u r a l of all t e n s e s in w h i c h it r e m a i n e d i n t e r v o c a l i c , a d e v e l o p m e n t w h i c h m a y h a v e b e e n of d i a l e c t a l origin, t h a t is, m a y h a v e a r i s e n in s o m e r e g i o n w h e r e i n t e r v o c a l i c d ( f r o m CI. L . t) a l w a y s fell, as is t h e c a s e t o d a y in S a n M a r t i n de T r e v e j o a n d o t h e r p a r t s of S p a i n ( R L , X X V I , 249). T h e s a m e c h a n g e t o o k p l a c e b u t a t a m u c h e a r l i e r d a t e in t h e d e v e l o p m e n t of *metipsimum t o meesmo, n o w mesmo. B. I n couto ( f r o m cautum) b y t h e u of t h e d i p h t h o n g . 75.

the t was prevented f r o m voicing

LIQUIDS.

1. CI. L. i n t e r v o c a l i c / fell in P o r t u g u e s e : dolor em > dor; salire

> sair;

soles

>

sois.

>

door

§ 75-77]

INTERVOCALIC SHORT

CONSONANTS

69

A. Intervocalic /, after becoming a t t a c h e d to the preceding vowel, became gutturalized as / is t o d a y at the end of a syllable; this gutturalized intervocalic / fell in the course of the t e n t h century (ZRPh, X X V I I I , 603; Li 9 oes, 291; E n t , 288; Sachs, 22). B. In learned and semi-learned words intervocalic / did not f a l l : *alecrem

( f o r alacrem)

> alegre;

calbrem

> color;

salarxum

> saldrio; scolam > escola. Along with these there are some regressive words with /, which have replaced popular words in w h i c h / h a d f a l l e n : dolorosum > doroso tium > seen$o ( o l d ) a n d silencio; ielum

( o l d ) a n d doloroso; silen> zeo ( o l d ) a n d zilo. It

is possible t h a t the I of *alecrem lengthened in Vulgar Latin in imitation of words with the prefix ad- plus initial /; t h a t is the explanation G r a n d g e n t gives of the long I of I t . allegro ( G I t , § 100). Palavra (from parabolam) is probably a borrowing from Spanish as the t y p e of metathesis it displays is characteristic of Spanish; the regular Portuguese form of the word was paracoa (old). Contamination also contributed t o the preservation of I: thus the / of pelo (from pUum) is explained b y the I of cabelo ( f r o m caplllum),

t h e I of melao

( f r o m melonem)

2. CI. L . i n t e r v o c a l i c r >

P t g . r: carum.

b y t h e I of

mel

(from mel). > caro;

erat

>

era. 76.

SIBILANTS.

CI. L . i n t e r v o c a l i c s > > coisa;

rosarn

>

P t g . s [ z ] : *ausdre

> ousar;

causa

rosa.

A. In Old Portuguese and in certain dialects of the north of Portugal, intervocalic s became [3] ( F M , I, xxiv; Esquisse, § 59 e; R L , X I , 277; R L , X I I , 307; R L , X X V I I , 101). This change is found in Spanish (JdeV, 74). I t was due to the close similarity of dialectal cacuminal z [z] to [3] ( E n t , 302). 77.

INTERVOCALIC M.

CI. L . i n t e r v o c a l i c m > >

P t g . m: fumare

> fumar;

sumus

somos.

A. Intervocalic m nasalized the preceding vowel (CA, I, xxii, n. 4) but this nasalization soon disappeared, except in dialects (cf. Opusculos, II, 313).

70

PHONOLOGY

[§ 77-78

B. Intervocalic m sometimes nasalized the following vowel as an initial m did: vimen > vime > vimem (old); amiddulam (ApPr) > amendoa. 78.

I N T E R V O C A L I C N.

1. I n t h e c o u r s e of t h e t e n t h c e n t u r y , intervocalic n nasalized t h e p r e c e d i n g vowel a n d fell. T h e nasal resona n c e of t h e v o w e l persisted a n d spread t o t h e following vowel. F r o m this p o i n t t h e r e were several d i f f e r e n t develo p m e n t s d e p e n d i n g on t h e n a t u r e of t h e vowel c o m b i n a t i o n , t h e p o s i t i o n of t h e accent, a n d t h e c o n s o n a n t following t h e second vowel. F o r p r e v i o u s studies of this p r o b l e m , see B e h r , 4 4 - 4 9 ; R C , X I , 8 4 - 9 3 ; M L N , L, 1 6 - 1 7 ; Z R P h , X L I , 5 5 5 - 5 6 5 ; N S , X I , 129-153; E n t , 289; R P h , I, 35-39. 2. If b o t h vowels were like vowels a n d t h e first was t o n i c , t h e second final, t h e nasal resonance r e m a i n e d a n d t h e vowels c o n t r a c t e d ; bonum > boo (§37, 6) > bom; lanam > laa > la; tines > tees > tens. A. The nasal resonance has fallen in some of these words in southern Galician and certain other dialects (Fab, Glossary, s.v. boo). B. Because unaccented final e became i and contracted with tonic j (§ 46, 9), words like fines belong to this category -.fines > fiis > fins. And because unaccented final o became u and contracted with tonic u (§ 48, 5), words like unum belong to this category: unum > uno > uu > um; tribunum > tribum (BF, II, 222). c. If the first vowel was a posttonic penult and the second vowel final, the nasal resonance fell, unless both vowels were i's: sabanam (Du Cange) > savaa (FM, II, Glossary) and diaconum > diagoo (old); but homines > homees > homens; ordines > ordees > ordens. Even if they were i's the nasal resonance fell in certain dialects, e.g., home (RL, III, 59; RL, X I , 277; RL, X I I , 308; RL, X X V I I , 127; RL, X X V I I I , 224). The nasal resonance remained in orfa (from orphanam) through the influence of orfao and perhaps also of irma.

§78]

INTERVOCALIC SHORT

CONSONANTS

71

3. If t h e first vowel w a s t o n i c in a n y of t h e p a i r s a-o, o-e, a n d a-e, t h e n a s a l r e s o n a n c e r e m a i n e d a n d t h e s e c o m b i n a t i o n s l a t e r b e c a m e n a s a l d i p h t h o n g s : germanum > irmáo; manum > mdo; lectiónes > lifoes; pones > poes; canes > caes. A. T h e nasal resonance fell in these words in southern Galician. B. Instead of -ao the suffix -ano is found in learned and regressive words: kumánum > humano; castellánum > Sp. castellano > OPtg. castelháo > M P t g . castelhano. c. T h e diphthong áo developed also with the pair a-o when both vowels came after the accented syllable: orgánum > órgao; orphanum > órfáo; Stephánum > Estevao. In some dialects this unaccented ao became o, e.g., orgo, orpho (Esquisse, § 56 j ; R L , X I , 275; R L , X I I , 307). An early stage of these forms, before contraction of the vowels and before the loss of the nasal resonance, is occasionally found in Old Portuguese documents, e.g., orfoos (RL, I, 334) and orgoons (RL, III, 116). D. T h e pair o-e with both vowels after the accented syllable is found both with and without the nasal resonance in the plural of demo: daemones > démoes ( F M , II, Glossary) and demoes ( F M , II, Glossary, s.v. demoes; Rom, X I , 376). 4. If t h e first v o w e l was t o n i c i a n d t h e s e c o n d v o w e l a or o, a p a l a t a l i z e d n a s a l d e v e l o p e d b e t w e e n t h e m a n d t h e n a s a l r e s o n a n c e d i s a p p e a r e d : gallinam > gallia > galinha; vicinam > vizia > vizinha; vinum > vio > vinho. A. T h e date of this development has not been determined. In the early Cancioneiros (CA, CB and CV), some of the poems of which were written as late as the first quarter of the f o u r t e e n t h century, derivatives of Lat. -inum and -inam did not rime with words in which the sound [p] came from Lat. n plus yod. Only one exception has been pointed out, namely, camho : miho : uiho: myho, i.e., camio ; Minho : vio : Minho (CV, N o . 912), where the poor rime or assonance arose through the necessity of using a proper noun (RF, X X I I I , 341-342). It must be remembered that assonance is fairly common in this early verse (cf. C D , cxxv).

72

PHONOLOGY

W e now know that the use of the graph nh in derivatives of L a t . -tnum and -inam in the critical edition ( C A ) of the " C a n cioneiro da Ajuda" is not in conformity with the manuscript (CACarter, x i v ) ; its use in the Italian apographs ( C B and C V ) is of no significance for pronunciation at the time the poems were written, because these manuscripts are of the late fifteenth or early sixteenth century. B. If the first vowel was pretonic and the second vowel was tonic i in hiatus with a following a or o, the nasal resonance spread to all three vowels. L a t e r a palatalized nasal developed between the last two vowels in the regular w a y : litaniam > lidaia > ladainha; venibam > venia > veía > viia > viinha > vinha. c. A palatalized nasal developed with the accent reversed, t h a t is, between i (or e, which in hiatus became i ) and a following tonic a or u: divinare > adivinhar (with prosthetic a) ; ordinäre > ordinhar (Crest, Glossary); * daemomätos > demöicuios (RL, I, 3 3 8 ) > demoinhados (RL, I, 3 4 5 ) ; *ne (for nec)-ünum > ne hum > nenbum. In certain dialects of the north of Portugal and in Galician, a palatalized nasal developed between e and a following tonic i: *veníram (for veneram) > v inker a (Esquisse, § 75 r). Perhaps this was due to the complete dissimilation of e( to [je] in the regions in question before the disappearance of the nasal resonance. T h e word dinheiro had the same development: denarium > dieiro > dinheiro, and is therefore probably a borrowing from a dialect. Compare with this the regular development of vieiro (from venarium). D. J u s t as a palatalized nasal developed between tonic J and final a, a labialized nasal [m] developed between tonic u and final a: ünam > hüa > uma (Lii^öes, 60). But this happened only in uma (and its derivatives). It was brought about by the intensification and preservation of the nasal resonance in the earlier form ña through the influence of the masculine form um [Ü], T h e form üa w a s preferred by the g r a m m a r i a n Joäo de M o r a e s M a d u r e y r a Feijó (Moraes, 71-72) in the e a r l y eighteenth century, is still popular in all parts of Portugal, and is used in verse before a noun beginning with m (Opúsculos, II, 139-140). Between tonic u and final a, m did not generally develop nor did the nasal resonance r e m a i n : *commúna > comüa > comua;

§ 78]

INTERVOCALIC SHORT CONSONANTS

73

lunam > lua > lua. T h e f o r m luma (for lua) is f o u n d in c e r t a i n d i a l e c t s ( C o m p , 216, n. 1; R L , V I I I , 167; R L , X X V I I I , 29). 5. If t h e first v o w e l w a s p r e t o n i c a n d t h e s e c o n d was

followed

by

a

dental,

a

consonantal

n

vowel

sometimes

developed between the second vowel and the dental,

the

nasal resonance remaining on the second vowel: *anethulum > endro; diviidade

*cinitia

> cèiza

> divindade

nutias

> miu(as

nìtentiam

;

> ciinza

*manutia

> miun(as;

> pendenza

> cinza;

divinitatem

> mau^a

tenetis

(old) ;

> tèedes

venitis

>

> maun$a;

mi-

> tendes;

> vèides

poe-

> viides

>

vinde s. A. T h i s c h a n g e a l w a y s t o o k place if b o t h v o w e l s w e r e alike or b e c a m e alike b y a s s i m i l a t i o n . I n o t h e r cases t h e p r e s e n c e of a d e n t a l a f t e r t h e second v o w e l did n o t a l w a y s a s s u r e t h e p r o d u c tion of a c o n s o n a n t a l n: *adminaclam > amea^a; monetarti > moeda; vanìtàtem > vaìdade; vènàtum > veado. B. A f t e r t h e d e n t a l e l e m e n t of F [ts] a n d z [dz] d i s a p p e a r e d , t h e [n] also d i s a p p e a r e d b u t t h e nasal r e s o n a n c e r e m a i n e d : cinza [sindze] > cinza [size]. 6. If b o t h v o w e l s w e r e a l i k e a n d t h e s e c o n d w a s f o l l o w e d b y a v e l a r , a v e l a r n [13] d e v e l o p e d : enecare dicàmus

> bèeigamos

> bengamos

> engar;

bene-

(old).

7. If a final c o n t r a c t e d n a s a l v o w e l , a n a s a l

diphthong,

o r a c o n s o n a n t a l n a s a l of s o m e k i n d d i d n o t d e v e l o p , n a s a l r e s o n a n c e d i s a p p e a r e d i n t h e c o u r s e of t h e century:

arenarti

> area

boa; corònam

> corda

tenére

> teer

> tèer

> area

> corda;

> areia;

mónàchum

ter; anellum

>

bonam > moago

dello

the

fifteenth > boa >

> eelo

>

mogo; >

elo.

A. T h e n a s a l r e s o n a n c e has r e m a i n e d in s o m e of t h e s e w o r d s in c e r t a i n d i a l e c t s , e.g., bòa ( R L , X I , 274; R L , X I I , 308). b. I n t e r v o c a l i c n r e m a i n e d u n c h a n g e d in l e a r n e d w o r d s , e.g., Jortunam > fortuna; unire > unir; a n d in b o r r o w i n g s , e.g., poenam > Sp. pena > P t g . pena; septimànam > Sp. semana > P t g . semana. T h e w o r d s menos ( f r o m minus), menor ( f r o m mindrem) a n d Jeno ( f r o m fenum) are regressive w o r d s a n d r e p l a c e O P t g . meos, meor a n d feo r e s p e c t i v e l y ; t h e y a r e p r o b a b l y b o r r o w ings f r o m S p a n i s h , a d o p t e d in o r d e r t o a v o i d c o n f u s i o n w i t h o t h e r

74

[ § 79-82

PHONOLOGY

words t h a t h a d identical forms, e.g., meos ( f r o m meos and feo (from foedum). Cf. B S C , X , 814. INTERVOCALIC 79.

LONG

mcdioj),

CONSONANTS

LABIALS.

1. C I . L . i n t e r v o c a l i c bb > P t g . b: abbatem > abade. 2 . C I . L . i n t e r v o c a l i c / / > P t g . / : offendo > ojendo. 3. C I . L . i n t e r v o c a l i c pp > P t g . p: cippum > cepo; stuppam > estopa. 80.

PALATALS AND

VELARS.

1. CI. L. i n t e r v o c a l i c cc f o l l o w e d b y a, o, o r u > P t g . c [ k j : buccam > boca; peccare > pecar; siccum > seco. 2. C I . L. i n t e r v o c a l i c cc f o l l o w e d b y e o r i > P t g . c [s]: acceptare > aceitar; accentum > acento. A. In the d e v e l o p m e n t of occidentem to ouciente (BF, I, 47; E l u c ) , t h e first c [k] p r o b a b l y changed t o p and vocalized (cf. § 9 2 , 7 c). Cf. P h M . I, 241. 3. C I . L . i n t e r v o c a l i c gg f o l l o w e d b y e o r i > P t g . g [5]: exaggerate > exagerar; suggerere > sugerir ( w i t h s h i f t of conjugation). 81.

DENTALS.

1. C I . L. i n t e r v o c a l i c dd > P t g . d: addere > adir ( w i t h s h i f t of c o n j u g a t i o n ) : adduco > adugo ( o l d ) . 2. C I . L. i n t e r v o c a l i c tt > P t g . t: mlttit > mete; sagittam >

seta.

82.

LIQUIDS.

1. C I . L . gallinam >

intervocalic galinha.

11 > P t g .

I:

caballum

>

cavalo;

A. W o r d s in which 11 b e c a m e Ih are borrowings f r o m S p a n i s h : beryllum > brilho; grillum > grilho (old). T h e word caballarlum b e c a m e cavalexro and cavalkeiro; and castelldnum b e c a m e casteldo a n d castelhano ( O P t g . castelhao) \ the f o r m s with Ih are borrowings f r o m Spanish. >

2. C I . L. i n t e r v o c a l i c rr > P t g . rr: ferrum torre.

> ferro;

turrim

§ 83-85] 83.

M E D I A L CONSONANT GROUPS

75

SIBILANTS.

ss > Ptg. ss [sj: ossum > osso;

1. CI. L. intervocalic

passum

> passo.

A. In a few words ss became x [ J ] : vessicam > bexiga (cf. I t . vescica, G i t , § 109, 6 ) ; Messlas > Mexias (old). T h i s p a l a t a l i z a tion of ss has been a t t r i b u t e d to the following ! (Comp, 1 1 8 - 1 1 9 ; R L , X X X I V , 307), but is probably the result of confusion of dialectal [S] with [$] (cf. § 6 5 , 1 A). 84.

NASALS.

1. CI.

comum;

L.

mm > Ptg. m:

intervocalic

flammam

commúnem

>

> chama.

A. T h e nasalization of the preceding vowel has d i s a p p e a r e d except in certain dialects. 2. CI. L. intervocalic

num > pano.

nn > Ptg. n: annum

> ano;

pan-

A. Words in which CI. L. nn became P t g . nh are borrowings from S p a n i s h : * ante-annum > Sp. antaño > antanho; plnnam > Sp. peña > penha; stannum > Sp. estaño > estanho. B. T h e preceding vowel was nasalized but this nasalization has disappeared except in certain dialects. MEDIAL 85.

CHANGES

CONSONANT

TO G R O U P S

IN VULGAR

GROUPS LATIN.

1. M o s t of the changes undergone b y medial c o n s o n a n t groups h a d their inception in V u l g a r Latin and it is genera l l y difficult to determine how f a r t h e y had progressed b y the beginning of the Portuguese period. T h e r e are a f e w , h o w e v e r , which took place clearly in V u l g a r Latin (see §12, 6)"

rs > V . L . and Ptg. ss: persicum > > pessoa; ursum > usso (old); versum

2. CI. L. medial

pessego;

personam

> ves so (old). A. M P t g . urso and verso are regressive words.

76

PHONOLOGY 3. CI.

gypsum

L. >

medial

ps > V . L .

and

Ptg.

ss:

ipse

>

esse;

gesso.

A. It h a s b e e n m a i n t a i n e d t h a t ps b e c a m e [ks] ( M - L , G r a m , I, § 4 5 8 ) ; t h i s a s s u m p t i o n w o u l d v e r y c o n v e n i e n t l y e x p l a i n caixa ( f r o m capsam) a n d queixo ( f r o m capsum). I t s h o u l d be p o i n t e d o u t , h o w e v e r , t h a t if t h i s is c o r r e c t , caixa w o u l d b e t h e o n l y w o r d in w h i c h a f o l l o w e d b y x did n o t b e c o m e ei (cf. § 33, 3). C f . H R , V, 349. 4 . CI. L . m e d i a l pt > V . L . tt a n d P t g . t: aptare captare turn

> catar;

> escrito;

inceptare septem

> encetar;

> sete.

ruptum

Cf. C a r n o y ,

>

> roto;

atar; scrip-

158.

A. T h e r e are m a n y w o r d s in w h i c h t h e p w a s p r e s e r v e d a n d h a d a d i f f e r e n t s u b s e q u e n t d e v e l o p m e n t (§ 92, 7 c ) . 5. CI. L. m e d i a l net > V . L . a n d P t g . nt: junctum sanctum

>

>

junto;

santo.

A. I t is difficult t o d e t e r m i n e t h e t i m e of t h i s c h a n g e ( C a r n o y , 165-166). I t did n o t t a k e p l a c e in t h e V u l g a r L a t i n of G a u l a n d c e r t a i n o t h e r regions ( G V L , § 267). 6 . CI. L . m e d i a l ns > V . L . a n d P t g . s: mensem pensare

> pesar;

sponsvm

>

>

mes;

esposo.

A. A s i m i l a r c h a n g e t o o k p l a c e in t h e g r o u p nstr, w h i c h b e c a m e sir: instrumentum > estormento ( o l d ) ; monstrare > mostrar. 7.

CI. L . m e d i a l nf > V . L .

cofortar

(old);

infantem

>

and

if ante

OPtg. f :

confortare

( o l d ) ; infernum

>

> iferno

(old). A. T h i s n w a s r e s t o r e d b y regression a t a n e a r l y d a t e , p e r h a p s t h r o u g h t h e i n f l u e n c e of t h e c o m m o n p r e f i x e s in- a n d con- (cf. M - L , G r a m , I, § 403, 3).

8. CI. L. * followed by a consonant > V.L. and Ptg. s: expertum

> esperto;

sextum

> sesto

(now spelled

sexto).

9. The final consonant of a prefix was assimilated to the initial consonant of the word to which it was attached: adversum

> avesso;

subterrare

( D u C a n g e ) > soterrar;

adven ire

avir;

obviare

subtilem

oviar >

sutil.

(old);

§ 85-86]

MEDIAL

CONSONANT

GROUPS

77

A. T h e final c o n s o n a n t of t h e prefix w a s o f t e n r e s t o r e d byr e g r e s s i o n : adventum > avento (old) a n d advento; advocatum > avogado (old) a n d advogado.

86.

MUTE

+

LIQUID.

1. A m u t e f o l l o w e d b y / o r r ( e x c e p t i n t h e g r o u p s cl, gr, a n d bl)

h a d t h e s a m e d e v e l o p m e n t in P o r t u g u e s e as a

s i n g l e i n t e r v o c a l i c m u t e ; if t h e l i q u i d w a s I, i t b e c a m e colubram jebrem

> coovra > jevre

> dobrar;

gl,

(RL,

XXVII,

24);

( o l d a n d p o p u l a r ) ; aprilem

leporem

sogro;

*anethiilum

pedra;

quadraginta

> lebre;

lacrimam

> *anedlo > quarenta;

> aedro hederam

laborare

>

> abril;

> lagrima; > endro; > edra

r:

lavrar; duplare

socrum

>

petram

>

>

hera.

A. A l o n g c o n s o n a n t f o l l o w e d b y / o r r h a d t h e s a m e d e v e l o p m e n t as a n i n t e r v o c a l i c l o n g c o n s o n a n t : approbare > aprovar; liticram > letra; supplicare > supncar (old). B. T h e r e a r e s o m e s e m i - l e a r n e d w o r d s in w h i c h b d i d n o t c h a n g e , e.g., nobUem > nobre; octubrem > outubro; a n d s o m e r e g r e s s i v e w o r d s in w h i c h b w a s r e s t o r e d , e.g., cobra (for coovra); fibre ( f o r jevre). c . T h e s e g r o u p s w e r e s o m e t i m e s b r o k e n u p b y m e t a t h e s i s of t h e r: pigritiam > *pegri(a > pregui(a; tenebras > teevras > irevas. D. W o r d s in w h i c h gr s u r v i v e d u n c h a n g e d a n d w o r d s in w h i c h cl a n d gl b e c a m e gr a r e s e m i - l e a r n e d w o r d s o r b o r r o w i n g s f r o m o t h e r l a n g u a g e s : nigrum > negro; eclesiam > igreja; regulam > S p . regla > regra; miraculum > milagre; saeculum > segre ( o l d ) . See Z R P h , X L I I , 2 2 7 - 2 3 0 . C f . Biblos, V I I , 518. F o r t h e p o p u l a r d e v e l o p m e n t of cl a n d gl, see § 92, 8; f o r t h a t of gr, see §92,4. E. T h e w o r d s mai ( f r o m matrem) a n d pai ( f r o m patrem) present a special problem. I n t e r m e d i a t e f o r m s made a n d pade h a v e b e e n e x p l a i n e d as h a v i n g a r i s e n in t h e p r o n u n c i a t i o n of c h i l d r e n b e c a u s e of t h e i r t r o u b l e in p r o n o u n c i n g r ( R L , X X I I I , 50 a n d 6 4 ) . T h e s e i n t e r m e d i a t e f o r m s h a v e also b e e n e x p l a i n e d as h a v i n g d e v e l o p e d b y a n a l o g y w i t h frade ( f r o m jratrem), in w h i c h t h e r fell b y d i s s i m i l a t i o n ( N a s c e n t e s , s.v. mae). The

PHONOLOGY

78

[§ 86-87

form pai is found in rime with vai (CA, line 8380) in a poem written in the first half of the thirteenth century (CA, II, 297); máe is not found in the early Cancioneiros (RL, X X I I I , 50). F. For the development of cadeira (from cathedram), see RFE, VII, 58. 2. In the regular d e v e l o p m e n t assimilated t o t h e l: fabüláre *tabularía > *tablaria > taleira; (Esquisse, § 72 a ; Opúsculos, I I , *sublo > solo (CV, N o . 507).

of t h e g r o u p bl t h e b was > fablar > fallar > falar: -ubi ilium > *ublo > ullo 19), i.e., ulo; sub ilium >

A. There are no words in which bl became vr. Words in which it became br are semi-learned words or borrowings from Spanish: obligare > obrigar; oblitdre (REW) > obridar (old); diabolum > Sp. diablo > diabro (FM, II, Glossary, s.v. diaboo). B. The development of bl in falar has been explained as due to the influence of calar (Grund, I, 975) and also as due to dissimilation of the v in an intermediate form *favlar (MP, XI, 349). But neither of these explanations would account for the same development in taleira, ullo, and solo. Cf. Nascentes, s.v. falar. 87.

GROUPS

ENDING

IN

YOD.

1. T h e s e g r o u p s consist of one or two c o n s o n a n t s followed b y u n a c c e n t e d e or i in h i a t u s with a following vowel. T h e yod m a y be of V u l g a r L a t i n or P o r t u g u e s e origin. Vulgar L a t i n yod arose t h r o u g h h i a t u s originally existing in Classical L a t i n ; P o r t u g u e s e yod t h r o u g h h i a t u s b r o u g h t a b o u t b y t h e fall in P o r t u g u e s e of intervocalic g, /, or n, or t h r o u g h original Classical L a t i n h i a t u s in l a t e r borrowings f r o m L a t i n , i.e., in semi-learned words. G r o u p s in which yod developed f r o m a p a l a t a l c o n s o n a n t are t r e a t e d separately(§ 92). 2. T h a t yod was relatively slow in f o r m i n g in the P o r tuguese t e r r i t o r y is shown b y t h e voicing of p and b in forms like saiba and raiva (§ 88 A), t h e lack of a t t r a c t i o n in early forms like sabia ( R L , X X I I I , 79), a n d t h e long retention of syllabic v a l u e b y t h e e in h i a t u s in forms like jemea (§ 91, 1 B).

§ 88-89] 88.

GROUPS

79

M E D I A L CONSONANT G R O U P S UNAFFECTED

BY

VULGAR

LATIN

YOD.

V u l g a r L a t i n y o d d i d n o t a f f e c t a p r e c e d i n g m e d i a l b, p, f , v, m or r, or a p r e c e d i n g g r o u p e n d i n g in b, v, m or r: *rabtam (for rabiem) > raiva; rubeum > ruivo; sepiam > siba; saplat > saiba; cuphiam ( D u C a n g e ) > c o i f a ; caveam > gaiva; vindemlam > vindima; corium > coiro; superblam > soberba; nervium ( A i n s w o r t h ) > nervo; dormiamus > durmamos; atrium > adro. A. It is evident t h a t yod did not prevent the voicing of a preceding medial b or p. It has been suggested that yod lengthened p to pp after a Classical Latin short vowel but not after a long vowel ( M - L , Gram, I, § 506): apium > *appiu > aipo, but sepiam > siba; and accordingly, the b of caibo (from capio), etc. and the b of saiba (from saplat), etc. have been explained as analogical with other forms of these verbs in which there was no yod (Grund, I, 959). But the form aipo is probably semi-learned; cf. Sp. apio. B. Preceding groups ending in p, /, /, or n (except mn), if there were any, were not affected by yod. For mn plus yod see § 89, 9 B. 89.

GROUPS

AFFECTED

BY

VULGAR

LATIN

YOD.

1. V u l g a r L a t i n y o d a f f c c t c d a p r e c e d i n g m e d i a l c, t, d, g, I, 11, n, s, a n d ss a n d a p r e c e d i n g g r o u p e n d i n g in c, t, d, a n d g. 2. L a t . c + i p r e c e d e d b y a v o w e l or a c o n s o n a n t > P t g . f or c [s]: bracchium > braqo; facio > fa^o; *incalciare > encah;aT; lanceare > lanqar; faciem > face. A. The intermediate stage was [ts], from which the dental element disappeared in the course of the sixteenth century (cf. K J , IV, P a r t I, 327). B. L a t . c plus yod, preceded by a vowel, sometimes became z: fid-uciam > fiuza (old); Gallaeclam > Galiza; judicium > juizo. This is the regular development of t plus yod. T h e confusion in these words arose as a result of the common use in Vulgar L a t i n of c plus yod for t plus yod (see section 4 A below).

80

PHONOLOGY

c. L a t . et plus yod became iç: destructiônem > destruiçâo; elcctiônem > eleiçâo. As a result of t h e use of ct for pt (§ 92, 7 c), pi plus yod likewise became iç: conceptiônem > conceiçào. 3. L a t . t + I p r e c e d e d b y a c o n s o n a n t > ç or c [s]: fortiam ( D u C a n g e ) > força; lenteum > lenço; tertiarlum > terceiro; silentium > seenço ( o l d ) . 4. L a t . t + i p r e c e d e d b y a v o w e l > z: pretiàre > prezar; ratiônem > razâo; satiànem > sazâo; vitium > vêzo; -ïtiam > -eza. A. L a t . t plus yod preceded by a vowel did not develop regularly in most cases b u t became ç: *metio (for mêtïor) > meço; oratiônem > oraçâo; palatïum > paço; plateam > praça; plùtèa > ckoça; pûtëum > poço; -ïtïam > -iça; -ïtïem > -IVÎ. This development arose in Vulgar Latin t h r o u g h a semi-learned mistaken use of c plus yod for t plus yod in an effort to avoid the popular and long established assibilated pronunciation of t plus yod; t h a t is, [kj] came to be used as the result of an effort to say [tj] instead of [tsj] ( T A P A , X L V I I , 147). Cf. G V L , § 276; Hanssen § 19, 14. It has been pointed out t h a t t plus yod had become assibilated somewhat earlier t h a n c plus yod (§ 12, 3). B. Words in which st plus yod became ch are borrowings f r o m Galician: bestlam > bicha; comestidnem > comichâo. 5. L a t . d + i p r e c e d e d b y a v o w e l > P t g . j [3]: adiutàre > ajudar; hodie > hoje; podium > pôjo; video > vejo. A. In a few words d plus yod became i [j]: medium > meio; modïum > moio; podium > poio; radium > raio. Some of these words are semi-learned words, some are borrowings f r o m Spanish (Behr, 37-38). Cf. T A P A , X L V I I , 150-151. Fastio (from fastidium) is a borrowing f r o m Spanish (cf. M a n u a l , § 53, 3). 6. L a t . d + i p r e c e d e d b y a c o n s o n a n t or a d i p h t h o n g > P t g . ç: ardeo > arço (old) ; *perdeo ( f o r perdo) > perço ( o l d ) ; verecundiam > vergonça ( o l d ) ; audio > ouço. A. It seems t h a t in certain regions d between n and yod fell; hence t h e v a r i a n t vergonha of O P t g . vergonça and O P t g . rigonha (from iracundiam). Cf. O F r . graignor (from grandidrem). See C a r n o y , 156. B. T h e word gozo (from gaudium) is a borrowing f r o m Spanish.

§ 89]

MEDIAL CONSONANT

81

GROUPS

7. Lat. g 4- ! preceded by a vowel or a consonant > Ptg- j esponja.

[3] : fugio

> fujo;

pulegium

> poejo;

spongiam

>

A. T h e words correla (from corrigiam) and ensaio (from exagium) are probably borrowings f r o m Spanish. Navio (from navigium) m a y be a borrowing from a dialect of Italy ( M P , X I I , 194). For the different d e v e l o p m e n t of t h e tonic vowel of correla and navio, see R H i , L X X V I I , 132.

8. Lat. / or // + i preceded by a vowel > Ptg. alienum mulher;

> alheio; filium > filho; mulierem > *mulierem allium > alho; malleàre > malhar.

Ih: >

A. This change has t a k e n place in learned words in certain dialects, e.g., familiam > familia > familha (Alentejano). B. In some words, yod did not p r e v e n t the fall of I: salio > saio;doleo > doyo (old); Juliànum > Juiào (old). T h i s has been explained as a d e v e l o p m e n t peculiar to Old Galician ( H u b e r , § 44, 1). It has been explained in the verb forms as due to analogy with the other persons in which there was no yod ( R L , X X X I I I , 198). 9. Lat. n + 1 p r e c e d e d b y a v o w e l > P t g . nh: > senhor; teneo > tenho; vineam > vinha.

seniorem

A. This change has t a k e n place in learned words in popular speech: Antdnium > Antonio > Antonho (popular); daemònium > demonio > demonho (popular). B. T h e r e are two examples with mn plus y o d : sòmnium > sonko; calumniam > caomia > coima. T h i s double development is found only in P o r t u g u e s e (cf. Sp. sueño and caloña, Fr. songe and OFr. chalonge, I t . sogno and O l t . calogna). T h e form calonha (Eluc) is p r o b a b l y a borrowing f r o m Spanish. T h e m of coima is probably due to c o n t a m i n a t i o n b y Arabic quima (Dozy, 257). For this and o t h e r a t t e m p t s to explain t h e development of coima, see N a s c e n t e s ; M P , X I , 349. T h e derivation of conha f r o m calumniam ( R L , I I I , 265) was a m i s t a k e which N u n e s did not repeat in his " C o m p e n d i o . " c. T h e g r o u p In followed by e or 1 in hiatus must have lost the I as soon as yod developed ( A L L G , I, 248): balneum > *banium > banho.

82

PHONOLOGY

[§ 89-91

10. L a t . s + i p r e c e d e d b y a vowel > P t g . / [3]: basium > beijo; caseum > queijo; eclesiam > igreja. 11. L a t . ss + 1 a n d sc + j > P t g . x [$]: *bassium > baixo; passionem > paixao; russeum > roxo; asciolam (Du C a n g e ) > enxó; fasciam > faixa a n d faxa. 90.

E F F E C T OF V U L G A R

LATIN YOD

ON V O W E L AND

CON-

SONANT.

I t m i g h t be said t h a t in g e n e r a l V u l g a r L a t i n yod spent its force e i t h e r on t h e p r e c e d i n g vowel or t h e preceding c o n s o n a n t . If it a f f e c t e d t h e p r e c e d i n g c o n s o n a n t , it h a d no effect on t h e p r e c e d i n g vowel. T h e o n l y exceptions are t h e closing of tonic f , p a n d 0 w i t h c o n s o n a n t s affected b y y o d : tertium > ter(o; hodie > hoje; cüneum > cunho; a n d t h e a t t r a c t i o n of yod w i t h s, ss a n d sc: caseum > queijo; passionem > paixao; fasciam > faixa. T h e a p p a r e n t effect of yod on p r e t o n i c e w a s r a t h e r t h e s u b s e q u e n t effect of t h e p a l a t a l i z e d c o n s o n a n t (see § 41, 3 A). A. Yod affected both vowel and consonant at the same time more frequently in dialectal Portuguese: veinho for venho; fuijo for fujo (Opúsculos, II, 204). B. The yod which came from the palatal consonant of a consonant group likewise had in general no effect on the preceding vowel if it affected another consonant of the consonant group, e.g., agnum > anho. 91.

GROUPS

ENDING

IN

PORTUGUESE

YOD.

1. T h e P o r t u g u e s e yod which arose t h r o u g h t h e fall of intervocalic d, g, I or n: a) did n o t a f f e c t t h e preceding c o n s o n a n t : acinum > azeo (old); búcinum > búzio; *cütinam > codea; litigare > lidiar > lidar; *reíinam > rédea; b) if it developed e a r l y , it was a t t r a c t e d to t h e preceding v o w e l : comedo > *comio > coimo (old); *exfaminare > esfaimar; or it closed t h e p r e c e d i n g vowel a n d fell: limpidum > limpo; türbidum > turvo; terminum > termio > termo; c) if it developed late, it did n o t a f f e c t t h e p r e c e d i n g vowel

§ 91-92]

MEDIAL CONSONANT GROUPS

and r e m a i n e d : minum > gemeo;

*cuiinam *retinam

> codea; > redea;

feminam similam

> femea; > semea.

83

ge-

A. In the hypothetical development of the word sujo a Portuguese yod affects the preceding consonant: sucidum > *fusio (by metathesis) > (ujo, now spelled sujo (RHi, V, 421). B. The e of codea, femea, etc. counted as a separate syllable as late as the verse of the "Cancioneiro Geral" (A.D. 1516) but became a yod soon after this time (Grund, I, 1001, § 265). 2. T h e y o d of l a t e r borrowings f r o m L a t i n (semi-learned words) did not affect the preceding consonant b u t affected the preceding vowel a n d f e l l : bestiam > besta; christianum > cristao; confessionem > confissao; reliquias > religas ( B a u s t e i n e , 5 8 2 ) ; studium > estudo; visidnem > visao; insidias > inssidas ( R L , X X , 197). A. In learned words the vowel did not close, e.g., sessao sessionem) and the yod did not generally fall, e.g., familia familiam), gremio (from gremlum). Cf. § 35, 2 B. B. In dialectal and popular Portuguese this yod was times attracted to the preceding vowel regardless of the of the preceding consonant: daemonlum > demonio > (popular); Timotheum > Timoteo > Temoito (Algarvio). 92.

GROUPS

CONTAINING

A

PALATAL

(from (from somenature demoino

CONSONANT.

1. L a t . c preceded b y a consonant a n d followed b y e or i > P t g . c [s]: mancipium > mancebo; mercedem > merce; vincendum > vencendo. A. In urze (from ulicem) and catorze (from quattuordecim) the c developed in the intervocalic position before the fall of the posttonic penult. 2. L a t . sc followed b y e or i > P t g . x [$]: fascem > feixe; miscere > mexer; piscem > peixe; vascellum > baixel. . A. The position of the accent does not seem to have anything to do with the development (cf. M-L, Gram, I, §473). In semi-learned words sc followed by e or i became c: discipulum > decipolo (cf. FM, I, xxiv); *-escere > -ecer.

84

PHONOLOGY

3. Lat. g preceded b y a c o n s o n a n t and followed by e or i > Ptg. g [3]: algentem > algente; angelum > angeo > anjo; argillam > argila; longe > longe; spargere > espargir (with shift of c o n j u g a t i o n ) ; virginem > virgem. A. Old and popular forms such as arzila, esparzir, etc. are probably the result of a semi-learned confusion of c and g in this position in Vulgar Latin (Carnoy, 157). B. T h e group ng followed by e or X became nh in a few words borrowed from Spanish: quingentos > OSp. quiñentos > quinhentos; "ringire (for ringi) > Sp. reñir > renhir. T h e word inhenho (from ingenüum) presents a special problem: ingenüum first became engeo (Eluc) or *engeo, which later became engenho (Eluc), perhaps by contamination of engenho (from ingenium); by assimilation of ng to nh, engenho then became inhenho (cf. Comp, 134). 4. Lat. gr > P t g . ir: fragrare cheirar; integrum > inteiro.

> flagrare ( G V L , § 292) >

5. Lat. gn > P t g . nh: agnum > anho; cognatum > cunhado; cognoscendum > conhecendo; insignem > insinhe (old); lignum > lenho; pügnum > punho. A. T h e changes through which this sound passed were the following: [gn] > [jn] > [jp] > [ji]. Reino (from regnum) was prevented from going beyond the second stage through the influence of rei. B. The g of gn fell in semi-learned words: dignum > dino (old); insignare (Du Cange) > ensinar; benignum > benino (old); malignum > malino (old); signum > sino. This g was restored in spelling and pronunciation in digno, benigno and maligno. 6. Lat. ngl > P t g . nh: singülos > senhos (Flor, G l o s s a r y ) ; singularium > senlheiro > senheiro (old); ungülam > unha. A. T h e intermediate stage was nlh, shown in the form senlheiro (Cd'A, I I I , Glossary). 7. Lat. ct > Ptg. {t: factum tem > noite.

> jeito;

ledum

> leito;

noc-

§ 92]

MEDIAL CONSONANT GROUPS

85

A. If the } of {t b e c a m e nasalized, a consonantal n developed (§ 78, 5) and the j fell: benedïctum > beêeito > ¿rifo > bento; hac node > *a node > aôite > oonte (RL, X X V I I , 55) > ontem; pedïnem > peite > pèitê > pentem > pente. T h e same change occurred where the i was of a different origin: P t g . muito > munto (popular, R L , IV, 231; H R , VI, 264). B. T h e i of it generally fell when preceded b y u: trudam > truita > truta; frûctum > fruito > fruto. I t c o n t r a c t e d with a preceding i (§ 36, 2) : *dïdum > dito. c. L a t . ct and pt were sometimes interchanged in Portuguese territory. T h e regular d e v e l o p m e n t of pt in Vulgar Latin was to it (§ 85, 4). A different development took place later, viz., pt > *ft > ut: aptum > auto (old); praeceptum > preceuto (old); raptum > rauto (RL, X V I , 10); adoptàre > adoutar (Eluc); baptizâre > boutiçar (Eluc). I t was this pt (or its derivative *ft) which was interchanged with d (or its derivative *[xt]). Cf. T r a i t é , 204; B F , I I I , 96-97. T h i s use of one g r o u p for the o t h e r was d e t e r m i n e d by the n a t u r e of t h e preceding vowel. W h e n preceded by e, pt h a d the same d e v e l o p m e n t as d: acceptum > aceito; conceptum > conceito; praeceptum > preceito; receptam > receita; septimum > seitimo (old); septuaginta > seitenta (old). T h i s development of pt to [xt], followed by vocalization of the palatal was characteristic of Celtic ( D o t t i n , 100). W h e n preceded by a or o, particularly pretonic o, d had t h e same development as pt: adivum > autivo ( E l u c ) ; actum > auto; contradum > contrauto (Elue); tradâtum > trautado (Fab, Glossary) ; docturn > douto; dodôrem > doutor; odâvum > outavo ( F M , I I , Glossary); octûbrem (Carnoy, 64) > outubro. T h e d e v e l o p m e n t of -od- in some regions t o -oit- and in others to -out-, followed by inter-dialectal influence, m a y have been the origin of the confusion of ou and oi, which existed much earlier t h a n has generally been believed, e.g., noute ( A H P , IV, 190, A . D . 1 385; BF, IV, 347, 1450?; Abraham, § 2 8 ) . As a consequence of this early confusion, the use of oi spread in the sixteenth century to words which originally had ou, e.g., coisa ( f o r cousa < causam) and the use of ou spread to words which originally had o 't not coming f r o m oc(t), e.g., couro ( f o r coiro < côrïuni), with the

86

PHONOLOGY

[§92

result that ou and oi have become generally interchangeable, although ou is more literary, oi more colloquial. It is significant that such spellings as noute are f o u n d earlier than such spellings as couro and coisa. Before the beginning of t h e s e v e n t e e n t h c e n t u r y , ou had lost its diphthongal character, which is still preserved, however, in the north of Portugal. For dialectal differences, see Esquisse, § 56 e; Fink, 98. T h e change f r o m ou t o oi has been explained as due to dissimilation of t h e t w o elements of ou ( R L , X X X I I , 288). Cf. Grund, I, 936; Bourciez, § 332 b; P h M , I, 75; AStNS, LXV, 47; BHi, X X X I X , 398. D. T h e R o m a n c e g r o u p c't became [dzd], t h e n [dz] by dissimilation, and finally z [z]: *amïcitàtem (for amïcïtïam) > amizade; placitum > prazo; recltàre > rezar. E. In learned words t h e c of ct generally fell: actum > acto [atu]; victimam > vitima; b u t factum > facto [faktu]. 8. V . L . ct a n d gl > P t g . Ih: apïculam > abelha; novaculam > navalha; oculum > olho; vermxculum > vermelho; rotulam > *roclam > rolha; situla tn *seclcwi sflha,' vitulutn )> veclum ( A p P r ) > velho; rêgùlam. > relha; tëgulam > telha. A. T h e combination tl regularly became cl in Vulgar Latin (GVL, § 284). B. In the group bl, b was assimilated to / (§ 86, 2). T h e same assimilation took place in some words in Spanish (Manual, § 57, 1). B u t in ralhar (from rabuldre) and trilho (from tribiilum), bl became cl, unless these words were borrowed from Spanish (which does not seem possible in t h e case of ralhar as Sp. rallar has a different meaning and comes f r o m *radulàre). c. T h e group pi in *scoplum (from scopulum) seems to have become cl: scopulum > *scoclum > escôlho. Sp. escollo was borrowed f r o m P o r t u g u e s e and perhaps f r o m French either directly or through I t . scoglio ( A G I , X I I I , 374—375). G r ô b e r adopts the same Vulgar L a t i n f o r m *scoclum for French, Provençal, and Italian b u t needlessly assumes the assimilation of p to / for Sp. escollo ( A L L G , V, 461-462). Meyer-Liibke borrows the French, Spanish, and P o r t u g u e s e forms f r o m I t . scoglio, which he believes t o be N e a p o l i t a n in origin (M-L, G r a m , I, § 491).

§ 92-93]

MEDIAL-CONSONANT GROUPS

87

D. T h e r e are m a n y words in P o r t u g u e s e in which the group cl did not develop because of lack of syncope: diabolum > diaboo > diabo; oracülum > oragoo > orago; pericülum > perigoo > perigo. Cf. M - L , I n t r o , § 133. E. T h e g r o u p cl preceded by a consonant b e c a m e ch: astülam > acha; conchülam > concha; mascülum > macho; sarcülum > sacho. T h e preceding consonant was lost in some of these words (RL, X I I I , 273). 9. L a t . m e d i a l x > P t g . x [$] o r ss: buxum > buxo; coxam > coxa; fraxinum > freixo; dixit > disse; *traxuit >

trouxe [trosa], A. T h e change of * t o [5] m a y h a v e been caused in some words by final i, t h u s dixl > dixe b u t dixit > disse, e.g., " E aa cima dixe-lhes eu . . . e o a b a d e me disse . . . " (Flor, 71); see also Carter, 30; R L , X X I I I , 30. See R L , X X X I V , 307. B. In learned words x is pronounced [s]: auxilio, máximo, próximo, sintaxe. In the prefix ex- plus a vowel, x is pronounced [z]: ex ame, exercicio. c. Followed by a voiceless stop, * is pronounced [$]: extraneum > extranho; sextum > sexto. T h e [k] of x in this position disappeared in Vulgar L a t i n (§ 12, 6).

10. Arabic kha and ha preceded by the I of the Arabic a r t i c l e > f : al-khass > alface; al-khaiyat > alfaiate; alkhomra > alfombra; al-khilél > alfinete; al-heire > aljeire; al-heláwa > alféloa; al-hofre > alfobre. Intervocalic kha

and ha sometimes have the same change: al-mokhadda almofada;

at-tahona

> atafona.

>

Cf. D e u x , 9 - 1 0 .

A. This change took place because t h e sound of / was the closest acoustical e q u i v a l e n t of these sounds in P o r t u g u e s e . I n learned words kha b e c a m e [k] while fya became h ( m u t e ) or was entirely lost. 93.

GROUPS

CONTAINING

A

U.

1. In many combinations u disappeared in Vulgar Latin: aestuarium > esteiro; battualia > batalha; februarium > fevereiro; mortuum > morto; quattuor > quattor (Seelmann,

88

PHONOLOGY

218) > quatro;

septuaginta

> setaenta

> setenta.

See § 10,

3. A. A u of Portuguese origin was slow in forming (cf. G r u n d , I, 1001, § 265); it has fallen in a few dialectal forms: consonantem > consoante > consante (Esquisse, § 4 8 d ) ; Ptg. Manuel > Nel. B. When preceded by medial b, p, c, t, s, or x, u was a t t r a c t e d to a preceding a or o: habuit > houve; sapuit > soube; placuit > prougue; potuit > pode (cf. § 33, 4; § 37, 4; § 40, 4; § 43, 5). T h e form prougue shows that attraction took place after the time of the palatalization of g followed by e or /. And that u was relatively slow in forming in the Portuguese territory is shown by the voicing of the medial consonant. Compare Ptg. soube and Sp. supe. W h e n preceded by medial g, u was attracted to a preceding e in dialectal Portuguese: equam > egua > euga (dialectal); regulam > regua > reuga (dialectal). c. T h e u of the diphthong au was a t t r a c t e d to the final syllable in certain dialectal forms of learned words: causarn > causa > cdsua; flautam ( R E W ) > flauta > fldtua (RL, V I I , 39).

2. Lat. qu preceded by a vowel and followed by a > P t g . gu [gw]: aquam

> agua;

aequdlem

> igual;

equam

>

egua.

A. Preceded by a consonant the q did not voice: quinquaginta > cinquaenta > cinquenta (§ 99, 3), the u remaining because t h e following a survived sufficiently late. It is not clear w h y the u fell in nunca (from numquam).

3. Lat. qu preceded by a vowel and followed by e, i, o or u > P t g . g [g] or gu [g]: aliquem > alguem; aguila; *sequo (for sequor) > sigo; antiquum >

aquilam antigo.

>

A. If q was long it did not voice: *acc%-{h)ic > aqui. The words jrequente (from frequentem) and tranqxiilo (from tranquillum) are learned. B. In a few words, u fell early enough in Vulgar Latin for [k] to be assibilated: coquere > *cocere > cozer; laqueum > *laceum > laqo; torquere > *torcere > torcer. GVL, § 254; G r u n d , I, 962; M a n u a l , § 52, 3. c. Lat. cu (unaccented) preceded by a vowel and followed by o had the same development: vacuum > vago.

4. Lat. gu preceded by a consonant and followed by a >

§ 93-94]

MEDIAL CONSONANT GROUPS

89

Ptg. gu [gw]: linguam > lingua. 5. Lat. gu preceded by a consonant and followed by e or o > Ptg. g [g] or gu [g]: distinguo > distingo; sanguen > sangue. A. The word unguento

(from unguentum)

is learned.

6. L a t . lu > Ptg. Iv: valulsset > valvesse (CV, No. 963); *doluerunt (for doluerunt) > dolveron ( C S M , I, No. 241, p. 336, col. 2). 7. L a t . du > Ptg. v: *creduit (M-L, Intro, § 191) > creve (old); *seduit (GVL, § 428) > seve (old). A. The development of this group was probably as follows: d y > dv > vv > v.

8. L a t . nu followed b y a > Ptg. ngu: manuale > mangual; *minuare (for minuere) > minguar. M - L , Gram, I, § 503. 9. L a t . nu followed b y e or by a which became e > Ptg. «: januarium > Janeiro; januellam > janela; manuaria > maneira. M - L , Gram, I, § 503. 10. A R o m a n c e u developed from bilabial v followed by g fabricam > fravega > *fravga > *frauga > fragua; -ificare > -evegar > *-evgar > *-eugar > -iguar (Manual, § 18, 2). A. These forms are borrowings from Spanish, as the posttonic penult of fravega and the intertonic of -evegar would not have fallen in Portuguese. The ending -evegar (or -ivigar) was fairly common in Old Portuguese and although it has been replaced by -iguar, the latter is not used as much in Portuguese as in Spanish. Cf. AStNS, CXXIV, 336, s.v. *adv'erifico. 94.

GROUPS

BEGINNING

WITH I.

1. Lat. /preceded b y a and followed by i, p or c [s] > OPtg. u in the diphthong ou, now pronounced [o]: altarium > outeiro; alterum > outro; saltum > souto; palpare > poupar; *talparla > toupeira; calcern > couce; falcem > fouce. A. That vocalization of I was an early change is proved by

90

PHONOLOGY

[ § 94-95

a) the fact t h a t it occurred in all Romance territory, except Rumania and Italy (literary Italian), b) the fact t h a t it did not occur in Arabic borrowings (notably initial al plus a consonant), c) the fact t h a t it did not occur in Germanic borrowings (names of people and place names), and d) the fact t h a t it did not occur in Romance groups. See Biblos, V I I I , 96-97; GVL, § 288; Grund, I, 976. B. T h e r e are many forms in which this change did not take place: altum > alto; palpare > palpar; calceare > calqar. Memendez Pidal thinks t h a t these forms were due to the strong learned reaction of the twelfth century, which allowed only a few popular forms with o (Ptg. ou) to survive (M-P, Orig, 122). But Piel thinks that both types, e.g., both poupar and palpar, are popular and show the instability of I in this position (Biblos, V I I I , 100-101). See also Z R P h , LVII, 630-631. 2. L a t . It p r e c e d e d by u > O P t g . it > M P t g . t: auscultare > ascuitar > escuitar ( w i t h c h a n g e of prefix) > escutar; cultellum > cuitello > cutelo; vulturem > abuitre > abutre. See B H i , X X X I X , 3 9 7 - 3 9 9 . A. T h e T of muito was perhaps prevented from falling through the influence of the apocopated form mui. There is a popular form munto, which developed according to § 92, 7 A.

95.

G R O U P S B E G I N N I N G WITH A N A S A L

CONSONANT.

1. If m or n was t h e first c o n s o n a n t of a g r o u p , it n a s a l i z e d t h e p r e c e d i n g vowel. T h i s n a s a l r e s o n a n c e h a s s u r v i v e d in M o d e r n P o r t u g u e s e b u t n h a s lost its c o n s o n a n t a l v a l u e b e f o r e /, r, s, f , c [ s ] , / , g [ 3 ] , / or v, e.g., enviar [ivjar]. Bef o r e d or n h a s p r e s e r v e d its c o n s o n a n t a l v a l u e , e.g., antes [5nti$], a n d b e f o r e c [k] or g [g] its v e l a r v a l u e , e.g., encantar [iqkintar]. I n t h e s e g r o u p s , m b e f o r e b or p h a s n o t l o s t its c o n s o n a n t a l v a l u e , e.g., ambos [imbu$]. F o r n p l u s y o d , see § 89, 9. F o r g r o u p s b e g i n n i n g w i t h an n w h i c h a r o s e f r o m a nasalized vowel, see § 78, 5 a n d 6; § 92, 7 A. 2. L a t . mn > P t g . n: autumnum > outono; columnam >

MEDIAL CONSONANT GROUPS

§ 95-96] coona >

( E l u c ) ; damnum

escano;

somnum

>

> dano; dominum

91

> dono;

scamnum

sono.

A. B e c a u s e of lack of s y n c o p e t h e R o m a n c e group mn, w h i c h b e c a m e mbr in Spanish, did not d e v e l o p in P o r t u g u e s e ( § 5 2 , 2; § 5 7 , 2 ) : feminam > Sp. hembra but P t g . jemea. Portuguese words in w h i c h R o m a n c e mn b e c a m e mbr are borrowings f r o m Spanish, e.g., deslumbrar (from Sp. des + lumbrar < luminare).

3. In t h e R o m a n c e group md, which arose through the fall of posttonic or intertonic e a f t e r t h e voicing of intervocalic t, m changed to n by assimilation to the d e n t a l : comitem

>

conde;

limites

>

lindes;

semitarium

>

sendeiro.

T h e s u b s e q u e n t development was the same as t h a t of original Classical Latin nd (section 1 above). 4. A f t e r the fall of p in t h e Classical Latin or Vulgar Latin g r o u p mpt, m b e c a m e n by assimilation to the d e n t a l : exemptum

>

isento;

promptum

> pronto;

computare

>

con-

tar. T h e subsequent development was the same as t h a t of original Classical Latin nt (section 1 above). 5. L a t .

mpl

plere > encher;

>

P t g . nch complere

jugation) ; implicdre

o r mpr:

amplum

> cumprir

(with

> empregar;

simplicem

>

ancho;

shift of > simprez

im-

con(old).

A. T h e c h a n g e to nch was m u c h earlier t h a n the c h a n g e to mpr. T h e g r o u p nfl also b e c a m c nch in inchar (from inflare).

6. T h e R o m a n c e group mbl (§ 113, 1) > P t g . mbr: lum

> combro;

similare

9 6 . MISCELLANEOUS

>

cumu-

sembrar.

GROUPS.

1. L a t . lb and rb > P t g . Iv and rv respectively: albam > alva; arbitrum > alvidro (old); arborem > arvore; carbonem >

carvao;

herbam

> erva;

sibilare

> silvar;

turbidum

>

turvo. A. T h i s w a s perhaps a Vulgar L a t i n c h a n g e (cf. C a r n o y , 141). B. Y o d did not interfere with this d e v e l o p m e n t : superbiam > sobirva (Crest, G l o s s a r y ) . See § 88. c. T h e original L a t i n b was restored in regressive words: barbam > barva (old) and barba; superbum > sobervo (old) and soberbo.

PHONOLOGY

92

2. A n a s a l c o n s o n a n t e n d i n g a g r o u p s o m e t i m e s n a s a l i z e d t h e f o l l o w i n g v o w e l as a n initial c o n s o n a n t d i d : vermem > verme > vermem ( o l d ) ; *remussicare (for re-mussitare) > remusgar > resmugar > resmungar (cf. N a s c e n t e s ) . FINAL 97.

CLASSICAL

LATIN

CONSONANTS

FINAL

CONSONANTS.

1. CI. L. final b fell in e a r l y P o r t u g u e s e : sub > so (old). A. T h i s word seems to be t h e only e x a m p l e and t h e b was later restored in it: M P t g . sob [soba]. 2. CI. L . final c fell in l a t e V u l g a r L a t i n o r e a r l y t u g u e s e : die > di ( o l d ) ; nec > nem;

sic > si > sim

Por-

(§ 112,

1).

3. CI. L . final d fell in V u l g a r L a t i n : ad > a; quid > que. 4. CI. L . final m fell i n V u l g a r L a t i n a n d final n in e a r l y P o r t u g u e s e ( G V L , § 3 1 0 ) : amorem > amor; nomen > nome. B e c a u s e of p r o c l i s i s , b o t h m a n d n r e m a i n e d as n in m o n o s y l l a b l e s in V u l g a r L a t i n . In P o r t u g u e s e t h e y nasalized the p r e c e d i n g v o w e l b u t l o s t t h e i r c o n s o n a n t a l v a l u e : cum > com; in > em; non > nom > nao; quem > quem; rem > rem ( o l d ) . A. In ja t h e m of L a t . jam was lost by wrong division of the word jamais. B. P t g . tdo a n d quao do not come f r o m tarn a n d quam b u t are r a t h e r a p o c o p a t e d forms of tanto and quanta ( P h M , I, 251; Comp, 362, n. 2; H a n s s e n , § 12, 5), in which c o n s o n a n t a l n was lost because it b e c a m e final a n d in which t h e d i p h t h o n g ao developed in t h e general fusion of -a, -o, a n d -ao (§ 157). c. F o r t h e sound of final -em, see § 34, 10 a n d § 46, 11. 5. CI. L. final r, b y m e t a t h e s i s w i t h a p r e c e d i n g u n a c c e n t e d v o w e l , b e c a m e t h e l a s t e l e m e n t of a m e d i a l g r o u p : inter > entre; quattuor > quattor > quattro > quatro; semper > sempre; super > sobre. A. T h i s c h a n g e p r o b a b l y took place in Vulgar L a t i n § 245).

(GVL,

FINAL

§ 97-98]

CONSONANTS

93

B. In *super-lo, in which the r of super was not final, this change did not take place: *super-lo > sobello (old and dialectal), e.g., sobellos bees ( R L , V, 136). For sobolo, see R o m , X , 339. T h e word cadaver (from cadaver) is learned. c. T h e preposition pro became *por in Vulgar L a t i n ( G V L , § 14). T h e r of *por and the r of per were really never final in Vulgar Latin or Portuguese because of the regular proclitic position of these words. B o t h per and por existed in early Portuguese with separate meanings ( R F , X X I I I , 351; R F , X X V , 667) but soon came to be interchanged as a result of phonological confusion in which per came to be pronounced por ( R o m , X I , 9 1 - 9 5 ) , per surviving in combinations with the definite article, e.g., pelo, por surviving when not combined. T h e combinations polo, pola, etc. have survived in dialects of the north and in Galician. 6 . CI. homens;

L.

final

vivis

>

s

remained

in

Portuguese:

homines

>

vives.

A. Final s (of a word or a syllable) began to palatalize in the thirteenth century (CA, I, xxiii). B u t see Bourciez, § 341 a. B. T h e [k] of final x became a yod and s remained: sex > seis (§ 34, 4). 7. CI. L . final t f e l l i n V u l g a r L a t i n : bibit >

cantam;

98.

post

>

CONSONANTS

> bebe;

cantant

pois. BECOMING

FINAL

IN

PORTUGUESE.

1. L a t . I, r a n d s, f o l l o w e d b y final e o r i, b e c a m e in P o r t u g u e s e t h r o u g h t h e f a l l of t h i s e o r i: amdrem amor;

male

>

mal;

mensem

>

mes;

sail

( 2 d sg. i m p v . ) >

final > sal

(old). A. Final I is pronounced [1] in Portuguese. 2. L a t . c f o l l o w e d b y

final

e o r i b e c a m e P t g . z, w h i c h

b e c a m e final t h r o u g h t h e f a l l of t h i s e o r i: facit faz;

vicem

> vez; feci

>

> faze

>

fiz.

A. Pronounced [dz] this z did not rime with final s in the early Cancioneiros (CA, I, xxii, n. 6). B. This z is spelled s in Modern Portuguese if the preceding vowel is not accented: pomicem ( G V L , § 207, 3) > pomes; simpllcern > simples.

PHONOLOGY

94

[ § 98-99

3. Lat. t -+- t followed by final e became Ptg. z, which became final through the fall of this e: -itiem > -ez. A. T h e final z in az (old) ( f r o m aciem) a n d faz (old) ( f r o m faciem) is a result of t h e confusion in Vulgar L a t i n of c plus yod and t plus y o d . Cf. § 89, 2 B.

4. L a t . n, followed by final e (chiefly acc. sg. 3d declension and 3d sg. of verbs) or final t (3d pi. of verbs), became final through the fall of this e or t. Later it nasalized the preceding vowel b u t lost its consonantal value ( Z R P h , XLI,

5 5 8 ) : venit

> vem;

bene > bem; finem

münem

> comum.

> fim;

Final

amant

> amam;

amarunt

-irn

and

debent

> amarorn

-um

had

>

devem;

(old);

no

com-

further

change b u t final -em became [ij] (§34, 10; §46, 11) and final -am and -om both became [iw] (§ 157, 2). A. T h e n of tanto and quanto b e c a m e final t h r o u g h a p o c o p e of t h e final syllable (§ 97, 4 B). B. T h e nasal resonance of a final vowel occasionally p r o d u c e d a c o n s o n a n t a l n in a sort of liaison b e t w e e n t h e final vowel a n d t h e initial vowel of a following word, e.g., tarn nasynha (for tarn asynha) (CG, facsimile ed., folha C C X I vo, col. 3). GENERAL 9 9 . ASSIMILATION, VOWELS

IN

PHONOLOGICAL DISSIMILATION,

PHENOMENA AND

CONTRACTION

OF

HIATUS.

1. T h e changes undergone by vowels in hiatus were not sporadic but took place with great regularity. Original Latin hiatus sometimes persisted until the word came into Old Portuguese, e.g., leonem > leom (old) (cf. Sp. león). B u t hiatus was generally eliminated in Vulgar Latin by reduction of the unaccented vowel to a semivowel, by the total disappearance of the unaccented vowel, or by contraction (GVL, §§ 222-227). Therefore, most of the cases of hiatus in Old Portuguese were due to the fall of an intervocalic consonant, e.g., dolére > doer. But this new hiatus of Old Portuguese was in turn eliminated in the same way, namely, by reduction of the unaccented vowel to a semivowel

§ 99]

GENERAL PHONOLOGICAL PHENOMENA

95

(5 and 6 below), by the total disappearance of the unaccented vowel (7 below), by contraction (2 below), or by assimilation and contraction ( 3 below), and in the case where none of these developments could take place (4 below), by the insertion of a [ j ] or a [ w ] , which is not shown in spelling, e.g., lua [Iuwe], The only case where this [ j ] is shown in spelling is between tonic e and a following a or o ( § 3 5 , 7 ) . 2. If the two vowels in hiatus were the same and of the same q u a l i t y , t h e y were contracted: videre > *veder > veer > ver; palatium > paa$o > pa¡o; oracülum > oragoo > orago. a . T h i s contraction, which had begun to t a k e place t o w a r d the end of the t h i r t e e n t h c e n t u r y , was completed b y the end of the fifteenth c e n t u r y (Linóes, 157—159; Cd'A, I, 358). An e x a m i n a tion of the r h y t h m of e a r l y verse shows t h a t it took place earlier b e t w e e n t w o unaccented vowels t h a n it did between t w o vowels of which one was accented ( R P h , I, 3 6 - 3 7 ) . b. T w o pretonic a's in h i a t u s were contracted to [a]: panatarium > padeiro; two pretonic e's in hiatus were contracted to [e]: creditorem > cr(dor; and two pretonic o's in hiatus were con-

tracted to [d]: coldratum

>

cprado.

3. An unaccented vowel in hiatus with an accented vowel was assimilated and contracted, if it was one or two steps more open (in the series a, g, e, i or the series a, g, o, u) than the accented vowel ( R R , X X I I , 4 3 ) : anellum > aello > ((lo > ¿lo; *cadetam > caeda ( R L , X X V I I , 18; F M , II, Glossary) > qu^da; calentem > caente > queente > quente; canalicülam > caelha > quelha; pelagum > peago > pgggo > p(go; portugalense (Nascentes, s.v. portugués) > portugalés > portugaes > portugués > portugués; sagittam > saeta > sggta > sgta; venire > veir > viir > vir; civiles > civiis > rivis; molam > moa > mpg > mó; monachum > moago > mppgo > mogo; solam > soa > SQQ > só; palümbum > paombo > poombo > pombo; nudum > nuo > nuu > nú; *mólútum (for mólítum) > inundo > mudo (BF I 203). a . If t h e unaccented vowel w a s two steps more open t h a n the

PHONOLOGY

96

[§99

a c c e n t e d vowel, b o t h vowels were a f f e c t e d , t h e u n a c c e n t e d vowel closing o n e s t e p , t h e a c c e n t e d vowel o p e n i n g one s t e p , e.g., saeta > s((ta; soa > sgg; majorent > maor (§ 73, S A) > mggr > mor. T h i s c h a n g e did n o t t a k e p l a c e b e t w e e n t o n i c e a n d a following a, w h e r e a different solution was adopted (see § 35, 7 ) . B. T h e e of quente a n d t h e o of pombo were closed b y t h e following n a s a l . T h e o of mogo was closed b y m e t a p h o n y . T h e e of portugués was closed t h r o u g h t h e influence of francés, inglés, etc. T h e e of quelha w a s closed t h r o u g h t h e influence of o t h e r words e n d i n g in -elha ( f r o m -Iculam). All t h e s e vowels m u s t h a v e closed a f t e r c o n t r a c t i o n h a d t a k e n place. c. B e c a u s e of some d i s t u r b i n g o u t s i d e influence, dissimilation i n s t e a d of assimilation t o o k place in t h e d e v e l o p m e n t of t h e word rainha ( f r o m rêgïnam). T h e w o r d s re i a n d reino, a l t h o u g h not a c c e n t e d on t h e t, m a y h a v e had some influence o n t h e Old P o r t u g u e s e f o r m reïa, as suggested b y D ' O v i d i o ( Z R P h , V I I I , 85). A t a n y r a t e dissimilation was f a v o r e d in this word b y t h e effect of r on u n a c c e n t e d e. A s s i m i l a t i o n was p r e v e n t e d in boa ( f r o m bonam), b y t h e nasal resonance, which was l a t e r lost (§45, 2 B). D. I n t h e d e v e l o p m e n t of réis t h e a c c e n t a p p a r e n t l y s h i f t e d for some u n k n o w n reason t o t h e vowel of t h e r o o t ; t h e n assimilation a n d c o n t r a c t i o n took p l a c e : regales > reais > *r'eais > reeis > réis.

4. An unaccented vowel in hiatus with an accented vowel remained u n c h a n g e d , if it was three steps more open (in t h e series a, ç, e, i or the series a, p, o, it) t h a n the accented paúl;

v o w e l : salire

múlam

>

> sair;

padülem

( f o r palüdem)

>

mua.

5. An unaccented vowel in h i a t u s with an accented vowel was dissimilated, if it was more close (in the series a, g, e, i or t h e series a, p, o, u) t h a n the accented vowel ( R R , X X I I , 4 3 ) : genestam vieira; vadit fidelem

vênarium > vai;

> ge¡sta

> veeiro malum

> *je(l

> giesta;

> vieiro;

> mao

(§ 35, 11) >

venerla

vênàtum

> man;

volare

> veeira

>

> veado [ v j a d u ] ; > voar

[vuar];

fiel.

A. L a t . colübram b e c a m e *colobram ( D ' O v i d i o , 16, n. 1), w h e n c e O P t g . coovra ( R L , X X V I I , 24). If b o t h o's of this f o r m were

§ 99-100] G E N E R A L

PHONOLOGICAL P H E N O M E N A

97

open, as the modern form cpbra indicates, Grandgent's opinion that pretonic o did not become o in Vulgar Latin seems to be confirmed (see § 7 A). For if the pretonic o was close, the modern form would be *cugvra. B. Further closing of the unaccented vowel of sudorem was not possible, hence the opening of the accented vowel: sudorem >

SUOT.

c. Dissimilation did not take place in the d e v e l o p m e n t of credentiam, because tonic e had been closed by the following yod

(§34, 2): credentiam > *creen(a > crenqa. D. Dissimilation did not take place in the development of avolum because tonic o was first closed by m e t a p h o n y : avolum

(Du Cange) > avgo > avoo > avb.

See § 123, 6 B.

6. An unaccented vowel in hiatus with an vowel became more close, if it was not in the same the accented vowel: caelum > ceu; pedonem > peao [ p j i w ] ; *venutum > viudo (old); soles > sois; doer [duer]; molinum > moinho [mwijiu].

accented series as peom > dolere >

A. Between tonic e and o an i [ j ] was inserted (see § 35, 7 ) .

7. An unaccented e in hiatus with an unaccented a or o often disappeared: monisterium > moesteiro > mosteiro; *me-lo > mo; *de-lo > do. 100.

METAPHONY.

1. M e t a p h o n y (umlaut) is the assimilating influence exerted by a final vowel upon a tonic vowel. Besides being found in Portuguese, it is found in Leonese (RA, X I V , 151-152), in some Italian dialects (M-L, It, 348-351), and in R u m a n i a n . In Portuguese, m e t a p h o n y is not an independent phonological p h e n o m e n o n ; it is indissolubly associated with inflection.1 I t seems to have occurred chiefly where discrimination or increased inflectional differentiation was felt 1

Inflection is u n d e r s t o o d t o include difference in declension such as t h e difference

between capelo a n d

capela.

98

PHONOLOGY

[§ 100

to be necessary. See RL, X X V I I I , 19. There was no metaphony in uninflected words, e.g., Iggo (from loco, R E W ) . And there was often no metaphony where there was no risk of ambiguity, e.g., do (from dolum, GVL, § 21), but avo (from avolum) because of avo. Because of this function of supplementing already existing inflection, metaphony is the least consistent phonological change in Portuguese and the one most easily interfered with by other forces, particularly by analogy. And because the interference of analogy continued long after the action of metaphony had ceased, the situation in Modern Portuguese is considerably different from that described by the grammarians of the sixteenth century. At the present time there is still some uncertainty in the use of metaphonic forms and great variation from one dialect to another. See Cavacas, 149, 162, and 163; Grund, I, 931, n. 4, and 1013, n. 2. 2. Evidence of the action of metaphony appears as early as the thirteenth century. Such forms as isto and tudo are found in the early Cancioneiros (cf. Cd'A, I, 370, n. 1; R L , X X I I I , 45, s.v. isto, and 88, s.v. todo). The form isto occurs in only one rime, disto : oantre tristo (i.e., o Antecristo) (CV, No. 1041) although there are several examples in C S M (pp. I l l , 141, 315, 474). Cf. BF, I, 348, n. 3. The form tudo does not occur in rime except in an obscure line of CV, No. 371 as emended by Nobiling (RF, X X V , 695). The rime medo : cedo is found repeatedly (CV, Nos. 241, 413, 421, 522, 865, 917) but the rime essa : abbadessa (CV, Nos. 944, 1137) seems to indicate that the e of essa had not yet opened. In the "Cancioneiro Geral," vowel changes which are evident in spelling, viz., from e to i and from o to u, have become much more numerous (cf. Cavacas, 184—185). It is therefore probable that the vowel changes which are not evident in spelling, viz., from g to e, e to Q to o, and o to p, have

§ 100-1013 GENERAL PHONOLOGICAL PHENOMENA 99 likewise become more numerous. Although this conclusion is not borne out by a study of the rimes (Cavacas, 167-172), the increasing number of unquestionably bad rimes seems to invalidate the testimony of rimes in this verse and in that of the following century. Perhaps printing and a consequent greater awareness of the written or printed word led poets to rime more for the eye alone than in the earlier period. Cf. Li esta > ¿sta (§35,8). 5. Final a opened tonic p to g: formgsa (§ 38, 7). 6. Final o closed tonic g to e: (§ 34, 8); and it closed tonic e to (§35,9). 7. Final o closed tonic g to (§ 37, 5); and it closed tonic o to (§38,8). 101.

formosam

> formosa

>

metum > mgdo > medo i: ipsum > esso > isso o: focum > fggo > fogo u: totum > todo > tudo

NASALIZATION.

1. T h e nasalization of vowels has been attributed to Celtic influence (cf. M - L , Intro, § 238). It was produced by nasal consonants in any position, initial, medial or final. Whether some form of nasal remained in the end and what that form was depended chiefly on the combination of sounds with which the nasal resonance was associated rather than on the original position of the nasal consonant. This is demonstrated by the following examples. Intervocalic n

Initial or Final Nasal

Resulting Nasal Sound

100

PHONOLOGY

tenes > tens (§ 78, 2) fines > fins (§ 78, 2 B) canes > cats (§ 78, 3) pones > poes (§ 78, 3) manum > mâo (§ 78, 3) vinum. > vinho (§ 78, 4) venitis > vindes (§ 78, 5) benedïcàmus > bengamos (old) (§78, 6)

[§ 101-102

nec > ne > nem (§ 66, 2 B) in > em (§ 97, 4) mihi > mi > mim (§ 66, 1 B) maires > mâes (§ 66, 1 B) *-ûdïnem > *-oen (§ 46, 2) > -de (old) vadunt > vâo nïdum > ninho (§ 66, 2 B) multum > muito [mûjntu] (§66, 1B) *nec-ûnum > ningum (old) (§66, 2 B)

ê ï âe ôe âo [n] toi

2. T h e nasal resonance generally closed an open tonic vowel: canto > canto [këntu]; tempus > tempos [tempu$]; pbntem > ponte [pônta]; but it prevented pretonic e and o from closing further: sentire > sentir [sentir]; voluntâtem > vontade [vôntada]. And it diphthongized final e, accented or unaccented, to [ i j ] : bene > bem [ b i j ] ; dëbent > devem [devij], 102.

ASSIBILATION.

1. T h e sound [s] in M o d e r n Portuguese comes f r o m : a) L a t . initial c followed by e or i: centum > cento; civitàtem > cidade (§ 62, 2). b) L a t . c preceded by a consonant and followed by e or i: mancipium > mancebo (§ 92, 1). c) L a t . c + i preceded by a vowel or a consonant : facio > faço; lanceâre > lançar (§ 89, 2). d) L a t . t + i preceded by a c o n s o n a n t : fortïam (Du Cange) > força"® 89, 3). e) L a t . initial s: sal > sal (§ 6 5 , 1). f ) L a t . medial ss: ossum > ôsso (§ 8 3 ) . g) L a t . medial x: dixit > disse; *traxuit > trouxe (§ 92, 9). 2. T h e sound [z] in M o d e r n Portuguese comes f r o m : a) L a t . c preceded by a vowel and followed by e or i: dices > dizes (§ 73, 2).

§ 102]

SPORADIC

CHANGES

101

b) L a t . i n t e r v o c a l i c s: causam > coisa (§ 76). c) L a t . t i p r e c e d e d b y a v o w e l : vitium > vezo (§ 89, 4). d) L a t . i n t e r v o c a l i c x in t h e prefix ex-: examen > exame (§ 92, 9 B ) . e) G r e e k initial f : zephyrum > zefiro (§ 65, 2). 3. T h e s o u n d [5] in M o d e r n P o r t u g u e s e c o m e s f r o m : a) L a t . sc p r e c e d e d b y a vowel a n d followed b y e or i: piscem > peixe (§ 92, 2). b) L a t . ss + i a n d sc + j : *bassium > baixo; fasciam > faixa (§ 89, 11). c) L a t . m e d i a l x: fraxinum > freixo (§ 92, 9). d) L a t . initial cl, fi, a n d pi: clavem > chave; flammam > chama; plagam > chaga (§ 67, 2, 3, a n d 5). e) Fr. ch: O l d F r . chapel > chefe (§ 62, 1 B). f) Lat. g) L a t . a final e feci > fiz 4. T h e a) L a t .

(Nascentes) > chapeu;

Fr. chef

s f o l l o w e d b y a s u r d : vestire > vestir (§ 97, 6 A). final s, a n d L a t . i n t e r v o c a l i c s a n d c followed b y o r i: duos > dois; mensem > mes; vicem > vez; (§ 97, 6 A; § 98, 1 a n d 2). s o u n d [3] in M o d e r n P o r t u g u e s e comes f r o m : initial g f o l l o w e d b y e or i: gentem > gente (§ 62,

4). b) L a t . g p r e c e d e d b y a c o n s o n a n t a n d followed b y e or i: longe > longe (§ 92, 3). c) L a t . g p r e c e d e d b y a v o w e l a n d followed b y e or i: vigilare > vigiar (§ 73, 4 c ) . d) L a t . i n i t i a l j: januarium > Janeiro (§ 62, 5). e) L a t . i n t e r v o c a l i c j: cujum > cujo (§ 73, 5). f) L a t . i n i t i a l d + j : diaria > geira (§ 70). g) L a t . m e d i a l d + j ; hodie > hoje (§ 89, 5). h) F r . j: F r . jardin > jardim; F r . forge > forja. i) L a t . s f o l l o w e d b y a s o n a n t : eleemosynam > esmola. A. T h e combination of s and a sonant did not exist in Classical Latin b u t came about in Vulgar Latin and early Portuguese through syncope or metathesis. Portuguese final s and z also have this sound when standing before a word beginning with a sonant.

PHONOLOGY

102

[§ 103-105

SPORADIC CHANGES 103.

ABBREVIATION.

Nicknames are sometimes made by omitting all that precedes the accented syllable: Manel (for Manuel) > Nel; José > Zé; Joaquim > Quim; Madalena > Lena; Manee a > Ñeca; Ricardo > Cardinho (with suffix -inho)\ Fernando > Nandinho; Francisco > Xico. In the language of children the tonic syllable sometimes alone survives but doubled: Helena > Lele; Herminia or Emilia > Mimi; Antonio > To to. 104.

ANALOGY.

Analogy, as frequently understood, m a y be defined as the process whereby a new form is invented in imitation of one or more other forms of the same paradigm or in imitation of the corresponding form of another paradigm. But analogy thus defined is really contamination, for true analogy involves at least four forms, viz., the form invented, another form of the same paradigm, and two forms of another paradigm which provide the model relationship on which the invention is based. 105.

ANAPTYXIS.

1. A parasitic vowel was sometimes inserted between the elements of a consonant group if one of these elements was I or r. This was brought about by the vocalic nature of I and the strongly vibrant nature of r: chronicam > caronica (old); clavicülam > cravelha > caravelha; sepulcrum > sapulcoro (FM, II, Glossary, s.v. sapulcro); februarium > fevereiro; fibram > jevera; florem > flor > felor (dialectal); glbriam > gloria > gueloria (dialectal); plantare > plantar > pelantar (dialectal); tertiarium > terceiro > tereceiro (FM, II, Glossary). See Esquisse, § 62; Opúsculos, II, 33. 2. A parasitic [a] is generally inserted in the pronunciation of learned words after the prefixes ab-, ad-, ob-

§ 105-106]

SPORADIC CHANGES

103

and sob- {sub-) in order t o a v o i d the j u x t a p o s i t i o n of the final c o n s o n a n t of the prefix w i t h the f o l l o w i n g initial c o n s o n a n t : P t g . advertir > [sdsvartir]; P t g . observar > [obasarvar]. See A S t N S , C X X V , 379; and R o m , X I I , 32, n. 1. A. If the [A] is not inserted, partial assimilation of the consonants takes place: obsequium > obsiquio [obzekju], 106.

APHERESIS.

1. T h e loss of the initial v o w e l of a word is a v e r y c o m m o n p h e n o m e n o n b u t it did n o t t a k e place w i t h a n y regularity: acumen > gume; P t g . alianqa > lian$a (popular, Barbosa, 2 3 ) ; apothecam > bodega; adulterium > dulterio (Eluc); *adluminäre > alumiar > lomear ( F M , II, G l o s s a r y ) ; hebdomäda > doma (old); eclipsem > cris ( o l d ) ; epistolam > pistola ( R L , X V I , 10); imaginäre > maginar (old and dialectal) ; *inodium ( R E W ) > enojo > nojo; *inamoräre > namorar; hominaticum ( M d ' A r ) > homenagem > menagem; hÖTologium > relogio; horrörem > TOT ( R L , V I I , 4 1 ) ; occasiönem > ocajom > cajom; iracundlam > *arigonha > rigonha ( R o m , X I , 9 5 ; I n e d i t o s , I, G l o s s a r y ; E l u c ) . A. T h e loss of initial a and o in nouns m a y have been occasioned by confusion of these vowels with t h e definite article. T h a t this confusion existed is shown by the change to the masculine gender of O P t g . cajom. B. I t has been argued t h a t namorar is a borrowing from Spanish, as initial in- would have become e in Portuguese (Liföes, 274-277). But it is probable t h a t apheresis took place in this word before the time of the fall of intervocalic n; apheresis took place in nojo, which does not appear to be a borrowing. I t is possible too t h a t the n of *inamöräre first lengthened as in Italian ( G i t , § 100). 2. In s o m e words, n o t o n l y the initial v o w e l b u t s o m e t i m e s the w h o l e initial syllable w a s lost: insaniam > sanha; spasmum > pasmo; P t g . absolutamente > solutamente (RL, V, 164). P t g . resurreifäo > surreiqäo ( R L , V, 164); P t g .

PHONOLOGY

104

[ § 106-108

destruir > struir ( R L , I I , 101); Ptg. sistema > sterna ( R L , X I , 278). A. T h e d e v e l o p m e n t of sanha f r o m insaniam

may have been

influenced by sanlem (RF, XX, 579). 107.

APOCOPE.

1. T h e loss of t h e last syllable of a word used as a proclitic is u n c o m m o n in Portuguese because of the comp a r a t i v e s t r e n g t h of unaccented syllables. T h e few cases t h a t are found are mostly of Spanish origin or due to Spanish influence: centum > cento > cem; dominum > donno > dom; grandem > grande > grao (used only in c o m p o u n d nouns); quantum > quanto > quao; tantum > tanto > tdo; sanctum > santo > sao; ille > ele > el (old); mille > mil; inde > em (old); multum > muito > mui. 2. Some apocopated forms are found only in stereotyped expressions, some only in place names: helium > bel, e.g., a bel prazer; casam > cas, e.g., em cas de; malum > mal, e.g., mal tempo; montem > mon, e.g., Monsanto; castellum > castel, e.g., Castel-Branco; vallem > val, e.g., Valverde. 108.

ASSIMILATION

AND

DISSIMILATION.

Assimilation and dissimilation took place between vowels and between consonants, sometimes when they were adj a c e n t and sometimes when they were remote, i.e., separated from each other by other sounds. For assimilation and dissimilation of a d j a c e n t vowels, see § 99. There was no dissimilation of a d j a c e n t consonants. I t does not seem possible to formulate very definite rules for the assimilation a n d dissimilation of vowels and consonants t h a t are remote from each other. T h e consonants involved are chiefly /, r, m, and n. A. Assimilation took place also between a vowel and a consonant, e.g., t h e opening effect of r on an adjacent pretonic e (see § 41, 9). B. M o s t of t h e changes in medial consonant groups (§§ 85-96) are examples of total or partial assimilation.

§ 109-110] 109.

SPORADIC

CHANGES

105

ASSIMILATION.

1. A s s i m i l a t i o n of r e m o t e v o w e l s : novaculam > navalha; P t g . riqueza > requeza ( o l d ) ; P t g . pedir > pidir (old a n d d i a l e c t a l ) ; P t g . cordura > curdura ( C V , N o . 6 9 0 ) . 2. A s s i m i l a t i o n of r e m o t e c o n s o n a n t s : parabolatn > paravra ( o l d ) ; animalia > alimalia ( o l d ) ; morbum > mormo; gloriam > groria (old). 3. A s s i m i l a t i o n of a d j a c e n t c o n s o n a n t s : molinarium > molneiro > moleiro; eleemosynam > esmolna (RL, X X V , 2 4 6 ; R L , X X V I I , 3 6 ; D C , 285) > esmola; lunulam > lula; *manere-a\o > manrei > marrei ( o l d ) ; sal nitrum > salitre; *amare-lo > amarlo > ama-lo; *totos-los > todollos (old); *deben-lo > devem-no; P t g . *fiz-lo > fi-lo. A. In these examples, the consonants were b r o u g h t t o g e t h e r by the fall of an intertonic vowel or a p o s t t o n i c penult, or b y new juxtapositions of closely related words. B. T h e assimilation was progressive in moleiro, esmola, etc. (see R L , I I I , 175; A S t N S , C X X I V , 337), unless m e t a t h e s i s of In took place before the assimilation (see R L , I I , 180-181); cf. monleiro (Eluc, s.v. conducteiro). T h e assimilation was progressive in devem-no, etc. also; this was due to t h e necessity of preserving the characteristic n of the ending of the third plural. 4 . A s s i m i l a t i o n of t h e n a s a l r e s o n a n c e : hac node > onte ( p o p u l a r ) > ontem; pectinem > peitem > peitem > pentem ( o l d a n d d i a l e c t a l ) ; nubem > nuve > *nuve > *nuvem > nuvem; mugilem > mugee > *muge > mugem. See L a n g u a g e , V I I , 142. 110.

DISSIMILATION.

1. D i s s i m i l a t i o n of r e m o t e v o w e l s : locustam > lagosta; sexaginta > sessenta > sassenta ( o l d ) ; dicebat > dizia > dezia (old) a n d dizia [dazin]; vxcinum > vezinho (old) a n d vizinho [ v a z i j i u ] ; divinum > devino (old) a n d divino [dav i n u ] ; rotatorem > redor; rotundum > redondo.

106

PHONOLOGY

[5 110-111

A. T h e dissimilation in v/zinho, devino and redondo originated in Vulgar Latin (GVL, § 229, 4 and 6). T h e pretonic i of modern vizinho and divino is merely orthographic. For dissimilation of remote vowels in Modern Portuguese, see Z R P h , X I X , 578, n. 1; AStNS, C X X V , 378-379. 2. D i s s i m i l a t i o n of r e m o t e c o n s o n a n t s : locale > lugar; animam > alma; memorare > nembrar > lembrar; priorem > priol ( o l d ) ; globellum ( D u C a n g e ) > lovelo > novelo; *ligaculum > legalho > negalho; arbitrum > alvid.ro (old). A. Although the two consonants in Lat. anima and in OPtg. nembrar were not the same, they were both nasals; the dissimilation was away from a nasal to an /. B. Sometimes the force of dissimilation was so intense t h a t it caused the disappearance of one of the like consonants: cribrum > crivo; rostrum > rosto; aratrum > arado; proprium > propio (old and dialectal). c. In the sixteenth century two b's or two p's in the same word were dissimilated so t h a t the first one was b, the second one v: blbere > bever (with shift of conjugation); vlvum > bivo. See SaM, Glossary, sub b. 3. D i s s i m i l a t i o n of t h e n a s a l r e s o n a n c e : campanam > campaa > campa; *ventanam > ventaa > venta; quintanam > quintaa > quinta; O P t g . pentem (Gil Vi, III, G l o s s a r y ) > pente. 111.

C O N F U S I O N OF P R E F I X E S A N D OF I N I T I A L

SYLLABLES.

1. B e c a u s e of its c o m m o n o c c u r r e n c e , initial es- ( f r o m L a t . ex- a n d f r o m p r o s t h e t i c e plus s followed b y a c o n s o n a n t ) s o m e t i m e s t o o k t h e p l a c e of initial as- ( f r o m L a t . abs, as- a n d aus-) a n d initial os- ( f r o m L a t . obs-): abscond.it > asconde > esconde; abstinentiam > estenga ( o l d ) ; asparagum > espargo; auscultdre > ascuitar > escutar; obscurum > escuro. A. T h e opposite exchange took place in OPtg. asperar (for esperar < sperdre), which may have developed in imitation of asconder, ascuitar, etc. (Cd'A, I, 366) or under the influence of

SPORADIC CHANGES

§111] Lat. aspectare 1 c).

107

( P h M , II, 163) or of Ptg. aspirar

(Huber, § 119,

2. T h e confusion between pretonic an- (or am-) and en(or em-) arose probably as a result of the similarity in the pronunciation of these nasal sounds before the time when en- closed to [i]: anguilam ( S o m m e r , 205) > anguia (old) a n d enguia; emcorar (old)

and

amperador resplandecente; *in-tum

ampullam

(see emcorou,

> empola; Crest,

amparar;

imperatdrem

(old);

*resplendescentem

inter

*ancorare

110);

>

> emperador

and

emparar

(old)

> resplendecente

(in p r o c l i s i s ) > entre

> entao a n d antam

> ancorar

*imparare

and and

a n d antre

(old);

(old).

A. This confusion is very common in certain dialects (see Esquisse, § 49 b and § SO d; RL, VII, 37; RL, X I , 271; RL, X I X , 1 7 2 ) ; sometimes even in the tonic position (Esquisse, § 44 m; RL, XII, 3 0 7 ) . B. T h e change of entre to antre has been explained as due to the influence of ante (RL, VIII, 70).

3. A sort of fusion or contamination took place between initial eix- (from L a t . ex- and ax- followed by a vowel and L a t . asc + i) and initial ens- (from Lat. ins-), resulting in t h e n e w p r e f i x enx- [1$]: exemplum

> eixempro

>

enxempro

(old); exheredare > enxerdar (Crest, G l o s s a r y ) ; exsúgere > enxugar (with s h i f t of c o n j u g a t i o n ) ; examen > enxame; *exuvialem ( R L , V I I , 1 2 0 - 1 2 1 ) > enxoval; *exaltiare > enxalfar; *exaquáre > enxaguar; axungiam > enxúndia (for di, see G r u n d , I, 9 9 1 ) ; asciolam > enxó; *asciatam > enxada; *insapidum > enxabido; insertare > enxertar; *insulfurdre > enxofrar. A. For the loss of j in eix-, cf. § 92, 7 A. B. Similar confusion has occurred in Old Portuguese and in modern dialects between initial e- and in-: eligere > OPtg. enieger (with shift of conjugation); aeternum > eterno > interno (Esquisse, § 50 a); educare > educar > inducar (RL, X I , 274); electionem > eleiqdo > inleifao (RL, X I , 140). c. The change of eix- to enx- has been attributed to the ease

108

PHONOLOGY

[ § 112

with which t h e vowel i became nasalized without a n y external influence (Grund, I, 981). D. T h e word ensaio (from exagium) is a borrowing from Spanish. E. T h e * and t h e prefix of enxofre (from suljurem) developed in imitation of t h e verb enxofrar. 112.

CONTAMINATION.

1. C o n t a m i n a t i o n is t h e p r o c e s s w h e r e b y t h e d e v e l o p m e n t of a w o r d is i n f l u e n c e d b y a n o t h e r w o r d of s i m i l a r s o u n d or b y a n o t h e r w o r d of s i m i l a r o r r e l a t e d m e a n i n g or u s e ; in s o m e cases t h e o t h e r w o r d is b o t h of s i m i l a r s o u n d a n d of s i m i l a r or r e l a t e d m e a n i n g or u s e : castellum > castrello ( E l u c ) w i t h t h e r of castro ( f r o m castrum); *stellam ( G V L , § 163) > estrela w i t h t h e r of astro ( f r o m astrum); astrologum > estrollogo (old) w i t h t h e e of estrela; caelestem > celestrtr (old) w i t h t h e r of terrestre ( f r o m terrestrem); *admanescere > amanhecer w i t h t h e nh of manha (from *maneanam); vinetum > vinhedo w i t h t h e nh of vinho ( f r o m vinum); repositam > resposta w i t h t h e s of responder (from respondere); sic > sim w i t h t h e n a s a l i z a t i o n of nom or nao ( f r o m non); puppem > popa w i t h final a f r o m proa ( f r o m prdram) t h r o u g h the expression de popa a proa; F r . camion > camiao > caminhao w i t h the nh of camtnho or caminhar. A. A form terreste in rime with celeste (Lusiadas, canto VII, stanza 6) arose either through the influence of celeste or by dissimilation of the r's. B. T h e nasalization in sim has also been explained as due to the presence of the sibilant a n d the vowel i (Grund, I, 981). c. T h e word asa has been explained (RL, X I I I , 258-261) as having developed by c o n t a m i n a t i o n : alas > aas > asas with medial s from the z of axes (from acies); but see P h M , II, 153 and G r u n d , I, 970. T h e word sarar has been explained (RL, X I I I , 389-392) as having developed by c o n t a m i n a t i o n : sandre > saar > sar > sarar with t h e additional syllable f r o m farrar; b u t see R o m , X I , 95-96. D. Endings have sometimes developed by c o n t a m i n a t i o n : devo-

§112-113]

SPORADIC CHANGES

109

tionem > dívo(om > devafom > deva(ao with the a of the common ending -a(om or -a$ao (from -ationem); institutionem > institui(ao with the i of the ending -ui(ao (from -uctionem); libcrtatem > liberdade

with -dade from nouns in which both t's were inter-

vocalic, e.g., caridade (from

caritatem).

2. Popular etymology is a form of c o n t a m i n a t i o n : aquaeductum > aqueduto > arqueduto through the influence of arco (from arcum) (RL, X X X I I , 283); sacristánum (Du Cange) > sacristao > sachristao through the influence of christao (from christianum), and sancristao through the influence of sam (from sanctum), and sanchristao through the influence of both these words; Fr. canapé > camapé through the influence of Ptg. cama; Satanas > Satanás > Santanás through the influence of sam (from sanctum) (Opúsculos, I, 344); litaniam > ladainha through the influence of O P t g . ladino (from latinum); ecclesiasticum > eccresiastico (RL, V, 134) > crelegiastico through the influence of O P t g . creligo (from clericum) (see R L , X X V I , 142); Ptg. tintura diodo > tintura d'ódio through the influence of odio (Esquisse, § 78 b); diluvium > dinuvio (dialectal) through the influence of nuvem in the expression nuvem de água (RL, X I I , 306). 3. False regression is a form of contamination, which was often only orthographic: innocentem > innocente > ignórente (Fab, Glossary; BF, I I I , 64); hymnos > innos > jgnos (Abraham, § 19, 2) with g from other words in which gn was etymologically correct although the g was silent; cf. inorancia ( P M H , Scriptores, 77); spissum > espesso > esperso (RL, X V I , 6) with r f r o m other words in which rs was correct. 113.

EPENTHESIS.

1. A b was inserted between m and r, and between m and l: hümerum > *omro > ombro; memorare > *memrar > nembrar > lembrar; cumulum > *comlo > combro; similare > semblar > sembrar; cücümerem > cogombro (with shift of declension); Arabic alkhomra > alfombra.

PHONOLOGY

110

[ 5 113-114

2. A d was inserted between n and r: konôrâre > (old); ingenerâre > engendrar.

ondrar

A. T h i s w a s n o t a P o r t u g u e s e c h a n g e b u t o c c u r r e d o n l y in w o r d s b o r r o w e d f r o m S p a n i s h ( C o m p , 144).

3. A v was inserted between t h e d i p h t h o n g ou a n d a following vowel: audit > ouve; laudat > louva; claudere > ckouvir (old) (with shift of c o n j u g a t i o n ) ; gaudibat > gouuja ( A H P , I I I , 95), i.e., gouvia; caulem > couve. Cf. F r . pouvoir. A. T h i s v w a s p r o d u c e d b y t h e labial e l e m e n t of t h e d i p h t h o n g ou. A t first it did n o t a p p e a r in v e r b f o r m s a c c e n t e d on t h e e n d i n g ; its l a t e r s p r e a d t o t h e s e f o r m s w a s d u e t o a n a l o g y ( R F , X X V , 664).

4. I n some words an r was inserted: tonâre >

troar.

A. T h e T of O P t g . celestre ( f r o m caelestem) is t h o u g h t t o b e a n e p e n t h e t i c r ( A S t N S , L X V , 40). B u t see § 112, 1.

5. A p h e n o m e n o n m u c h like epenthesis is the exchange of one consonant for a n o t h e r : mêdicam > melga; magidam > *madiga > malga; pallidum. > pardo; papyrum > papel; ulicem > urze; judicâre > julgar; portatïcum (Du Cange) > portadego > portalgo (old). A. T h e c h a n g e f r o m d t o / is f o u n d in V u l g a r L a t i n , e.g., non alipes ( A p P r ) . T h e c h a n g e f r o m / t o r is c o m m o n d e v e l o p m e n t of c o n s o n a n t g r o u p s in P o r t u g u e s e (see § 86, §95,5). B. I n t h e Old P o r t u g u e s e f o r m s juigar a n d portadego, in I did n o t develop, t h e i n t e r t o n i c vowel a n d t h e p o s t t o n i c d i d n o t fall (see § 53 a n d § 58).

adipes in t h e 1 and which penult

6. T h e d e v e l o p m e n t of a consonantal n f r o m a preceding nasalized vowel should not be confused with epenthesis, e.g., minûtias > mïuças > miunças. See § 78, 5. 114.

HAPLOLOGY.

An unaccented group of sounds sometimes fell when t h e preceding or following group was identical or nearly so, or

§ 114-115]

SPORADIC

CHANGES

111

b e g a n or e n d e d w i t h t h e s a m e c o n s o n a n t : perditam > * perdida > perda; vendïtam > vendida > venda; simplices > simprezes > simprez a n d simples ( p l . ) ; aurïfïces > ourivezes > ourives ( p l . ) ; sóror > sór; rotatôrem > *rtdador > redor; jejunàre > jejuar > juar ( R L , V I I , 2 4 5 ) ; *-tatôsum > -dadoso > -doso, e.g., *aetatôsum > *idadoso > idoso. A. T h e a d j e c t i v e suffix *-tatôsum was formed by adding -ôsum to the noun ending -tàt(em). Other examples are bondoso, maldoso, and cuidoso. But see Traite, 336. B. Haplology occurred also in popular and dialectal P o r t u guese in combinations of two or more words: duas vezes > duas bês (Esquisse, § 70 a); Madre de Deus > Madre Deus (popular); pelo amor de Deus > pelo amor Deus (popular). P e r h a p s the expressions, en nom Deus, found in a document of 1271 (RL, V I I I , 40) and filio deus (Crest, 8) are examples of haplology. 115.

METATHESIS.

1. T w o a d j a c e n t s o u n d s w e r e s o m e t i m e s r e v e r s e d . T h e s e s o u n d s c o u l d b e t w o v o w e l s : genücülum (Du Cange) > geolho > joelho; dehonestâre > deostar > doestar; t w o c o n s o n a n t s : merülum > melro > merlo ( d i a l e c t a l , R L , I V , 2 2 0 ) ; sibilâre > silvar; a v o w e l a n d a c o n s o n a n t : *inodium > enojo > enjôo; instrumentum > estormento (old). 2. S o m e t i m e s a c o n s o n a n t ( u s u a l l y r) w a s s h i f t e d f r o m o n e s y l l a b l e t o a n o t h e r : fenestram > feestra > fresta; pigrïlïam > pegriça > preguiça; tënebras > teevras > trevas; fabrïcam > fravega ( R L , X X V , 247) ; capistrum > cabresto; praesaepem > pesebre; satis/acere > satisfazer ( w i t h s h i f t of c o n j u g a t i o n ) > sastijazer (popular). A. Learned fábrica had the same change in dialectal fráveca ( R L , X X V I I I , 230). B. A similar shift of the nasal resonance is also f o u n d : anâtem > ánade > *¿ade (RL, X X V I , 112) > adem (NS, X I , 144). For criticism and a different explanation, see BF, I, 164. 3. S o m e t i m e s t w o c o n s o n a n t s in d i f f e r e n t s y l l a b l e s ex-

112

PHONOLOGY

[§ 116-117

c h a n g e d p o s i t i o n s : eleemosynam > e smoina > esmola; plantare > chantar > tanchar; machinam > maquina > manica ( p o p u l a r ) ; catalògum > catalogo > catàgolo ( p o p u l a r ) ; P t g . bicarbonato > bicabornato > bitabornaco ( R L , X X V I I , 112); P t g . Ilhavo > Ivalho. A. This type of metathesis took place as a regular change between the / of an initial group and the r of a following syllable: clysterem > cristel; clerxcum > creligo (old); fidrem > frol (old). Perhaps the I first changed to r according to § 67, with subsequent dissimilation of the r of the following syllable. See AStNS, LXV, 44, s.v. frol; and H - M P , I, 608, footnote. 116.

PARAGOGE.

1. A n s w a s o f t e n a d d e d t o a d v e r b s e n d i n g in a v o w e l : ante > antes; *extunce (cf. G V L , § 47) > estonce > estonces (old); numquam > nunca > nuncas (old); -mente (GVL, § 41) > -mente > -mentes (Esquisse, § 77 a). A. This change arose perhaps through the influence of adverbs ending in s such as mais (from magis), OPtg. pos (from post), etc. 2. A p a r a g o g i c vowel (usually e) was a d d e d c o m m o n l y in dialectal P o r t u g u e s e t o words e n d i n g chiefly in I a n d r: animal > animale; mare > mar > mare. See Esquisse, § 59 d a n d f, § 66 c ; R L , X X V I I , 122; C d ' A , I, 367; S-A, 2 0 - 2 1 . B u t see M e i e r , 44. 3. A n e is a d d e d in M o d e r n P o r t u g u e s e to borrowings e n d i n g in a c o n s o n a n t which does n o t occur as a final in n a t i v e w o r d s : F r . chic > chique; F r . bric-à-brac > bricabraque; F r . kiosk > quiosque; E n g . beef > bife; E n g . club > clube; E n g . film > filme; A r a b i c al-khaiyat > alfaiate. 1 1 7 . PROSTHETIC

a.

I n e a r l y P o r t u g u e s e an a was prefixed to words beginning in r: remittere > remeter (with shift of c o n j u g a t i o n ) > arremeter; ranam > ràa > arra (old a n d p o p u l a r ) . See R o m , X I , 7 5 - 7 9 ; M u s , I I I , 2 1 6 - 2 1 7 ; A b r a h a m , § 32, 2.

§ 117-119]

SPORADIC

CHANGES

113

A. T h e r was lengthened in order to preserve the sound it had in the initial position. Prosthetic a before r is a different phenomenon from OPtg. er (and ar), which was a separable particle arising from the Latin prefix re- (see Rom, I X , 580-589). B. In some words, initial a came from the Latin prefix ad-: adplicare (AStNS, C X X I V , 339) > achegar; *adoculdre (Rom, X I , 90-91) > aolhar > olhar; advenire > avir. In others, it is thought to have arisen through agglutination of the definite article: mora (pi.) > mora > amora; nanam > ana; medielatem > metade > ametade. See Comp, 266-267. c. Through the influence of words with initial a from the Lati n prefix ad- and from the agglutinated article, prosthetic a soon spread to other words which did not begin in r: memorare > lembrar > alembrar (old); monstráre > mostrar > amostrar; *minatiáre > meaqar > ameaqar (AStNS, C X X I V , 335). See Comp, 58-59. This was particularly common in dialectal Portuguese (see Esquisse, § 66 a; RL, X I , 142 and 272; RL, X I I , 307). T h e use of the Arabic article al as a prefix m a y have also been a contributing factor (M-L, Gram, I, § 383). D. OPtg. atal (from talem) and atanto (from tantum) may have developed through the spread of prosthetic a. T h e y have been explained as having arisen in Vulgar Latin through the influence of words like aqui and aquel (Huber, § 108, 1). 118.

SANDHI.

T h e initial or final sound of a word o f t e n d e v e l o p e d as in the medial position because of the j u x t a p o s i t i o n of the word with the preceding or following word. T h i s is a c o m m o n p h e n o m e n o n in old and dialectal P o r t u g u e s e , b u t with the t e n d e n c y toward word individuation of the m o d e r n literary language, it has largely disappeared e x c e p t b e t w e e n words which were closely related s y n t a c t i c a l l y , where it m a y be called syntactical phonology, e.g., c o m b i n a t i o n s of conj u n c t i v e pronouns with verbs and of prepositions with articles and pronouns. a) T o n i c i followed by a or o: Ptg. vim aqui > vi nh aqui (Esquisse, § 159); Ptg. dá me a mim o pau > dá m'a mim nho pau (Opúsculos, I, 342). See § 78, 4.

PHONOLOGY

114

[5119

b) Assimilation of vowels in hiatus: Ptg. até agora > ateegora > ategora (old); P t g . outra hora > outrora; Ptg. é urna hora > é urn hora [eumoru] (colloquial). See § 9 9 , 3. c) Assimilation of adjacent consonants: nao na tinha ( R L , X I , 1 4 0 - 1 4 1 ) for nao a tinha; chamem no conde ( R L , V I , 151) for chamem o conde; se fezer mha senholo que tern no corado ( C B , No. 223) for se fezer mha senhor o que tern no corafo. See § 109, 3. 119.

WRONG

DIVISION.

T w o words standing in frequent juxtaposition were sometimes wrongly divided, with the result that a vowel, a consonant, or a whole syllable was permanently attached to or detached from the beginning of the second of them. a) T h e word maluta arose from the addition of a syllable of the indefinite article uma to luta ( R H i , V, 4 2 3 ) . b) T h e word zorate arose from the addition of the j of the masculine plural of the definite article os to the plural form orates ( R H i , V, 4 2 5 ; see also R R , I I I , 3 1 0 - 3 1 2 ) . c) T h e word dil (dialectal for ir) arose from the addition of the d of the preposition de to il in the expression hei-d'il (for hei de ir) ( R L , X X X I , 199). d) T h e word Thiago arose from the addition of the t of San/' (for Santo) to lago. T h e graph th was used for t by false regression. e) T h e word ume arose from the separation of the initial a of aume, which had contracted with the final a of -pedra in the compound pedra-ume (from petra alümen). See RL, V, 59-60. f) T h e word ameixa arose from the separation of the initial d of *damascea in the expression pruna *damascea (for pruna damascena). g) T h e word aleijao arose from the addition of the definite article a to leijáo (from laesidnem). Later, the final a of the feminine indefinite article, perhaps in its earlier form üa, was contracted with this initial a and the contracted

§ 119]

PHONOLOGY

115

vowel was detached f r o m the article, thus changing the g e n d e r of t h e n o u n : ûa aleijâo > ûaleijào > û aleijâo, i.e., um aleijâo ( R L , I I I , 131). h) T h e w o r d léste a r o s e f r o m t h e a d d i t i o n of t h e F r e n c h d e f i n i t e a r t i c l e to t h e w o r d este. A. When the article is attached to the following word, as in the last two examples, the phenomenon is really one of agglutination and not of wrong division.

MORPHOLOGY NOUNS 120.

DECLENSIONS.

There was little left of the Classical Latin declensions in Vulgar Latin. T h e fourth and fifth declensions, the neuter gender, and all cases except the nominative and accusative disappeared. With the disappearance of the nominative in Portuguese, case distinction came to an end. Inflection for number alone remained. T h e form derived from the Latin accusative now has to function as subject, as object of a verb, and as object of a preposition. 121.

NOMINATIVE

SINGULAR.

There are a few exceptional nouns which have survived in the nominative instead of the accusative. T h e most important of these are : bübo > bufo; cancer > cáncer; daemon > demo; deus > deus; draco > drago; index > éndes; gürgülio > gorgulho; magister > mestre; homo > ome (old indef. pron.); sóror > sor; Carôlus > Carlos; Marcus > Marcos. A. An analogical plural of these nouns was formed by the addition of s or es, e.g., deuses, gorgulhos. Deuses replaced an older plural deus (from dens), e.g., todos ot Dens (Inéditos, II, 1 1 7 ) . An analogical feminine deusa was formed from deus. OPtg. deessa was probably borrowed from French déesse, which came from a Latin form with the Greek ending -ïssa. B. The accusative of learned cancer survived with a different meaning: cancrum > cancro. T h e form drago has generally been supplanted by dragâo (from the accusative draconem). Both drago and dragom were used interchangeably in Old Portuguese, e.g., Rom, X I , 382. c. The nominative *serpes (from serpens) was apparently taken for an accusative plural, with the result that a new singular, serpe was formed by dropping the final s. 117

118

MORPHOLOGY

[§ 121-122

D. I n mestre m e t a t h e s i s of t h e last syllable took place according t o § 97, 5. F o r a d i f f e r e n t e x p l a n a t i o n of the final e of mestre, see § 48, 6. E. T h e g e n i t i v e s u r v i v e d in t h e n a m e s of the d a y s of the week: Martis > martes (old); Jovxs > joves (old); Veneris > vernes ( R L , V I I , 194). T h e s of O P t g . lues (from lunae) and of OGal. mercores ( f r o m Mercurii) developed b y analogy with martes, joves a n d vernes. T h e s e n a m e s did not survive except in Galician, w h e r e only lus a n d martes h a v e survived t o d a y . See Z R P h , X I X , 614. B e i n g b a s e d on t h e h e a t h e n n a m e s of gods and p l a n e t s t h e y were p r o s c r i b e d in t h e y e a r 316 b y P o p e Sylvester, w h o o r d a i n e d t h a t t h e y should h e n c e f o r t h be replaced b y feria secunda, etc. See H a m p s o n , I, 137-138, s.v. feria. Vestiges of t h e a b l a t i v e are f o u n d in agora; hac node > ontem; hoc anno > meliore > melhor a n d pejore > pior. On t h e s u r v i v a l of t h e n o m i n a t i v e a c c u s a t i v e , see R H i , I I , 117-119; R o m , L i f o e s , 3 9 - 4 7 ; H - M P , I, 6 0 7 - 6 0 9 . 122.

FIRST

a few a d v e r b s : hac hora > ogano (old); and p e r h a p s and o t h e r cases t h a n t h e X I , 79-81; M u s , I I I , 217;

DECLENSION.

1. T h e e n d i n g s of t h e a c c u s a t i v e singular a n d plural of n o u n s of t h e C l a s s i c a l L a t i n first d e c l e n s i o n a n d t h e e n d i n g s d e r i v e d f r o m t h e m in P o r t u g u e s e are:

Sg. PI.

Classical L a t i n -am -as

-as

2. T h e s e e n d i n g s a n d a p r e c e d i n g t o n i c o in h i a t u s w e r e a s s i m i l a t e d a n d c o n t r a c t e d : molam

> mo.

S e e § 99, 3.

3. T h e s e e n d i n g s a n d a p r e c e d i n g t o n i c a in h i a t u s b e c a m e -a a n d -as: lanam irmas.

> laa > la; germanas

> irmaas

>

See § 78, 2.

A. I n c e r t a i n dialects t h e plural e n d i n g -as b e c a m e -aes b y a n a l o g y with n o u n s of t h e t h i r d declension in -aes ( f r o m -anes): mattidnas ( R E W ) > magas > ma(aes (RL, X I , 142; R L , X I I , 309; Esquisse, § 70 a).

§ 122-123]

119

NOUNS

B. In a few words the nasal resonance was lost by dissimilation and the accent shifted from the ending probably by analogy with other nouns ending in unaccented a: campanam > campa > campa; *ventanam > venta > venta; quintanam > quinta > quinta. Rodrigues Lapa does not consider this explanation satisfactory (BF, I, 163-164). However, it is noteworthy t h a t those words in which the accent did not shift, viz., avela, irma, manha, mafa, etc. are words in which the nasal resonance remained on the ending and most of them words without a nasal in the preceding syllable. 123.

SECOND

AND

FOURTH

DECLENSIONS.

1. T h e e n d i n g s of t h e a c c u s a t i v e s i n g u l a r a n d p l u r a l of m a s c u l i n e a n d n e u t e r n o u n s of t h e C l a s s i c a l L a t i n s e c o n d a n d f o u r t h declensions a n d t h e e n d i n g s in P o r t u g u e s e resulting f r o m their fusion are:

Sg. PI.

CI. L. CI. L. CI. L. CI. L. 2d Decl. 4th Decl. 2d Decl. 4th Decl. Masc. Masc. Neut. Neut. -um -um -um -u -os -us -a -ua

Portuguese -o (§ 12, 8; § 48, 1) -os

T h e f u s i o n c o n s i s t e d in t h e a d o p t i o n in V u l g a r L a t i n of t h e e n d i n g s of m a s c u l i n e n o u n s of t h e s e c o n d d e c l e n s i o n b y m a s c u l i n e n o u n s of t h e f o u r t h d e c l e n s i o n a n d n e u t e r n o u n s of t h e s e c o n d a n d f o u r t h d e c l e n s i o n s : passus > passos; castella > castelos; cornu > corno; cornua > cornos. See G V L , § 347 a n d § 355, 1. 2. T h e s e e n d i n g s in h i a t u s w i t h a p r e c e d i n g t o n i c a or g b e c a m e - u a n d -us: caelum > ceu; palos > paus. See § 48, 2. T h e y c o n t r a c t e d t o -u a n d -us w i t h a t o n i c u w i t h w h i c h t h e y w e r e in h i a t u s : culum > cu. See § 48, 5. 3. T h e s e e n d i n g s a n d a p r e c e d i n g a in h i a t u s b e c a m e -ao a n d -aos: granum > grao; orphanos > orjdos. See § 78, 3 a n d 3 c. A. In m a n y nouns of late Latin formation or adoption the ending derived from -anum was replaced by the ending derived

120

MORPHOLOGY

[ § 123

from -ánem: sacristanum (Du Cange) > sacristam (old); cappellánum (Du Cange) > capelam (old); castellánum > castelam (old); alemannum > alemam (old). The plurals accordingly end in -aes: sacristies; capelaes; castelaes; alemáes. See Manual, § 83, 4; and D'Ovidio, 22, n. 2. On Sp. alemán, see Hanssen, § 15, 1. The singular ending -ao of all these nouns in Modern Portuguese developed according to § 157, 2. B. In dialectal Portuguese the plural ending -aos became -oes and -aes by analogy with nouns which in Latin belonged to the third declension and ended in -ones and -ánes: germanos > irmdes and irmaes (RL, XI, 142 and 280; RL, XII, 309; Esquisse, § 70 a; Grund, I, 1013). 4. These endings and a preceding tonic -o in hiatus became -om and -ons: donum > dom; sonos > sons. See § 78, 2. A. The form does, the OPtg. plural of dom, was formed on the analogy of nouns which in Latin belonged to the third declension and ended in -ones. The form which has survived in Modern Portuguese is dons. B. OPtg. padrom (from patronum) became padrao by analogy with nouns which in Latin belonged to the third declension and ended in -onem. Later padrao became patrao through the influence of learned patrono. 5. These endings and a preceding tonic ú in hiatus became -um and -uns: jejunum > jejum (see Estudos, II, 73-74). See § 4 8 , 5 A; § 7 8 , 2 B. 6. In nouns with Classical L a t i n radical o m e t a p h o n y generally took place in the singular (§ 100, 7 ) : popülum > povo; hortum > horto; but not in the p l u r a l : populos > pgvos; hórtos > hprtos. T h e pronunciation of final o became [u] at an early d a t e (see § 48, 1 A). T h i s extreme closeness of the final vowel increased its assimilating effect on the tonic vowel. I t must be assumed t h a t the o of final -os did not acquire this close pronunciation until after the time of the action of m e t a p h o n y ; therefore, final -os did not have a n y effect on

§ 123]

NOUNS

121

t h e t o n i c v o w e l a t all. T h i s is w h a t A l m e i d a C a v a c a s m e a n s w h e n he s a y s t h a t t h e e n d i n g of t h e p l u r a l m u s t h a v e h a d " u m s o m d i f e r e n t e m e n o s s u r d o " t h a n t h e e n d i n g of t h e s i n g u l a r ( C a v a c a s , 145). I t is i n t e r e s t i n g t o n o t e t h a t u n a c c e n t e d i n i t i a l o, f o l l o w e d b y s, is still p r o n o u n c e d [o] in s o m e w o r d s , e.g., hospedagem, hospital. a . A t t e m p t s have been made to account for the supposed difference between Old Portuguese final -o and -os by tracing it to the difference between Classical Latin u and o in -um and -os (see Grund, I, 464; M-L, Gram, I, § 641; Huber, § 93,2; M-P, Orig., § 35, 3). T h e Classical Latin difference has apparently survived in Asturian, Central Italian and the dialect of Logudoro in northern Sardinia (see M-L, Gram, I, § 3 0 8 ) . However, metaphony was caused in Portuguese not only by final o from CI. L. u but also by final o from CI. L. o. e.g., volvo > volro ( § 176, 2 ) . In an effort to explain the lack of m e t a p h o n y in the plural, Gon^alves Viana likewise goes back to an early distinction between singular and plural endings. H e suggests t h a t the radical vowel of the plural remained open in neuter nouns because of the ending -a, t h a t it continued open even after this ending was replaced b y the masculine ending -os, and t h a t the open radical vowel of originally masculine nouns remained open by analogy with originally neuter nouns (see Rom, X I I , 74 and 80). B u t this t h e o r y is untenable as it is certain t h a t -os had replaced -a long before the action of m e t a p h o n y had ceased (cf. Z R P h , I X , 143). b . M e t a p h o n y is found also in nouns in which the ending and radical p c o n t r a c t e d : avolurn (Du Cange) > avo; tertiolum (Du Cange) > terqb. See § 37, 5 a . c. T h e r e are m a n y learned and semi-learned nouns in which m e t a p h o n y did not t a k e place: globum > glgbo; modum > mpdo. d . Some nouns with Classical Latin o (u) adopted a radical p in t h e plural by analogy with nouns in which m e t a p h o n y had t a k e n place in the singular: furnos > fgrnos; turdos > tprdos. B u t there are m a n y nouns with Classical Latin o (m) in which neither the action of m e t a p h o n y in the singular nor of analogy in t h e plural took place: lupum > lobo and lupos > lobos; lutum > lodo and lutos > lodos. e . T h e o of olho was due to the palatal (see § 37, 3 b ) . And

122

MORPHOLOGY

[ § 123

the g of t h e plural glhos developed by analogy with nouns in which m e t a p h o n y had t a k e n place in the singular. F. N e i t h e r m e t a p h o n y nor the analogical simulation of met a p h o n y took place if radical o was followed by a nasal consonant, because a nasal consonant closed an open o and kept a close o u n c h a n g e d : dominum > dgno and domlnos > donos; lumbum > lombo and lumbos > lombos. See § 37, 6. G. N o u n s with t h e suffix -ottum (GVL, § 37) have o in t h e plural b y analogy with t h e singular: -otium > -oto and -ottos > -otos. H. I n dialectal P o r t u g u e s e m e t a p h o n y is sometimes found in both t h e singular and t h e plural and sometimes in the plural and not in t h e singular. A n d in a given dialect there is generally much inconsistency f r o m one word to another. See RL, IV, 328; R L , V I I , 45; R L , X I , 142 and 270; RL, X I I , 309; R L , X I V , 83; R L , X I X , 172; R L , X X V I I , 131; G r u n d , I, 931. T h i s wide variation b e t w e e n dialects and between words in the same dialect is doubtless d u e to a variation in the relative force of m e t a p h o n y and analogy. i. L a t . ovum developed as follows: ovum > gum (GVL, § 324) > gum (GVL, § 167) > gvum (GVL, § 167) > gvo (cf. Sp. huevo) > gvo. T h e last step was b r o u g h t a b o u t by m e t a p h o n y . 7. I n n o u n s w i t h C l a s s i c a l L a t i n r a d i c a l e m e t a p h o n y t o o k p l a c e in b o t h t h e s i n g u l a r a n d p l u r a l : catellum > cadelo a n d catellos > cadelos; cappellum ( R E W ) > capelo; cappellos > capelos. A. While final -os did not affect radical g it was probably sufficiently close t o affect radical Or perhaps the e of t h e plural of these nouns developed by analogy with the singular. 8. S o m e f e m i n i n e n o u n s of t h e s e c o n d a n d f o u r t h d e c l e n sions a d o p t e d t h e e n d i n g s d e r i v e d f r o m t h e first d e c l e n s i o n : amethystum > ametista; sapphirum > safira; fagum > faia; socrum (f.) > sogra; w h i l e o t h e r s did n o t c h a n g e t h e i r f o r m b u t b e c a m e m a s c u l i n e : fraxinum > freixo; pinum > pinho ( o l d ) ; alaternum > aderno. O n e n o u n , mao ( f r o m manum), did n o t a d o p t t h e endings derived f r o m the first declension o r c h a n g e its g e n d e r .

§ 123-124]

NOUNS

123

9. Some neuter plurals retained the ending -a and became feminine singulars: arma > arma; vota > boda; dona > doa (Crest, Glossary); ligna > lenha; folia > folha; cornua > coma. These nouns formed a new plural by t h e addition of s and their forms correspond accordingly to those derived from feminine nouns of the first declension. Some of t h e m still retain the collective force which t h e y all probably acquired in the period of transition f r o m plural to singular, e.g., lenha—"firewood." 1 2 4 . THIRD AND FIFTH DECLENSIONS.

1. T h e endings of the accusative singular and plural of masculine, feminine, and neuter nouns of the Classical Latin third declension and of nouns of the Classical Latin fifth declension and the endings in Portuguese resulting f r o m their fusion are:

Sg. Pl.

Cl. L. 3d Decl.

Cl. L. 3d Decl.

Cl. L. Sth D e c l

Masc. Fem. -em -es

Neut. -(e) -(i)a

-em -es

Portuguese

T h e accusative singular and plural of nouns of the fifth declension are identical with the accusative singular and plural of masculine and feminine nouns of t h e third declension. T h e fusion consisted in the a d a p t a t i o n of the stems and endings of neuter nouns to correspond with masculine nouns (see sections 5, 6 and 7 below). T h e nasal resonance survived in O P t g . rem (from rem) as the word was a monosyllable (see § 97, 4). 2. T h e ending of the singular fell when preceded by a short /, n, r, s, or c, or by the group t + j (§ 46, 2): solem > sol; canem > cam (old); seniorem > senhor; mare > mar; mensem > mes; vicem > vez; duritiem > durez. B u t the e of the ending of the plural of these nouns did not fall (§ 46, 3): soles > sois; canes > caes; seniores > senhores; menses > meses; vices > vezes.

124

MORPHOLOGY

[§124

3. Nouns with stems ending in a short I lost this I in the plural because it was intervocalic; and the ending generally became - j s because of the hiatus (see § 46, 10) : soles > sóis; *animàles > animais. A. This / was not lost in dialectal Portuguese if the plural was formed by adding s to singular forms which had a paragogic e: animal > animale (see § 116, 2) and animale s > animates. See Esquisse, § 70 a. See also M-L, Gram, II, § 53. B. T h è form soles is found in Old Portuguese (see Fab, 106). The noun mal came from the adverb male. There was a plural maes in Old Portuguese (see R L , X V I , 104 and 107; Abraham, Glossary) and Old Galician (Crest, Glossary). But males was much commoner and is the form which has survived. The plural consules (from consules) is learned. c. Final e did not fall when preceded by long I: vallem > vale; pellem > pele. In dialectal Portuguese, however, it sometimes fell with consequent shortening of the consonant (Esquisse, § 67). Analogical plural forms were then made by adding -es to the new singulars and the / now short and intervocalic fell: vallem > valle > vai and vai -+- es > vaes; pellem > pelle > pel and pel + es > pets. See Esquisse, § 70 a ; R L , II, 27; R L , X I , 280; R L , X X V I I , 128. 4. Nouns whose stem ended in short n lost this n after it had nasalized the preceding vowel (see § 98, 4 ) : canem > cam ( o l d d r a c ò n e m > dragom ( o l d ) ; f i n e m > Jim; hominem > omen > homem. In the plural the nasal resonance spread to the ending and the t w o vowels either formed a nasal diphthong or contracted: canes > caes (§ 78, 3); dracbnes > dragdes (§ 78, 3); fines > fins (§ 78, 2 B) ; homines > omees > homens (§ 78, 2 c). A. For the later development of cam to cào and dragom to dragào, see § 157, 2. For the loss of the final nasal resonance in the noun pente (from pectinem), see § 110, 3. B. It is evident that in Old Portuguese the singular of nouns ending in unaccented -en or -em (from -inem) was correctly spelled with one e, the plural with two i ' s ; for example, in the

§124]

NOUNS

125

"Vida de S. Nicalau" (Bausteine, 581-586), all the singular forms of homem (three) have one e, all the plural forms (four) have two *-oen > -oe or de; its plural developed as follows: *-udines > -oees > -des. T h e singular -oe then became -om (and later -ao) by analogy with the ending coming f r o m L a t . -onem (see dragom above), as both types were t h e same in the plural. 5. M a n y Classical L a t i n neuter a c c u s a t i v e s did n o t end in e but in the s t e m - e n d i n g I or r; t h e y a d o p t e d in Vulgar Latin the masculine e n d i n g -e{m): mel > *melem > mel; uber > *uberem > ubere; marmor > marmorem > marmore; sulfur > sulfurem > enxufere (old a n d popular) and enxofre. See G V L , § 347. A. T h a t *melem had the short I of the original Classical Latin accusative and not the long I of the other oblique cases is shown by the plural meis (from *meles). B. For the final e of ubere, etc., see § 46, 2 B. c. CI. L. robur became V. L. roborem (GVL, § 347) b u t P t g . roble is doubtless a late borrowing f r o m Spanish. T h e medial group bl never survived in Portuguese words; even in borrowings from Spanish it generally changed to br (§ 86, 2 A). 6. N e u t e r a c c u s a t i v e s e n d i n g in -en did n o t a d o p t the e n d i n g of the m a s c u l i n e a c c u s a t i v e , b e c a u s e -en was t a k e n to be the e q u i v a l e n t of -em. T h e n final n fell, j u s t as final m had fallen at an earlier d a t e : nomen > nome; lumen > lume; culmen > cume; legumen > legume; sanguen > sangue; acumen > game; aeramen > arame (§ 97, 4 ) . N e w plurals were accordingly formed by adding s to the singular as in masculine nouns, e.g., nomes. See G V L , § 369A. These nouns generally adopted the masculine ending in the Vulgar Latin of the Spanish t e r r i t o r y : nomen > *nominem > Sp. nombre. I n a s m u c h as t h e ending -inem did not spread in the Vulgar Latin of the Portuguese territory, it is not likely

126

MORPHOLOGY

[§ 124

t h a t OPtg. vermem came from a V. L. *verminem nor that OPtg. vimem came from a V. L. *viminem. The spelling vimee (FM, II, Glossary) is of no significance as unaccented single vowels were often written double in this document (FM, I, xxv). Nor is it likely t h a t Ptg. sangue came from the Latin masculine form sanguinem through an OPtg. * sangue. See Comp, 116. For vermem, see § 96, 2 and for vimem, see § 77 B. B. T h e suffix -úmen belongs to this group. It became -ume in Portuguese and through a V. L. *-üminem, -umbre in Spanish. In many words it replaced the suffix -udinem, e.g., consuetudinem > Ptg. costume and Sp. costumbre (Manual, § 83, 4). Cf. D'Ovidio, 25, n. 3. 7. A few n e u t e r a c c u s a t i v e singulars e n d i n g in -us were t a k e n for m a s c u l i n e a c c u s a t i v e plurals of t h e second declension a n d a n e w s i n g u l a r was f o r m e d b y d r o p p i n g t h e s: tempüs > tempos; stercüs > estercos; corpus > corpos; pectus > peitos; latus > lados; plgnüs > empenhos (with prefix em-). T h e n e w singulars a r e tempo, estéreo, corpo, etc. A, M e t a p h o n y took place in the new singular corpo. B. T h e neuter caput became cabo and adopted the plural ending of nouns coming from the Classical Latin second declension. 8. T h e L a t i n a c c u s a t i v e -icem of p r o p a r o x y t o n i c n o u n s a n d a d j e c t i v e s b e c a m e -ez, which is now spelled -es: aurificem > ourivez > ourives; simplicem > simprez > simples. See § 98, 2 B. T h e p l u r a l of these words was s h o r t e n e d b y h a p l o l o g y (§ 114): aurifices > ourivezes > ourives. A. Haplology occurred in paroxytonic plurals in dialectal Portuguese: vices > vez es > bes (§ 114 B); felices > felites > feliz (RL, IV, 220). B. A new popular singular ourive has been formed by dropping the final s of the shortened plural ourives. Some plurals, erroneously taken for singulars, have been pluralized again in popular Portuguese by the addition of -es: pedes > pés > peses (Esquisse, § 70 a; RL, VII, 45; and RL, X I I , 309). Such plurals are common in Modern Galician: reges > reis > reises; leges > leis > leises.

§124]

NOUNS

127

9. A f e w nouns of t h e fifth d e c l e n s i o n a d o p t e d in V u l g a r L a t i n t h e endings derived f r o m t h e Classical L a t i n first d e c l e n s i o n : rabiem > raiva; dies > dias. A. Some nouns belonged to both declensions in Classical L a t i n , e.g., materiam and materiem. 10. S o m e nouns of t h e third (and fifth) d e c l e n s i o n a d o p t e d t h e endings derived f r o m t h e first d e c l e n s i o n or t h o s e derived from t h e s e c o n d d e c l e n s i o n : axem > eixo; aclem > a(o; cucumerem > cogombro; passerem > pdssaro; gruem > *gruam > grua (§ 38, 9 ) ; gruem > *gruum > grou (§ 38, 10); os > bssum ( G V L , § 356, 3) > osso; vas > vasum ( G V L , § 356, 3) > vaso; gramen > grama; inguen > ingua; septembrem > setembro; octubrem ( C a r n o y , 64) > outubro; novembrem > novembro; decembrem > dezembro. A. T h e shift of declension in setembro, outubro, etc. was occasioned by analogy with the names of most of the o t h e r months, Janeiro, fevereiro, etc. B. T h e plural daemones became demoes in Old P o r t u g u e s e b u t this form later adopted the plural ending of nouns derived f r o m the second declension and became demos. T h i s shift was facilit a t e d by the fact t h a t the singular was demo (from the n o m i n a t i v e daemon) (§ 121). 11. S o m e m a s c u l i n e nouns a n d s o m e n o u n s of c o m m o n gender formed a distinguishing feminine modeled on nouns of t h e Classical L a t i n first d e c l e n s i o n : leonem > leom > leao, f e m . leoa; pavonem > pavom > pavao, fem. pavoa; seniorem > senhor, f e m . senhora; infantem > infante, fem. infanta; parentem > parente, f e m . parenta; hospitem > h6spede, f e m . hospeda. A. T h e formation of these feminines began sporadically very e a r l y : senhora (Cd'A, I, 368-369; R L , X X I I I , 85; F M , I, xxx); parenta (RL, X X I I I , 64); ospeda ( A H P , I I I , 18, A . D . 1325). But it did not become general until t h e sixteenth c e n t u r y . T h e feminines chorona and ladrona were obviously f o r m e d m u c h later t h a n leoa and pavoa.

MORPHOLOGY

128

[§ 125-126

ADJECTIVES 125.

DECLENSIONS.

T h e declensions of adjectives had the same fate in Portuguese as t h e declensions of nouns. All cases except the accusative disappeared and as there were no longer any neuter nouns to modify, the neuter endings fell into disuse. T h e neuter singular, identical with the masculine singular, has survived with the force of an abstract noun. 126.

FIRST AND

SECOND

DECLENSIONS.

1. T h e endings of the masculine and feminine accusative singular and plural of adjectives of the Classical Latin first and second declensions and the endings derived f r o m t h e m in Portuguese are: Mase. Sg. Masc. PI. Fem. Sg. Fern. PI.

Classical Latin -üm -os

Portuguese -o (§ 12, 8) -os

-am

-a (§ 12, 8)

-as

-as

2. These endings and a preceding a in hiatus became -ao, -aos, -a, and -as: sanum > sao; sanos > saos; vanam > va; vanas > vas (§ 78, 2 and 3). 3. In hiatus with a preceding tonic a or £ the masculine endings became -u and -us (§ 48, 2): malum > mau; maids > maus; reum > r'eu (§ 34, 7 B) ; reos > reus; t h e y contracted with tonic u and became -ú and -ús (§ 48, 5): nudum. > nú; nudos > nús; they contracted with tonic o and became -om and -ons (§ 78, 2): bonum > bom; bonds > bons. 4. In hiatus with a preceding tonic a the feminine endings contracted and became -á and -ás: ma'iam > má; malas > más; with tonic ( and o assimilation took place: ream > ree > r'e (§ 34, 6 A); solam > soa > só (§ 99, 3 A).

§ 126]

ADJECTIVES

129

A. T h e masculine f o r m s of so had close o at first (see r i m e : soo : nego-o in CA, N o . 416). B u t t h e masculine plural b e c a m e sos b y analogy with a d j e c t i v e s of the novo t y p e and t h e m a s c u line singular b e c a m e so, p e r h a p s in order to avoid confusion with the v e r b f o r m sou. T h e early feminine f o r m s soa and soas still survive in certain dialects ( R L , I I I , 60 and 325). B. Assimilation was p r e v e n t e d in boa (from bonam) b y t h e nasal resonance (§ 45, 2 B). 5. In s o m e a d j e c t i v e s an a n a l o g i c a l f e m i n i n e h a s r e p l a c e d t h e p h o n o l o g i c a l f o r m , e.g., parva h a s r e p l a c e d parvoa ( f r o m parvulam) as t h e f e m i n i n e of parvo ( f r o m parvulum)\ antiga h a s r e p l a c e d *antigua ( f r o m antiquam, § 93, 2) as t h e f e m i n i n e of antigo ( f r o m antiquum, § 93, 3 ) ; sadia has r e p l a c e d *sadiva ( f r o m sanativam) as t h e f e m i n i n e of sadio ( f r o m sanativum, § 72, 2 A). 6. A f e w a d j e c t i v e s of t h e first a n d s e c o n d d e c l e n s i o n a d o p t e d t h e e n d i n g s d e r i v e d f r o m t h e third d e c l e n s i o n : liberum > litre; contentum > contente; firmum > firme. 7. I n a d j e c t i v e s w i t h Classical L a t i n radical o m e t a p h o n y g e n e r a l l y t o o k p l a c e in t h e m a s c u l i n e singular ( § 1 0 0 , 7 ) : novum > novo; grossum > grosso; b u t n o t in a n y of t h e o t h e r f o r m s : ngvos, ngva, ngvas; grgssos, grgssa, grgssas. S e e § 123, 6. A. For dialectal variations, see references in § 123, 6 H. B. T h e v of novum fell in Vulgar L a t i n , b u t it was soon restored b y a n a l o g y with t h e feminine f o r m s and t h e masculine p l u r a l ( G V L , § 324). 8. M e t a p h o n y or t h e a n a l o g i c a l s i m u l a t i o n of m e t a p h o n y d i d n o t g e n e r a l l y t a k e p l a c e in a n y of t h e f o r m s of a d j e c t i v e s w i t h C l a s s i c a l L a t i n o ( u ) : ruptum rota, fosco;

rotas;

totum

a n d foscos,

> todo; fosca,

a n d todos,

toda,

> roto; a n d tgdas; fuscum

rotos, >

fgscas.

A. M e t a p h o n y t o o k place in t h e n o u n rgta ( f r o m ruptam, N a s c e n t e s ) , which was not affected b y a n a l o g y with f o r m s of t h e a d j e c t i v e . It likewise t o o k place in t h e indefinite p r o n o u n tudo

130

MORPHOLOGY

[§126

(from the neuter totum), which was not affected by analogy with forms of the adjective. B. M e t a p h o n y took place in the feminine singular and plural of the adjective ending -oso, e.g., formdsam > formgsa; formosas > formgsas. T h e masculine singular, instead of becoming -*uso, remained unchanged by analogy with adjectives of the novo t y p e and the masculine plural remained unchanged also, as final -os was neutral with respect to t h e radical vowel (see § 123, 6). T h u s the inflection of formoso toward the end of the sixteenth century was formoso, formosos, formgsa, formgsas. These forms are given by Nunes de Liao in 1576 ( N - L , Orth, 17 r o ) . But there is evidence that forwpsos had become formgsos even earlier (Oliv, Chapters 8 and 1 8 ) . And thus analogy with the novo type became complete. 9. In s o m e a d j e c t i v e s w i t h Classical Latin radical e met a p h o n y took place in t h e masculine singular and plural, as in masculine nouns w i t h radical e (§ 123, 7): adversum > avesso; adversos > avessos; graecum > grego; graecos > gregos; laetum > ledo; laetos > ledos; and the e of the feminine singular and plural b e c a m e close b y a n a l o g y with t h e m a s c u l i n e singular and plural: adversam > avessa; adversas > avessas; graecam > grega; graecas > gregas; laetam > leda; laetas > ledas. In o t h e r a d j e c t i v e s w i t h Classical Latin radical e, m e t a p h o n y took place in the f e m i n i n e singular and plural, t h a t is, final a k e p t t h e radical v o w e l o p e n : bellam > b(la; bellas > b(las; caecam > C(ga; caecas > c(gas; feram > f(ra;feras > feras; and the e of the masculine singular and plural remained o p e n b y a n a l o g y w i t h the feminine singular and plural: bellum > b(lo; bellos > b$los; caecum > cggo; caecos > c£gos; ferum > /¿ro; feros > fgros. A. When a n y of these forms occurred as other parts of speech t h e y were not affected by analogy, e.g., avessas (fem. pi. noun); *mure(m)-caecum > morcego. 10. A d j e c t i v e s w i t h radical e (i) were not affected b y m e t a p h o n y ; their f e m i n i n e forms developed b y a n a l o g y w i t h their m a s c u l i n e forms. T h e y , therefore, have e

§ 126-127]

131

ADJECTIVES

t h r o u g h o u t like t h e avesso t y p e : quietum > quedo a n d quedos, queda, quedas; secretum > segredo a n d segredos, segreda, segredas; siccum > seco a n d secos, seca, secas. A. From this section and the above section it is clear that there is no alternation in the radical e of any adjectives. 127.

THIRD

DECLENSION.

1. T h e endings of t h e (masculine and feminine) L a t i n t h i r d declension a n d in P o r t u g u e s e a r e : Classical

a c c u s a t i v e singular a n d p l u r a l of a d j e c t i v e s of t h e Classical the endings derived from them Latin

Portuguese

Sg. -em -(e) (§ 12, 8) PI. -es -es or -is 2. T h e singular e n d i n g p r e c e d e d b y a s h o r t I, n, r, s or c fell: *hispanidnem ( M a n u a l , § 66, 2) > espankol; commúnem > comum; meliorem > melhor; *cortensem ( G V L , § 39) > cortes; simplicem > simprez (old). A. For the development of comuns, the plural of comum, see §38, 4 A . 3. A d j e c t i v e s w i t h s t e m s e n d i n g in a s h o r t I lost this I in t h e p l u r a l because it was i n t e r v o c a l i c a n d t h e e n d i n g g e n e r a l l y b e c a m e -is b e c a u s e of h i a t u s (see § 4 6 , 10): personales > pessoais; possibiles > possíveis. A. The plurals of the learned adjectives fácil and útil have popular endings: facéis and úteis. 4. S o m e a d j e c t i v e s of t h e t h i r d declension a d o p t e d t h e e n d i n g s derived f r o m t h e Classical L a t i n first a n d second d e c l e n s i o n : acrern > acrum ( A p P r ) > agro a n d agra; veterem > vedro a n d vedra (old); rüdem > rudo a n d ruda (old a n d d i a l e c t a l ) ; latro (nom. sg.) > ladro a n d f r o m this, ladra was formed. A. In the fusion of the endings -ao, -am and -om, gra, the apocopated form of grande, became grao (§ 157, 1 A). This form was taken to be a masculine adjective derived from the Classical

MORPHOLOGY

132

[ § 127-128

Latin first and second declension and formed in the manner of sao and vao (§ 126, 2) while gra, t h e original apocopated form, was taken to be its feminine and formed in the manner of sa and va (§ 126, 2). These two forms have survived in compound nouns, e.g., grao-mestre and gra-cruz. 5. Some adjectives of the third declension, chiefly adjectives of nationality and adjectives ending in -or (from -orem), formed a distinguishing feminine m o d e l e d o n the feminine of adjectives of the Classical Latin first and second declension: *hispanionem ( M a n u a l , § 6 6 , 2) > espanhol, fem. espanhola; incantatdrem > encantador, fem. encantadora; *anglensem > ingles ( N a s c e n t e s ) , fem. inglesa. A. These forms are first found in Vulgar Latin, e.g., -ensam as the feminine of -ensem (M-L, Gram, II, § 6 0 ; Carnoy, 243). But m a n y were formed very late, e.g., espanhola, which was formed after the period in which intervocalic / fell; cf. the masculine plural espanhois. See Comp, 232, n. 2. B. T h e only adjectives ending in -or which did not form a separate feminine by the addition of a were comparatives: meliorem > melhor (masc. and fem.). c. T h e adjective comum formed a feminine comua, which was used as late as the eighteenth century but has survived today only as a noun. 128.

COMPARISON.

1. T h e substitution of c o m p a r a t i v e and superlative forms b y an analytical m e t h o d of comparison, which had b e g u n in Vulgar Latin (see § 13, 2 c), was c o m p l e t e l y accomplished in Portuguese. T h e o n l y forms t h a t have survived are the following irregular c o m p a r a t i v e s : M . & F. Acc.

Classical Latin meliorem pejorem majorem mindrem Neut. Acc. minus plus

Portuguese melhor peior (old) and pior (§ 73, 5 A) maor (old), mor, and maior (§ 73, SA) meor (old) and menor (§ 78, 7 B) meos (old) and menos (§ 78, 7 B) chus (old)

NUMERALS

§ 128-129]

133

The tonic o of these adjectives was close in Old Portuguese, cf. the rime maor : amor in CV, No. 697 and see R F , X X V , 650. It became open first in mór (§ 99, 3 A) and through the influence of mór spread to melhgr, pigr, maigr and mengr. But it did not spread to the learned forms superior, inferior, etc. A. It is thought that modern dialectal maor is not a survival from Old Portuguese but rather a new development from maior (Opúsculos, II, 195). B. T h e adverbs melhor and fior probably came from neuter ablative forms (§ 121 E). c. T h e comparative seniórem has survived only in the forms of the noun senhor. And junior has survived only as a learned word. D. There is a form maire in Old Portuguese which may come from the nominative major (see Rom, X I , 81).

2. The superlatives óptimo, máximo, etc. and those ending in -issimo are learned and derived directly from Latin forms. But semi-learned forms are found in dialectal Portuguese, e.g., -issemo and -essíssemo (RL, X X V I I , 132) and -ismo (Gal). Sometimes a superlative is made by attaching the learned suffix to a popular form, e.g., docissimo for the learned dulcissimo. NUMERALS 129.

ONE

TO

THREE.

1. The first three cardinal numerals were declined Classical Latin but only the first two are declined Portuguese. 2. Unum > úu > um (§ 48, 5 A; § 78, 2 B); ünarn > > urna (§ 78, 4 D). 3. Düos > dous (§ 38, 10) and dois (oi for ou, § 92, 7 düas > duas (§ 38, 9). 4. Tres > tres. Neuter tria disappeared.

in in üa c);

MORPHOLOGY

134 130.

F O U R TO

[§ 130-131

TEN.

1. Quattuor > quattor (§ 93, 1) > quatro (§ 69, 1; § 97, 5). 2. Qulnque > cinque (Carnoy, 214) > cinco. T h e r e were two u's in this word in Classical L a t i n ; the first of t h e m fell b y dissimilation in Vulgar Latin (GVL, § 379); cf. cerquinho (from *querquinum). Final o developed through t h e influence of quatro; b u t see Lexique, 93-94. 3. Sex > seis (§ 97, 6 B). 4. Septem > s(te (§ 85, 4). 5. Odd > oito (§ 37, 3). 6. Novem > ngve. 7. Decern > dgz ( § 4 6 , 2; § 9 8 , 2). 131.

E L E V E N TO

SIXTEEN.

1. T h e posttonic p e n u l t seems to have fallen in these six numerals, c o n t r a r y to § 53. T h i s probably took place in t h e following m a n n e r : tredecim and sedecim became dissyllabic t h r o u g h contraction a f t e r the fall of intervocalic d\ then the remaining numerals of the group shortened in imitation of treze and seze and through the influence of the Spanish forms, which m u s t have been considerable in view of the i m p o r t a n c e of numerals in commerce. Final e was preserved in doze, treze and seze, contrary to § 46, 2. T h i s was due to t h e influence of onze, catorze and quinze, in which z a f t e r syncope was preceded by a cons o n a n t . T h e e m a y have been preserved also in order to avoid confusion with dous, tres and seis respectively, somew h a t similarly pronounced, inasmuch as the slightest cause of confusion in numerals is always carefully avoided because of their importance in exchange. 2. Undecim > *undece (GVL, § 166) > onze. 3. Dubdecim > dbdece (GVL, § 225) > doze. 4. Tredecim > tredece > *treeze (§ 74, 1) > treze. T o n i c £ became e t h r o u g h the influence of tres.

§ 131-132]

NUMERALS

135

5. Quattuordecim > *quattordece (§ 93, 1) > catorze (§ 69, 2; 92, 6. Quindecim > quindece > quinze. 7. Sedecim > sedece > *seeze (§ 74, 1) > jrz