No CrossRef data available.
Published online by Cambridge University Press: 22 April 2026
[Iguaria, recorded since the 14th century, is traced back to Late Latin iequāria ‘giblets’ and thus connected with the iecur family, reputedly extinct in Romance, where hēpar and *fīc
tum replaced the traditional Latin word for ‘liver’. In conjunction with the proposed etymology, new theories and data are presented on controversial points of Portuguese historical grammar, including the development of initial yod and of pretonic e, the transmission of the suffix -āriu and the stress shift. For a summary of the results, see §5.]
1 On the use of originally abstract terms like dignitās and celebritās to express the idea of ‘tidbit’ in Romance, see W. Meyer-Lübke, Cillérveda, RFE 7.368–9.
2 A. R. Gonçalves Viana, Apostilas aos Dicionários Portugueses, 2 vols. (Lisbon, 1906); see also J. P. Machado, Acrescentos de Gonçalves Viana às suas Apostilas; outras notas a propósito, BF, from vol. 7 (1941).
3 C. Michaëlis de Vasconcelos, Etymologies espagnoles, Rom. 2.86–91; Etymologisches, JREL NF 1.202–17, 308–27; 3.57–64; Studien zur romanischen Wortschöpfung (Leipzig, 1876); Portugiesische Etymologien, ZRPh 7.102–15; Studien zur hispanischen Wortdeutung, Miscellanea Caix-Canello 113–66 (1885); Etimologias portuguesas, RL 1.117–32, 298–305; Fragmentos etimológicos, RL 3.129–90; Contribuïções para o futuro dicionário das línguas hispânicas, RL 11.1–62; Estudos etimológicos: contribuïções para o futuro dicionário etimológico das línguas românicas peninsulares, RL 13.222–432; Etymologische Einfälle, RL 20.316–9; Glossário do Cancioneiro da Ajuda, RL 23.1–95; Miscelas etimológicas, Hom. R. Men. Pidal 3.441–73.
4 R. Menéndez Pidal, Etimologías españolas, Rom. 29.334–79; Notas para el léxico románico, RFE 7.1–36.
5 V. García de Diego, Dialectalismos, RFE 3.301–18; Divergentes latinos, RFE 5.133–42; Etimologías españolas, RFE 6.113–31, 7.113–49; Notas filológicas, RFE 11.333–50; Miscelánea filológica, RFE 12.1–15; Etimología idealista, RFE 15.225–43; Notas léxicas, RFE 15.337–42; Notas etimológicas, RFE 18.1–14, 20.353–62; Miscelánea etimológica, BAE 7.249–63; Contribución al diccionario hispánico etimológico (Madrid, 1923; RFE Anejo 2).
6 A. Castro, Adiciones hispánicas al diccionario de Meyer-Lübke, RFE 5.21–42, 6.337–45, and numerous notes in RFE.
7 O. J. Tallgren, Glanures catalanes et hispano-romanes, see index of Portuguese etymologies NM 16.104.
8 L. Spitzer, Lexikalisches aus dem Katalanischen und den übrigen iberoromanischen Sprachen (Geneva, 1921; Bibl. dell'Arch. Rom. II-1); Notas etimológicas, RFE 10.167–72, 373–9; Notes étymologiques, RFE 12.229–51, 14.243–55, 16.148–53, 18.235–8, 20.169–72; Notes linguistiques, RFE 24.24–36; Estudios etimológicos, Anales del Instituto de Lingüística 1.30–70 (1941), and numerous notes in various journals.
9 F. A. Coelho, Sôbre a Língua Portuguesa, in Domingos Vieira, Grande Dicionário Português, vol. 1, pp. ix-ccvii (Porto, 1871). The same writer's A Língua Portuguesa: Noções de Glotologia Geral e Especial Portuguesa (Porto, ca. 1881) was not available for consultation.
10 J. Cornu, Die Portugiesische Sprache, Gr. Grundriss2 1.916–1037 (Strassburg, 1904–6).
11 J. J. Nunes, Compêndio de Gramática Histórica Portuguesa2 (Lisbon, 1930).
12 J. Leite de Vasconcelos, Lições de Filologia Portuguesa2 (Lisbon, 1926).
13 J. Huber, Altportugiesisches Elementarbuch (Heidelberg, 1933).
14 E. B. Williams, From Latin to Portuguese (Philadelphia, 1938).
15 R. de Sá Nogueira, Curso de filologia portuguesa, LP 1.86–92, 113–21, 158–62, 188–96, 232–40, 287–97, 327–31, 377–84; 2.33–8, 77–80, 117–9, 154–8, 428–33; 3.75–81, 146–8. A. G. Ribeiro de Vasconcelos, Gramática Histórica da Lingua Portuguesa (Lisbon, 1901) was not available for consultation.
16 A. A. Cortesão, Subsídio para um Dicionário Completo (Histórico-Etimológico) da Língua Portuguesa (Coimbra, 1900–1).
17 None of the four printings of F. A. Coelho's antiquated Dicionário Manual Etimológico da Lingua Portuguesa was available for consultation.
18 See S. R. Dalgado, Glossário Luso-Asiático, 2 vols. (Coimbra, 1919–21). The Portuguese word did not penetrate into any of the Asiatic languages, either, on the authority of S. R. Dalgado, Influência do Vocabulário Português em Línguas Asiáticas (Coimbra, 1913).
19 See Dozy-Engelmann, Glossaire des mots espagnols et portugais dérivés de l'arabe (Leyden, 1869); A. A. Fokker, Quelques mots espagnols et portugais d'origine orientale, ZRPh 34.560–8, 38.481–5; D. Lopes, Alguns vocábulos arábico-portugueses, Misc. C. Mich. Vasc. 23–34; A. Steiger, Contribución a la fonética del hispano-árabe y de los arabismos en el ibero-románico y el siciliano (Madrid, 1932; RFE Anejo 17); J. P. Machado, Alguns vocábulos de origem árabe, BF 6.1–33; idem, Comentários a alguns arabismos do Dicionário de Nascentes, BF 6.225–328; M. L. Wagner, Sobre alguns arabismos do português, Biblos 10.427–53; idem, Etimologías españolas y arábigo-hispánicas, RFE 21.225–47.
20 See the reviews of DELP by F. Rebêlo Gonçalves, BF 2.282–7; by R. de Sá Nogueira, LP 3.236–8; by I. Xavier Fernandes, LP 5.113–9, 231–8; also Eduardo de Lisboa (pseud.), O Dicionário do Sr. Nascentes e o REW: Rectificações (Rio de Janeiro, 1937). REW3 has been reviewed by M. Rodrigues Lapa, LP 1.392–3, and addenda have been published by J. M. Piel in Biblos 8.379–92, 9.244–62, 10.124–40; see also the same scholar's Miscelanea Viscentina: notas lexicográficas e etimológicas, Biblos 14.39–69. No reference to iguaria is made in the fragments of a new etymological dictionary of Portuguese by A. Magne, RFH 1.231–42, 336–69, 502–24, 546–7; the same writer's Apostilas etimológicas, RLP 19, 21–7, 47, 55–7 were not available for consultation.
21 The word occurs in Cod. Harleianus 5792, known at some time to have been in the hands of a French owner. Authorship and country of origin do not seem to have been ascertained. See Götz-Gundermann, Glossae latinograecae et graecolatinae 2.351 (Leipzig, 1888). Κοιλιήπατα for κοιλήπατα is an emendation by Heraeus.
22 See Du Cange 4.58c: 'Gigeria forte zizeria'; ibidem, 8.432b: 'Zizerium pro gigerium, quo nomine Latini appellant gallinarum intestina'. On gigerium, which survives in Fr. gésier, see REW3 §3760.
23 To judge by J. Cuveiro Piñol, Diccionario gallego (Barcelona, 1876); M. Lugris Freire, Gramática do idioma galego 135–52 (La Coruña, 1922; the 2nd ed. [1931] was not accessible); A. Cotarelo y Valledor, El castellano en Galicia, BAE 14.82–136; L. Carré Alvarellos, Diccionario gallego-castellán2 (La Coruña, 1933). The available fascicles of (R.) Academia Gallega, Diccionario gallego-castellano (La Coruña, 1913–28) do not seem to extend beyond the letter C; M. Valladares Núñez, Diccionario gallego-castellano (Santiago, 1884) was not accessible. From V. García de Diego, Elementos de gramática histórica gallega 190 (Burgos, [1909]) it follows that in Modern Galician two (regionally differentiated?) suffixes, -aria and -eiria, correspond to Ptg. -aria.
24 V. García Rey, Vocabulario del Bierzo (Madrid, 1934); F. Krüger, El dialecto de San Ciprián de Sanabria 119–28 (Madrid, 1923; RFE Anejo 4); idem, Vocablos y cosas de Sanabria, RFE 10.153–66; idem, Mezcla de dialectos, Homen. R. Men. Pidal 2.121–66; idem, Die Gegenstandskultur Sanabrias und seiner Nachbargebiete 295–322 (Hamburg, 1925); J. Leite de Vasconcelos, Estudos de Filologia Mirandesa 2.145–225 (Lisbon, 1901); J. de Lamano y Beneite, El dialecto vulgar salmantino (Salamanca, 1915); B. Acevedo y Huelves and M. Fernández y Fernández, Vocabulario del bable de occidente (Madrid, 1932); R. Menéndez Pidal, El dialecto leonés, RABM, 3rd set, 14.128–72, 294–311. S. A. Garrote, El dialecto vulgar leonés (Astorga, 1909) could not be consulted.
25 Å. W. Munthe, Anteckningar om folkmålet i en trakt af Vestra Asturien 57–92 (Upsala, 1887); A. de Rato y Hevia, Vocabulario de las palabras y frases bables que se hablaron antiguamente (Madrid, 1891); P. de Mugica, Dialectos castellanos: montañés, vizcaíno, aragonés (Berlin, 1892); E. de Huidobro, Palabras, giros y bellezas del lenguaje popular de la Montaña (Santander, 1907); G. A. García Lomas y García Lomas, Estudio del dialecto popular montañés (San Sebastian, 1922).
26 F. Krüger, Studien zur Lautgeschichte westspanischer Mundarten (Hamburg, 1914).
27 J. Alemany, Voces extremeñas recogidas del habla vulgar de Albuquerque y su comarca por Don Aurelio Cabrero, BAE 3.653–66, 4.84–100.
28 On the homogeneous character of that area, see R. Menéndez Pidal, Sufijos átonos en español, Bausteine zur romanischen Philologie (Festgabe A. Mussafia) 386–400 (Halle, 1905); F. Krüger, Die nord westiberische Volkskultur, WS 10.45–137. The former provides linguistic, the latter ethnological evidence.
29 On the use of comeres, see Menéndez Pidal, Cantar de Mio Cid 582–3; cf. Libro de Buen Amor, lines 303a, 902a.
30 See Menéndez Pidal, op. cit. 588.
31 Quotations from Cantar de Mio Cid, Poema de Alexandre, Libro de Buen Amor, Crónica de 1344, and Guillelme will be found in Menéndez Pidal, op. cit. 604–5.
32 Manjar (correctly listed as a Gallicism by J. B. de Forest, RR 7.400) is absent from the Cantar. Early examples include Duelo, line 35d; Libro de Buen Amor, lines 176b, 483d, 502a, 815d, 944d, 1127d; Rimado de Palacio (MS N) lines 104a, 108c, 480c, 1624c.
33 See Menéndez Pidal, op.cit. 784.
34 See Menéndez Pidal, op.cit. 585; add Libro de Buen Amor, lines 85a, 86d, 113d, 176a, 293a, 568a.
35 See Menéndez Pidal, op.cit. 715; add Libro de Buen Amor, lines 292b, 770c, 1083d, 1266b, 1315c, 1372d (according to H. B. Richardson, An Etymological Vocabulary to the Libro de Buen Amor [New Haven, 1930]).
36 In Old Spanish, golosina signified ‘gluttony’ exclusively; see Libro de Buen Amor, lines 291a, 295a, 297a, 1168a. It remains to be seen whether it developed its secondary connotation on the pattern of Fr. gourmandise.
37 See, for instance, the recently published dictionaries of A. Moreno (1938), J. A. Ferreira (1939), A. Franco, 3rd ed. (1940), Lima-Barroso, 3rd ed. of the revised version (1942).
38 See A. Cândido de Figueiredo, Novo Dicionário da Língua Portuguesa4 (s.d.); the 5th ed., now in course of publication, could not be consulted. Figueiredo's source of information was presumably RL 16.246 (Oscar de Pratt, Notas à margem do Novo Dicionário da Língua Portuguesa, RL 16.206–79, 17.338–48, 18.65–162). C. Basto, Nótulas ao Novo Dicionário, 4 vols. (Viana-do-Castelo and Elvas, 1913–6) was inaccessible.
39 See C. Teschauer, Novo dicionário nacional2 (Porto Alegre, 1928). Numerous dialectal word-lists were checked, including F. Braga Barreiros, Vocabulário barrosão, RL 20.137–61, 35.239–97; F. Santos Serra Frazão, Sucinto vocabulário coligido numa aldeia da Serra de Albardos, RL 36.73–167; J. A. Pombinho Jr., Vocabulário alentejano, [letter E] RL 34.266–90, 35.155–60, 36.197–217.
40 Many of the older dictionaries were unfortunately out of reach, including those of J. Cardoso (1570), A. Barbosa (1611), B. Pereira (1647), A. de Morais Silva (1789) and J. I. Roquete – J. da Fonseca (1850). Fr. J. de Santa Rosa de Viterbo, Elucidário (Lisbon, 1798) contains no information.
41 R. Bluteau, Vocabulário Português e Latino (Coimbra-Lisbon, 1712–28).
42 Frei Domingos Vieira, Grande Dicionário Português; Tesouro da Lingua Portuguesa 3.1015 (Porto, 1873).
43 Pois o Turco Gran-Soldão Não come tanta iguaria (Farças) ‘Since the Turkish Grand-sultan does not eat quite so much choice food’.
44 Tão frescos que os estimou elle por a milhor iguaria do mundo (Décadas) ‘So fresh that he thought they were the best food in the world’; Inda que a cea nao era mui abastada de iguarias (Clarimundo) ‘even though few dishes were served at supper’.
45 Nos devemos abster ... das iguarias delicadas (Compêndio de Espiritual Doutrina) ‘We must abstain from delicate food’.
46 Que as iguarias que dava Vinhão nos pratos pintados (Redondilhas) ‘For the tidbits that he offered were just painted in the plates’.
47 Só d'elles os tres levavam iguaria (Carta de Guia de Casados) ‘Only three out of them (viz. of the plates) carried any food’; Mandava levar as iguarias a sua casa (ibid.) ‘He ordered the dishes to be brought to his home’.
48 Entre as iguarias mais selectas usavam de carne dos bogios (Portugal Médico) ‘Among their choicest dishes they used the meat of monkeys’; Na cosinha onde se preparavão as iguarias (ibidem) ‘In the kitchen where the food was being prepared’.
49 Depois lhe estenderaõ diante hũa alcatifa, e sobrela hũas toalhas ... e sobrelas muytas e diversas igorias em bategas de prata (s.v. bategada) ‘They then stretched out a carpet before him, and [laid] some napkins upon it ... and placed numerous and varied foods on them in golden dishes’. The form igoria is a vulgarism comparable to quorenta ‘forty’ beside quarenta, quoresma ‘Lent’ beside quaresma, common in Old Portuguese.
50 Tem o ventre recheado de granos de ova de que fazem iguarias (s.v. balancás) ‘His stomach is replete with hard roe, out of which they make tasty foods’.
51 Daim, iguaria ordinaria destes paisanos, he leyte coalhado (s.v. daim) ‘Daim, the ordinary food of these people, is coagulated milk’.
52 Marco Paulo: O Livro de Marco Paulo, o Livro de Nicolao Veneto, Carta de Jeronimo de Santo Estevam, ed. F. M. Esteves Pereira (Lisbon, 1922). Most of the quotations in the body of the text are from the Livro de Nicolao Veneto, translated from Latin by Valentim Fernandes, not a native speaker of Portuguese, but a Moravian printer known to have come to Portugal prior to 1490 and to have died in 1519; the original draft of the translation may have been revised (see Introduction, p. xvi). Opinions are divided on the authorship of the translation of the Livro de Marco Paulo (from the Latin version of Francisco Pipino). It would seem that the book was translated as early as the first half of the 15th century, possibly by the Infante D. Pedro, and subsequently modernized (see Introduction pp. xxi-xxv).
53 O Manuscrito ‘Valentim Fernandes’, ed. A. Baião (Lisbon, 1940). The book is a compilation of materials from different accounts and records arranged by the person mentioned in the preceding footnote.
54 I owe knowledge of this passage to J. P. Machado's quotation in BF 7.154.
55 Elrey tem cada dia 35 ygoarias de diversas maneiras (Val. Fern. 28) ‘The king is served thirty-five different dishes every day’; Quãtas molheres tever cada huũ tãtas ygoarias ha de ter no jãtar e outras tãtas na çea, porq̃ cada mother the ha de mãdar sua ygoaria (Val. Fern. 76) ‘According to the number of wives that each one has, he is served a commensurate variety of dishes at dinner and at supper as well, inasmuch as each wife has to send him a dish of her own’; Ygoarias, as quoaes vem em hũas bategas que são bacias douro (Cronica 70) ‘Tidbits which are served each in a batega, i.e. a gold dish’.
56 A brief description of this Tratado de Cozinha and a fragment of a few lines are found in J. Leite de Vasconcelos, Textos Arcaicos3 99–100 (Lisbon, 1922–3).
57 Garcia de Resende, Miscelânea, ed. Mendes dos Remédios (Coimbra, 1917), especially the description of the Indies (24–8). For one example of igoaria in a different work of this author see RL 1.123.
58 Garcia da Orta, Colóquios dos Simples e Drogas, 2 vols., ed. Conde de Ficalho (Lisbon, 1891–5). The statement is based on the perusal of the first 25 dialogues (volume 1).
59 Cancioneiro Geral de Garcia de Resende, 5 vols., ed. A. J. Gonçálvez Guimarãis (Coimbra, 1910–7). The word under study has not been encountered in the first two volumes; but see RL 13.423.
60 Apart from Cancioneiro da Ajuda and Textos Arcaicos the following sources were consulted: Die Lieder des Trobadors D. Joan Garcia de Guillade, ed. O. Nobiling (Erlangen, 1907); Um trovador do periodo post-dionisiano : Dom Afonso Sanches, ed. A. Magne, RFH 1.58–88; The Eleven Songs of Joan Zorro, ed. A. F. G. Bell, MLR 15.58–64; Cancionero de Payo Gómez Chariño, ed. A. Cotarelo y Valledor, BRAE 16.467–91, 615–41, and 17.523–38, 664–87; Zwanzig Lieder des Joan Ayras de Santiago, ed. A. Gassner, Misc. C. Mich. Vasc. 385–418.
61 The following sources were consulted: Contemplação de São Bernardo, ed. J. P. Machado, BF 6.97–157; Vida do cativo monge confesso, ed. A. Roseira, BF 1.40–52, 125–62; Dois fragmentos de uma vida de S. Nicolau do século XIV, ed. P. A. d'Azevedo, Festschrift Mussafia 580–6; Vida de Eufrosina, ed. J. Cornu, Rom. 11.357–65; Maria Egípcia, ed. J. Cornu, Rom. 11.360–81; Traité de Dévotion, ed. J. Cornu, Rom. 11.381–90; A Visão de Tundalo, ed. J. J. Nunes, RL 8.239–62.
62 The sources consulted include: Fragmentos de una traducción portuguesa del Libro de Buen Amor de Juan Ruiz, ed. A. G. Solalinde, RFE 1.162–72; K. S. Roberts, Orthography, Phonology, and Word Study of the Leal Conselheiro (Philadelphia, 1940); Duarte Galvão, Cronica delrey Dom Affomsso, ed. A. R. Nykl (Cambridge, 1942); Fragmento de una versión galaico-portuguesa de Lanzarote del Lago, ed. M. Serrano y Sanz, BRAE 15.307–14; Vida e Feitos de Julio Cesar, ed. M. Rodrigues Lapa and J. B. Aquarone, BF 2.207–23, 315–26; 3.59–76, 207–17, 350–66; 4.92–108, 341–57; H. H. Carter, Paleographical Edition and Study of the Language of a Portion of Codex Alcobacensis 200 (Philadelphia, 1938), glossary.
63 However, iguaria is absent from several treatises of falconry and hunting, e.g. Uma tradução portuguesa desconhecida do Tratado de Cetraria do Rei Dancus, ed. G. Tilander, BF 6.439–57; Livros de falcoaria, livro de citraria e experiencia de aigũs caçadores, ed. M. Rodrigues Lapa, BF 1.199–234; Livro de Falcoaria de Pero Menino, ed. Rodrigues Lapa (Coimbra, 1931). Mestre Giraldo's works edited by G. Pereira, Livro de Alveitaria, RL 12.1–60, is not available to me at present, but C. Michaëlis, who studied its vocabulary thoroughly, does not mention iguaria. At any rate, iguaria does not seem to have been a technical term at that early date.
64 Livro da Montaria feito por D. João I, Rei de Portugal, ed. F. M. Esteves Pereira 41 (Coimbra, 1918).
65 See Introduction, pp. xii-xiv.
66 The quotation is from J. Leite de Vasconcelos' edition in RL 8 and 9. The work is anonymous. Its date is discussed RL 9.66–7; the reason why A. F. G. Bell, Portuguese Literature 60 (Oxford, 1922), assigns it to the early 15th century is not transparent. On the style of the fables, full of colloquialisms (“simples e familiar”, “dialogado em parte”), see RL 9.64; on the spelling with h-, RL 9.51. The word has not been included by Leite in his vocabulary; I owe knowledge of its occurrence to the index in Huber's grammar.
67 For the use of manjar (borrowed from French), see Maria Eg. f. 62r; M. Paulo ff. 8v, 16v, 21r, 33v; MS Val. Fern. 78; Garcia da Orta 1.81.
68 Aside from the above-mentioned example from Fr. de Sousa (1697), compare the following passage from Cottineau de Kloguen, Bosquejo Histórico de Goa (1829), quoted by Dalgado s.v. caril: Caril he a sua principal iguaria, e o arroz cozido a comida de manhã e de tarde ‘Curry is their chief food, and boiled rice is what they eat in the morning and in the afternoon’.
69 Examples of the use of comer(es) are found: Livro da Montaria 63; MS Val. Fern. 42; M. Paulo ff. 20v, 87v; Garcia da Orta 1.83, 185, 267.
70 Manti(i)mento was more specifically akin in its use to ‘provisions, victuals’, but could also mean ‘food’; see Maria Eg. ff. 53v, 54r; L. Mont. 60, 61, 65; M. Paulo ff. 11v, 13v, 24v, 39r; MS Val. Fern. 46, 48.78, 87; Garcia da Orta 1.82; Garcia de Resende, Miscelânea 66.
71 See Pero Menino ff. 38, 39, 40, 42; L. Mont. 64; Fernão Lopes, Crónica de D. João I, ed. Braamcamp Freire 41 (Lisbon, 1915); M. Paulo f. 21v; Canc. Geral 1.216.
72 See M. Paulo ff. 78v, 39v, 43v, 44v; Garcia da Orta 1.109. The derivative especiaria will be discussed in the following chapter.
73 See Pero Menino f. 41. Compare the similar development of Süssigkeiten in German and of sladosti in Russian.
74 The importance of this shift has not been duly recognized by Ernst Tappolet, Die romanischen Verwandtschaftsnamen mit besonderer Berücksichtigung der französischen und italienischen Mundarten (Strassburg, 1895). My own findings are as follows. Padre ‘father’ prevailed until the end of the 15th century: Traité de Dévotion ff. 158v, 161r; S. Nicolau (twice); Cod. Alc. 200 f. 200r; Eufrosina f. 43r; Nobiliário T. Arc. 41; Dial. de S. Greg., T. Arc. 45; Crón. João I 57, 60, 61; Hist. Geral de 1344, T. Arc. 55; Crón. Condestabre, T. Arc. 83; M. Paulo ff. 18v, 79v; MS Val. Fern. 39; Canc. Ger. 1.306. Pai ‘father’ appeared in Fernão Lopes: Crón. João I 82; M. Paulo f. 42r; MS Val. Fern. 27, 35, 54, 55, 56, 76, 79; Canc. Ger. 1.54, 55, 306, 321; Garcia de Resende, Miscelânea 23, 40. Madre ‘mother’ appears to have been used even more consistently throughout the 15th century: C. V. Nos. 227, 255, 295, 336, 515 et passim; Eufr. f. 43r; Cod. Alc. 200 f. 196v; Crón. Condestabre T. Arc. 83; Crón. João I 57, 60, 66; M. Paulo f. 20v; Canc. Ger. 1.361. Mai ‘mother’ seems to have been unknown to the Middle Ages; the earliest occurrence known to me is in Flos Sanctorum, quoted by J. J. Nunes, Crestomatia Arcaica2 212 (Lisbon, [1921]); mãi must be an even more recent form. It would be worthwhile to study the rise of iguaria in conjunction with the obliteration of such typical Old Portuguese words as filhar ‘to seize’, leixar ‘to let, to leave’, and the like.
75 For instance, the spread of mochacho ‘boy’ and nadie ‘nobody’ in the course of the 15th century. See the writer's article Old Spanish otri(e), nadi(e), to be published in Hisp. Rev.
76 The phenomenon has been discussed by W. v. Wartburg, Zur Benennung des Schafes in den romanischen Sprachen (Berlin, 1918; Sitzungsber. d. Preuss. Ak. d. Wiss., Phil.-Hist. Kl.).
77 M. Grammont, La dissimilation consonantique dans les langues indo-européennes et dans les langues romanes 10 (Dijon, 1895); for criticism, see G. Paris, Mélanges linguistiques 129–49 (Paris, 1909).
78 E. Schopf, Die konsonantischen Fernwirkungen: Fern-Dissimilation, Fern-Assimilation und Metathesis 149–56 (Göttingen, 1919).
79 R. de Sá Nogueira, Subsidios para o estudo da dissimilação em português, BF 5.115–62 (particularly 158).
80 E.g. Pero Menino, Falcoaria, ff. 31r, 43v. On *Pair' eres > Paieres, see LP 1.7 On the place-name Bafoeira < baforeira ‘fig-tree’ see RL 1.298.
81 W. Meyer-Lübke, La sonorización de las sordas intervocálicas latinas en español, RFE 11.1–32, militated in favor of a late sonorization, but not all of his arguments have been found valid. The opposite view has been taken by W. von Wartburg, Die Entstehung der romanischen Völker 55–6 (Halle, 1939).
82 On accent-shift in Spanish, see R. Menéndez Pidal, Manual de gramática histórica española6 36–41 (Madrid, 1941); Rufo Mendizabal, Acentuación de los grecismos, BAE 11.408–26; H. Gavel, Essai sur l'évolution de la prononciation du castillan depuis le XIVme siècle 494–510 (Paris, 1920); S. G. Morley, La modificación del acento de la palabra en el verso castellano, RFE 14.256–72, and especially Amado Alonso, Cambios acentuales, BDHA 1.317–70 (Buenos Aires, 1930). A new monograph on this subject is being written at present: M. R. Lida, Los grecismos en español; con especial atención al estudio de la forma y de la acentuación (announced by M. W. Nichols, A Bibliographical Guide to Materials on American Spanish, under No. 119).
On problems of accentuation in Portuguese, see A. R. Gonçalves Viana, Essai de phonétique et de phonologie de la langue portugaise d'après le dialecte actuel de Lisbonne, BF 7.165–243, especially 230–9 (revised edition); A. C. de Figueiredo, O que se não deve dizer: Bosquejos e Notas de Filologia Portuguesa3 1.58–65 (Lisbon, 1916); idem, Anormalidades Prosódicas, in Novas Reflexões sôbre a Lingua Portuguesa 9–24 (Lisbon, 1923); idem, A Linguagem de Camões nas suas Relações com a Linguagem de Hoje 33–4, 49–55 (Lisbon, 1930). Also, Leite de Vasconcelos, Lições2 79–86; A. Magne, Achádego ou achadêgo?, RFH 2.80–2; J. P. Machado, Campa, Quinta, Venta, BF 6.407–18 (with references to earlier publications on this subject).
83 On -ia and -aria in Vulgar Latin and Romance see W. Meyer-Lübke, Rom. Gramm. 2.452; Grandgent, Introduction 65; F. T. Cooper, Word Formation in the Roman Sermo Plebeius 74 (New York, 1895); G. N. Olcott, Studies in the Word Formation of the Latin Inscriptions 173–6 (Rome, 1898); Veikko Väänänen, Le latin vulgaire des inscriptions pompéiennes 156–63 (Helsinki, 1937); T. Claussen, Die griechischen Wörter im Französischen, RF 15.774–883, especially p. 812 (with reference to other Romance languages). On the accentuation of -ia in Portuguese, see C. de Figueiredo, O que se não deve dizer2 3.204–5, 271–3 (Lisbon, 1916) showing the contrast in the treatment of medical and botanic terms.
84 On Greek elements in the Portuguese lexicon, see F. Rebêlo Gonçalves, Os elementos gregos no vocabulário português, BF 1.37–46, 145–54, 319–26; idem, De vocibus Graecis in sermone Latino vulgari (unpublished thesis, not accessible to the writer); A. Magne, Palavras portuguesas derivadas do grego RFH 1–382–400, 547–50, 2.407–28; Ramiz Galvão, Vocabulário das Palavras Portuguesas Derivadas do Grego, includes only erudite formations.
85 On suffix-change in Romance see G. Cohn, Suffix-wandlungen im Vulgärlatein und im vorliterarischen Französisch nach ihren Spuren im Neufranzösischen (Halle, 1891); A. Alonso, op.cit.; E. G. Wahlgren, Un problème de phonétique romane: le développement d > r (Upsala, 1930). The phenomenon in Portuguese has been studied by J. da Silva Correia, Associação vocabular natural, BF 2.347–73, and R. de Sá Nogueira, Subsídios para o Estudo das Conseqüências da Analogia em Portugués §§76, 154–8 (Lisbon, 1937).
86 See Nachschrift, ZFSL 58.306–11. The opposite view has been taken by G. Moldenhauer, Zur Entstehung des französischen Suffixes -erie, ZFSL 58.296–306, 381–2, who believes that the shift -ária > -aría occurred in Church Latin as late as the 11th century and thence radiated into the vernaculars. D. Scheludko, Über den Ursprung und die Geschichte des rumänischen Suffixes -ie, NM 33.82–92, attempts to dissociate Roum. -ie from Western Romance -ia, tracing it to -īlia. Moldenhauer's and Scheludko's theories are untenable. See also Meyer-Lübke, Histor. Gramm. d. frz. Spr. 2.65, 68 and L. Spitzer, Warum ersetzt frz. -erie (dtsch. -erei) das alte -ie (-ei)?, ZRPh 51.70–5.
87 For judaria, see Crón. João I 29; Garcia de Resende, Miscelânea 52; for mouraria, Miscelânea 52 and Canc. Ger. 2.176; see also cafraria ‘negroes’ in Dalgado s.v.
88 For especiaria see M. Paulo ff. 5r, 8v, 32v; Miscelânea 98; Garcia da Orta 1.67, 257; Dalgado, s.v. coco (1516) and laca (1508). Variant forms are espeçeria M. Paulo f. 78v, Garcia da Orta 1.201; speçaria MS Val. Fern. 23, 24; espiçiaria, quoted from A. de Albuquerque by Dalgado, s.v. candura. A similar formation is drogaria, see Dalgado s.v. gengibre. Especiaria ‘spice’ and drogaria ‘drug’ are likely to have influenced the development of iguaria.
89 Pedraria was used in reference to the wealth of the Indies, e.g. Canc. Ger. 1.350; Miscelânea 15, 28, 37, 43, 96; Garcia da Orta 1.217, and in numerous authors quoted by Dalgado, e.g. A. de Albuquerque (1513) s.v. almiscar; G. Correia (1514) s.v. gomedar; D. Barbosa (1516) s.v. canafístula and jagonça; Crónica de Bisnaga (1525) s.v. aloés; F. Mendes Pinto (1541) s.v. andor, cabisondo; Castanheda (1552) s.v. balais; Damião de Góis (1566) s.v. balais, jagonça; P. Valignano (1581) s.v. boião; P. F. Guerreiro (1605) s.v. leque; Diogo do Couto (1612) s.v. chapim; J. de Cabreira (1635) s.v. bazar, and so forth. Pedraria could also be associated with iguaria because of their common reference to the pomp and exquisiteness of courtly life.
90 Derivatives in -aria with a collective meaning include argentaria ‘silver ware’ Canc. Ger. 1.350; bategaria ‘set of dishes’ Fr. Gaspar da Cruz (1569) apud Dalgado; chaparia ‘plates’ Crónica de Bisnaga (1525), see Dalgado, s.v. cofo; rouparia ‘clothing’ P. F. de Queirós (1687), see Dalgado s.v. bentó; tapeçaria ‘tapestry’ Garcia de Resende Misc. 79, P. F. de Queirós in Dalgado s.v. anil; vozearia ‘shouting’ Diogo do Couto (1612) in Dalgado s.v. atabale.
91 C. Basto, Formação popular de nomes-de-unidade, Misc. C. Mich. Vasc. 374–84 (with reference to Galician and Minhoto dialects, and indication of earlier bibliography).
92 On the development of the ai diphthong in Ibero-Romance, see R. Menéndez Pidal, Orígenes del español2 78–103 (Madrid, 1929), presenting evidence of the slower development of the diphthong in the West and the South than in the Center and the East; [aj] as contained in the suffix -ariu > -airu is shown to have made more rapid progress than primary and secondary [aj] found elsewhere. On the strength of this argument, an early date must be posited for the stress-shift in iequária; the result of an undisturbed development of the Latin word in Portuguese would have been *(g)igueira.
The chronology will be better understood in the light of the following data. Lt. -ariu has been transmitted into Old Portuguese:
(a) in vernacular words, as -eiro; see Huber 244–5;
(b) in learned words, as -ario; cf. cõcubinario Cod. Alc. 200 f. 218 v, herbolario ‘herbalist’ Orta 1.258, neçessario Mar. Eg. ff. 55r, 65r, Eufr. f. 52r, Cod. Alc. 200 f. 196r, Crón. João I 89 notario T. Arc. 109 (A. D. 1299), solitario Boosco del. in T. Arc. 67, Canc. Ger. 1.356, somario ‘summary’ L. Cons. f. 12v, Canc. Ger. 1.42; sudario ‘holy handkerchief’ M. Paulo f. 17v; tributario M. Paulo f. 4v; veterinario Orta 1.85; voluntario Canc. Ger. 1.42, 219. To these words must be added a vernacular distortion of a semi-learned formation, alimaria ‘beasts’ Fab. f. 19r, L. Mont. 44, L. Cons. f. 38r, M. Paulo ff. 20r, 20v, 22v, 23r, 23v, etc., Canc. Ger. 1.356, Tenreiro BF 7.134, traceable to animalia Mar. Eg. ff. 54v, 62r. Fab. f. 1r, L. Mont. 36, 50, 61, L. Cons. f. 11r, M. Paulo ff. 5r, 11r, 11v, 12r, 12v, etc., MS Val. Fern. 31, 42, 52;
(c) in semi-learned words, as -airo (this pattern disappeared about the year 1600, through Latin or Castilian influence): aversairo ‘adversary’ L. Cons. f. 7v, Crón. João I 85, Canc. Ger. 2.191; boticairo ‘druggist’ Orta 1.34, 60, 86; contrairo ‘contrary’ Mar. Eg. f. 56v, Cod. Alc. 200 f. 198r, Fab. ff. 18r, 23v, 41v, L. Mont. 4, 6, 8, Crón. João I 1, 2, 5, 14, Canc. Ger. 1.186, 2.121, 191, 301; co(r)sairo ‘corsair’ MS Val. Fern. 22, Canc. Ger. 1.187, J. de Barros BF 7.127; fadairo ‘fate’ Canc. Ger. 2.121; falsairo ‘forger’ Canc. Ger. 1.123; lapidairo ‘stonecutter’ Orta 1.131; necessairo ‘necessary’ Orta 1.364; notairo ‘notary’ Nobiliário T. Arc. 43, M. Paulo f. 38r; repairo ‘resource’ MS Val. Fern. 63, Canc. Ger. 1.44 (where -airo strictly speaking is not a suffix); sudairo ‘holy handkerchief’ M. Paulo f. 88v; vigairo ‘vicar’ Cod. Alc. 200 f. 200v, Canc. Ger. 1.316. Modern Portuguese uses adversário, boticário, contrário, fadário, vigário instead. Note the extension of use in mundairo ‘mundane’ Canc. Ger. 1.247 and the possible influence on ventuira ‘luck’ and its derivatives.
To make the situation even more complicated, Lat. -eriu has been transmitted into Old Portuguese:
(a) as -eiro: cativeiro ‘captivity’ MS Val. Fern. 18, Canc. Ger. 1.270, 357, 2.201; lazeyra ‘misery’ Mar. Eg. f. 57v; refrigeiro ‘refreshment, comfort’ Canc. Ger. 1.63; (p)salteiro ‘psalter’ Tundalo 258, 259, Cod. Alc. 200 f. 201r, Canc. Ger. 1.328. To these must be added marteiro < martyriu Tundalo 259, Fab. f. 31v, Misc. 103, Canc. Ger. 1.24, 295, 360, 374 (beside martirio M. Paulo f. A vi r and marterio Canc. Ger. 2.42);
(b) as -erio: adulterio Tr. Dév. f. 163r, Tundalo 258, Canc. Ger. 1.225; materia P. Menino f. 39; mesterio < mysteriu Crón. João I 41; refrigerio M. Paulo f. 78v; vituperio (also viltoperio) ‘infamy’ Cod. Alc. 200 f. 215r, Canc. Ger. 1.225.
This distribution is peculiar to Portuguese; semi-learned reflexes of -ariu and -eriu are unknown to Spanish. It will be observed that there was some vacillation between -erio and -eiro (refrigerio ~ refrigeiro) and also between -ario and -airo (necessario ~ necessairo, notario ~ notairo, sudario ~ sudairo), whereas the line of demarcation between -eiro < -ariu and -airo was a strict one. This situation is further proof of the early segregation of -ariu in the vernacular stratum and compels us to assume an early date for the stress shift in iequãria.
93 See, in addition to the above-mentioned manuals, G. Baist, Spanische Sprache, Gr. Grundr.2 1.899; A. Zauner, Altspanisches Elementarbuch2 34 (Heidelberg, 1921) and quite recently R. K. Spaulding, How Spanish Grew 91–2 (Berkeley and Los Angeles, 1943): ‘y is an early step in the evolution of the sound and Castile has passed beyond it’.
94 W. Meyer-Lübke, Zur Geschichte von Ge, Gi und J im Romanischen, Vox Romanica 1.1–31, especially 29–31. The author came to the conclusion that in Castilian yema ‘bud’ yolk', yerno ‘son-in-law’, yeso ‘gypsum’, yuso ‘below’, yugo ‘yoke’, yunta 'couple (of oxen)“ yace 'he lies', yanta ‘he dines’, Sant-iago were representative of the vernacular stratum' whereas junio ‘June’, junco ‘rush’, justo ‘just’, juro ‘I swear’, juzgo ‘I judge’, junto ‘I join’, Jorge ‘George’, joven ‘young’, jornada ‘journey’, jornal ‘day-work’, jamás ‘never’ were formations borrowed from mediaeval Latin, French, Provençal, or Catalan (note the contrast to Spaulding's statement). In pretonic position, yod was altogether dropped, as in enero ‘January’, hermano ‘brother’, hiniesta ‘Spanish broom’, hinojo ‘knee’. Before ue, [ž] developed in vernacular words: juego ‘play’, jueves ‘Thursday’, juez ‘judge’. Uncir, uñir ‘to join’ < iungere was due to dissimilation of palatal groups (and so were, according to Baist, echo ‘I throw’ and ayuno ‘I fast‘). A different and more illuminating explanation has been offered by R. Menéndez Pidal, Manual de gramática histórica española6 124–5 (Madrid, 1941), who makes the development of initial yod dependent on its environment: it has been preserved before a stressed front vowel (including a), eliminated before an unstressed front vowel, shifted to [ž] before any back vowel; the remaining exceptions are discarded as dialectalisms.
95 On the possible phonetic value in Old Spanish of h- in words like hermano and hinojo see W. B. Hawkins, Flight from Assimilation and Trial and Error in Spanish Linguistics, HR 10.273–84.
96 See Gr. Grundr.2 1.991. This explanation has been accepted by Nunes, Compendio2 168, and has appealed likewise to Huber, APtg. Elementarbuch 112; Williams mentions it without endorsing it.
97 Tappolet, Romanische Verwandtschaftsnamen, passim, has recorded numerous instances of apocope and aphaeresis, but has failed to produce any evidence of a leaning toward amalgamation with the preceding possessive pronoun (comparable, for instance, to the place names Santervas < Sancti Gervasii, Santibáñez < Sancti Johannis, Santurde < Sancti Georgii, mentioned by Baist in Gr. Grundr.2 1.899).
98 See his Romanische Namenstudien: Altportugiesische Personennamen (Kais. Ak. d. Wiss. Wien, Sitzungsb. phil.-hist. Kl., vol. 149, 184).
99 That germānu was a word used in the vernacular (at least of the Iberian and Apennine peninsulas), whereas frāter was rather favored by ecclesiastic writers, has been conclusively demonstrated by Paul Aebischer, L'italien prélitteraire a-t-il dit germano et germana pour ‘frère’ et ‘soeur‘? Etude de stratigraphie linguistique, ZRPh 57.211–39.
100 There may exist a fourth reflex, to wit ye-; compare OSp. iermano (a graph that may be interpreted in a variety of ways) and Yelvira, listed by Huber, Aptg. Elem. 112.
101 Ptg. janeiro is an outgrowth of iānuāriu, Sp. enero goes back to *ienuāriu; see REW3 §4576; Einführung3 §138; M. L. Wagner, La stratificazione del lessico sardo, RLiR 4.7–8.
102 It is generally true that Latinisms merged more intimately with vernacular elements in Portugal than in Spain, thus interfering with native phonological and morphological processes and deeply affecting the lexicon; for the contrast between Sp. interrumpir ~ Ptg. interromper and similar cases, see the writer's Some Contrasts Between Verbal Derivations in Spanish and Portuguese, Univ. of Wyoming Publications 9.53–67; for the conflicting reflexes of the suffix -tiōne, see the article Spanish deleznar ‘to slide’, lezne ‘smooth, slippery’, Hispanic Review 12.57–65; for the different development in the two areas of -entia, see the forthcoming monograph The Development of the Latin Suffixes -antia and -entia in the Romance Languages with Special Reference to Ibero-Romance.
103 See F. Hanssen, Gramática histórica de la lengua castellana 47–8 (Halle, 1913) and the sources mentioned in footnotes 23, 24, 25. In the East of the peninsula, the consonant has not been dropped.
104 A cursory survey of Hispanic place-names of other than Latin provenience did not yield for Portugal further examples of the treatment of initial yod as in [jermanu], [jekwarja]; see G. Sachs, Die germanischen Ortsnamen in Spanien und Portugal 19 (Jena-Leipzig, 1932; Berliner Beitr. z. roman. Phil. vol. 2) and J. M. Piel, Os nomes germânicos da toponímia portuguesa, BF 2.105–40, 224–40, 289–314; 3.37–53, 218–42, 367–94; 4.24–56, 307–22; 5.35–57, 277–88; 6.65–86, 328–50 (final installment not yet available), especially Nos. 326–51, 568–604, and 843–51. Both authors illustrate the preservation of [ge-] as [že-]. No pertinent examples have been found in David Lopes, Toponímia árabe de Portugal, RH 9.35–74; the same author's later essay in RL 24.257–73 was not available for consultation. Generally speaking it is questionable whether Germanic and Semitic words, absorbed in the early Middle Ages, could yield the same result in regard to this shift as the inherited Latin stock.
105 On this influence, see Hanssen, Gramática histórica 33; Menéndez Pidal, Manual6 59–60; Baist, Gr. Grundr.2 1.893; Huber, Altptg. Elementarbuch 63–5; Williams, From Lt. to Ptg. 44.
106 A tentative statement to this effect is found only in Huber, op.cit. 64.
107 Eguariço ‘hostler’, eguariça ‘horse's stable’ developed along different lines because of the secondary stress on the initial vowel through association with égua < equa ‘mare’; see Cornu, Gr. Grundr.2 1.944; for old examples, see Elucidário s.v. (the meaning is given inaccurately). On iguar and igualha see RFH 1.358–9. Examples of igualdança ‘equality’: Elucidário s.v.; L. Mont. 162; Crón. João I 286. Examples of igual(d)eza ‘equality’: Livro da Virtuosa Bemfeitoria 52, 83 (ed. J. Pereira de Sampaio; Porto, 1910). Huber, op.cit. 64 quotes igualmente as of the year 1214. Other examples are: igualmente ‘equally’ Fab. f. 4v; iguais P. Men. f. 56v; igualhar ibid. f. 52r; iguallar Crón. João I 56; iguaes Côrte Imp. T. Arc. 61. Noteworthy is igalmente Eufr. f. 43v, suggesting the existence of a variant form igal (cf. Ptg. antigo ‘old’ beside Sp. antiguo) and iugua(a)es Cod. Alc. 200 ff. 210r, 216v, iugualmente ibid. f. 215v, reflecting the same development as aqua > augua. In Spanish, igual superseded egual at a considerably later date.
108 To begin with, the etymology of minguar is uncertain. Huber and Williams give *minuāre, so does A. Alonso BDHA 1.408 (and quite recently Menéndez Pidal, Manual6 60) for Sp. menguar, but C. Michaëlis, Homen. Men. Pid. 3.458 argued in favor of a derivative in -ificāre. The next difficulty is the problem of the dependence of the vowel on the accent, because of the co-existence of stressed and unstressed forms. Furthermore, it is not clear whether mingua directly perpetuates the Latin i or has passed through the stage mengua; yet even if we knew whether the vowel had been preserved or reintroduced, it would still remain problematic whether the given treatment was due to [w] or to the preceding consonant cluster; compare for methodological guidance P. Fouché, Etudes de philologie hispanique: ō et ŭ latins en castillan, RH 77.2–24. The thorniest question is that of the regional distribution; against MSp. mengua, lengua there exists OSp. mingua, see Menéndez Pidal, Cantar de Mio Çid 760. In Old Portuguese continued wavering can be observed, with the i forms prevailing from the outset; Mar. Eg. f. 63r mingua, minguado; Tr. Dév. f. 149v mingua; Julio César f. 18c mingua, f. 32b mingasse (suggesting a variant form?), f. 47d minguada, f. 49d mingu[av]ã; Cod. Alc. 200 f. 199v mengua (twice), f. 204v linguagẽ, f. 205v mjngua, f. 216v amẽguamento ‘decrease’; Tundalo 254 lingua, 261 mingua; Fab. ff. 3r, 10v limgua, ff. 3v, 9v, 18v, 30v mingua, f. 11v vinguança ‘revenge’; M. Paulo f. 80v lenguoa, f. 92r lenguagẽs, f. 92v lingua. The question deserves special study.
109 Giolho is, indeed, by far the predominant form throughout the Old Portuguese period: see Mar. Eg. ff. 55r, 62v, 63r; Barlaam and Josaphat f. 33v (A Portuguese Version of the Life of B. and J., ed. R. D. Abraham; Philadelphia, 1938); M. Paulo ff. 33v, 34v, 82r, 83r, 88r; MS Val. Fern. 21, 28, 46, 58, 62, 67, 77, 84; Canc. Ger. 1.306; Garcia da Orta 1.265. However, geolho is also found, e.g. Josaph. Barl. f. 43v; Milagres de S. António, see Huber 237; MS Val. Fern. 16. The modern form joelho, arrived at through metathesis, is recorded early: Tr. Dév. f. 156r, Crón. João I 27, M. Paulo f. 19v, Canc. Ger. 1.200. The oldest form, gẽolho, is encountered in the 13th century Cancioneiros; see Nunes, Cr. Arc. 392 and Huber, Elementarbuch 54. The metathesis must have occurred while the e was nasalized, since two consecutive oral vowels would have been dealt with in a different way. On Gal. geonlho see RL 13.311.
110 Nunes, Compêndio2 61 was at a loss how to explain giesta, suggesting two tentative solutions: dissimilation of consecutive e sounds or development as in hiatus. Either interpretation is unsatisfactory. Why indeed should genesta > *geesta not have experienced the vowel contraction which struck similar words? In the Documentos de Beira, ed. P. A. d'Azevedo, RL 8, we find Meendiz (1270), meesmo ‘same’ (1270), perteemça ‘property’ (1292), séélo ‘seal’ (1292), which subsequently yielded Mendes, mesmo, pertença, selo. Tundalo contains sééstro ‘left’ (p. 249), tréévas ‘darkness’ (p. 249), enpéécer ‘to hinder’ (p. 250), freestra ‘window’ (p. 257), which evolved into sestro, trevas, empecer, fresta. If genesta swerved from this line, it can only have happened owing to the palatalizing effect of [ž].
111 See the interesting data assembled by N. P. Sacks, The Latinity of Dated Documents in the Portuguese Territory 10 (Philadelphia, 1941), who quotes (Latin) iermana, iermano as of the years 897 and 908, respectively, and irmanos, irmanibus (visibly affected by the vernacular) as of 1097. Irmana instead of irmãa, not infrequent in Old Portuguese lyrics, has been explained by H. R. Lang and more recently by M. Rodrigues Lapa LP 1.15 as a significant survival of pre-literary Portuguese (C. Michaëlis preferred to see a Castilianism in it). Huber, Elementarbuch 64, gives a very late date for the persistence of the archaic variety germão (1288). The spelling was in the 13th century: irmão, irmáo, see Docum. Beira as of 1271, 1274 and Docum. do Mosteiro de Chelas (ed. P. A. d'Azevedo, RL 9) as of 1273; later yrmão, hirmão. Irmão has influenced the development of erēmita ‘hermit’ (REW3 §2890) on Portuguese soil; cf. Tr. Dév. ff. 157r, 160 v irmitãao ‘hermit’; Fab. f. 24v ermida beside irmida ‘hermitage’; M. Paulo f. 13r irmitãaes; MS Val. Fern. 38 irmida, 91 hermida.
112 Mentioned by Nunes, Compêndio2 60.
113 Compare the reflexes of creātūra ‘creature’ and leōne ‘lion’: criatura occurs in Tundalo 256, Cod. Alc. 200 f. 201r, Fab. f. 5v, M. Paulo f. A vi v as against the lonely testimony of creatura in the archaic Vida de S. Nicolau; liom in Fab. f. 4v, lyões in M. Paulo ff. 35r, 36v, 40r, 41r, MS Val. Fern. 56, 68, Miscel. 18 as against leom Fab. ff. 5r, 11v, leam Fab. f. 21v.
114 In peiore ‘worse’ the [j] was dropped so early as to justify piores Tundalo 254, pyor MS Val. Fern. 48 beside peor Fab. f. 11r, discussed RL 9.33 (note the retarded development of meão < mediānu ‘middle’ Misc. 54, caused by association with meio ‘midst’). Also, to judge by the testimony of meolo Fab. f. 1v, miólo M. Paulo f. 50r, myolo MS Val. Fern. 70 (< *medullu ‘marrow’), intervocalic -d- must have disappeared prior to -l- and -n-. Otherwise, only e is recorded: Docum. Chelas teor ‘tenor’ (1292); Tundalo 254 jeada ‘hoar-frost’, 259 lealmente ‘loyally’; Fab. f. 8r ameaçar ‘to threaten’, f. 35r semear ‘to sow’; Cast. Perig. T. Arc. 50 theudo ‘held’; L. d. Alveit. T. Arc. 45 amende ‘often’; Eufr. f. 42v amende; Mar. Eg. f. 53v terreaaes ‘earthly’, f. 62v geada; Tr. Dév. f. 149r terrea[e]s; P. Men. f. 41v ameude beside amiude; L. Mont. 49 amende; Crón. João I 37, 66 amende, 81 meudo ‘small’; M. Paulo ff. 1v, 6r terreal, f. 39r amende ‘often’, f. 91v areoso ‘sandy’; MS Val. Fern. 53 meudo, 54 esmeuça ‘he minces’, 54 çeoso ‘jealous’. Note the difference between the development of the product of *genuculu and those of minūtim, minūtu, ex-minūtiat, *terrēnāles, sēmināre in reference to the vowel of the pretonic syllable.
115 Docum. Chelas adeante (1292, 1298); Tundalo 257 adeante beside adiante; Fab. f. 3v per diante; Cast. Perig. T. Arc. 47 deante; Tr. Dév. f. 139r adiantado, f. 154v diante; P. Men. ff. 30r, 33v adiante; Cr. João I 44 adeamte; M. Paulo f. 283 adiante, f. 93r dianteiro.
116 Of Castilian origin; see Nunes, Compêndio2 60.
117 This variant form is listed by Nunes, Compêndio2 61.
118 Throughout the Middle Ages, milhor and melhor were equally common variants. Milhor occurs S. Nicolau (twice), Fab. ff. 3v, 7v, 22v, 32v, 39r, Demanda T. Arc. 44, P. Men. ff. 33v, 63r, 65v, L. Cons. f. 4v, M. Paulo ff. 8v, 93r. Melhor is recorded in Tundalo 254, L. Cons. f. 9v, Crón. João I 4, 44, 70, MS Val. Fern. 14, 56; cf. melloramos ‘we improve’ Docum. Chelas (1277), melorado ibid. (1296), melhorar Fab. f. 36v. The form vilhice ‘old age’ (instead of velh-ice) Eufr. f. 49r can be likened to milhor, but lends itself to other explanations as well.
119 Portuguese medial s, especially as the introductory element of a consonant cluster, is known to have a palatal coloring; see Gonçalves Viana's statement, reprinted BF 7.196. The shift was a latent possibility rather than a ‘law’ and did not affect all words at the same time nor to the same extent. Mizquinho, borrowed from Arabic in the 8th century, is by far the prevalent form: Tr. Dév. ff. 145v, 152r; Mar. Eg. f. 64v; Eufr. ff. 49r, 49v; Tundalo 250, 253, 254, 255; Fab. ff. 16r, 28Br, 31r, 35v; Barlaam f. 35r beside mezquinho Boosco del. Cr. Arc. 215; mizquindade ‘misery’ Mar. Eg. f. 57r, Eufr. f. 49r, Barlaam f. 2v beside mesquījndade Nobil. Cr. Arc. 52. The other extreme is represented by mester ‘need’, borrowed several centuries later from Gallo-Romance. The form with -e- prevails: Docum. Chelas (1292); Nobil. T. Arc. 41; L. Alveit. T. Arc. 45; Tr. Dév. f. 139v; Mar. Eg. f. 53r; Tundalo 258; Fab. f. 43v; Crón. João I 41, MS Val. Fern. 74, as against mister P. Men. ff. 30r, 43r; L. Mont. 46, 67. If in Modern Portuguese the relation of the variants has been reversed (mesquinho, mister), this may be attributed to the increasing supremacy of the trend to dissimilate consecutive vowels over other, in part overlapping phonetic tendencies. In Old Portuguese, mesturar and misturar ‘to mix’, messegeiro and missegeiro ‘messenger’, tesouro and tisouro ‘treasury’ co-existed over a period of centuries.
120 The question may be raised why geada did not yield *giada in harmony with giesta, OPtg. giolho, dialectal jijum and also irmão and iguaria. The answer may be that it was steadily associated with gelo ‘ice’, in which the stem syllable was tonic, hence bound to develop differently.
121 See K. Voretzsch, Einführung in das Studium der altfranzösischen Sprache6 189, 198 (Halle, 1932); M. K. Pope, From Latin to Modern French 239 (Manchester, 1934); U. T. Holmes Jr. and A. H. Schutz, A History of the French Language 33 (New York, [1938]). A. Dauzat, Histoire de la langue française 68 (Paris, 1930), quoting mercides ‘mercies’ as of 627, merely says that the diphthong ei was affected by the preceding palatal (even as Meyer-Lübke phrased it in 1908 in Hist. Gr. d. frz. Spr.1 1.65). Suchier expressed the belief that the shift e > i may have preceded the diphthongization of ē in Northern Gaul (Gr. Grundriss2 1.731). Wartburg, Evolution et structure2 52, speaks vaguely, in reference to merci, of a ‘réaction de la consonne sur la voyelle’. Gamillscheg, in his lecture course on French phonology (1937), disproved the theory of the triphthong. H. Rheinfelder, Altfranzösische Grammatik 1.103–4 (Munich, 1937) is undecided which theory to accept. Nyrop, Gr. hist.2 1.204 and Suchier, Les voyelles toniques du vieux français 48 (Paris, 1906) offer no explanation.
122 A. Darmesteter, Cours de grammaire historique9 1.105; L. Clédat, Grammaire élémentaire de la vieille langue française 286 (Paris, 1887); Meyer-Lübke, Rom. Gramm. 1.112; also Schwan-Behrens and Anglade. Modern treatises that favor this view are scarce: L. Jordan, Altfranzösisches Elementarbuch 72 (Bielefeld-Leipzig, 1923); Brunot-Bruneau, Précis de grammaire historique de la langue française 110 (Paris, 1937) and, surprisingly enough, E. Richter, Beiträge zur Geschichte der Romanismen 95, 116, 168–9, 206–7 (Halle, 1934). A similar palatalization of a front vowel by the preceding c does not seem to be known to Hispanic dialects; see W. Meyer-Lübke, La evolución de la c latina delante de e e i en la península ibérica, RFE 8.225–51.
123 Aside from the questions discussed in the preceding footnotes, the development of pretonic e presents numerous difficulties which await clarification. J. Moreira, RL 8.281 explained lectione > lição ‘lesson’ through assumption of an intermediate stage *leição (cf. eleição ‘election’); assimilation to the -itiōne > -ição pattern is more likely to underlie the shift. The theories proposed by J. Leite de Vasconcelos, RL 13.433–4, are subject to caution: to illustrate the reputed shift ei > i in pretonic syllable, he posits unnecessarily complicated etymologies, e.g. idade ‘age’ < aevitāte, not aetāte; iguar < adaequāre, likening these cases to isento < exemptu, igreja < ecclēsia, disregardful of chronological discrepancies and phonological contrasts.
Only an approximate idea of the diversity of the factors involved can be given here. Fezerom > fizeram ‘they did’ is due to analogical influence of fiz ‘I did’. Enquerer ‘to inquire’ (Docum. Beira, 1292; Docum. Chelas, 1298) yielded to inquirir (M. Paulo f. 2r) because of (1) conjugation shift -er > -ir, (2) assimilation of pretonic e to tonic i (responsible for common pidir ‘to ask’, aligria ‘mirth’ Tundalo 255, 261, Tr. Dév. f. 149v, mancibia ‘youth’ Tr. Dév. f. 143v, riliquias ‘relics’ Tundalo 258), (3) vacillation between en- and in-, the vernacular and the learned outgrowths of the same Latin prefix. The extreme unpopularity of such sequences as e – e, e – ei, ei – ê(i) is an agency whose power in modifying words within the period of time 1200–1600 has been underrated; it led to the dissimilations i – e, i – ei. Examples: dereito ‘right, law’ (cf. Sp. derecho) Docum. Chelas 1277–96, Docum. Beira 1292, Tundalo 250, 261, Fab. ff. 4v, 42v, Crón. João I 37, 80, but direito M. Paulo f. 28r, MS Val. Fern. 28 (co-existence of learned directo ‘straightforward’ Tundalo 250, 253 may have paved the way); pequeno ‘small’ used currently, but piqueno P. Men. ff. 40v, 57r; L. Mont. 77; enteiro ‘entire’ (cf. Sp. entero) as late as M. Paulo ff. 18r, 89v, but the new form inteiro as early as P. Men. 36v, 64r; deleitar ‘to delight’ Tundalo 258, but dyleito ‘delight’ ibid. 255, 256. Idade, whose i- dates from the earliest time (Eufr. f. 43v, Hist. Ger. T. Arc. 57, Crón. João I 29, 59, 60) must be dissociated from igreja, whose i- evolved toward the close of the Middle Ages: eygleyga Docum. 1193 T. Arc. 15, eygreya Docum. Beira 1274; eygreia S. Nicolau; egreja Mar. Eg. passim, Eufr. ff. 44v, 45v, Crón. João I 24, 57, 62, 80; igreja Eufr. 43v, 50v, Tr. Dév. f. 162v, Tundalo 249, 260, M. Paulo f. 23v (parallel to direito, dileito, piqueno, inteiro; cf. also Theresia > Tareija, ceresia ‘cherry’ > OPtg. cereija, çareija [Compêndio2 56, 59]). Popularity of the sequence i – í, documented above, is responsible for the transformation of vezinho ‘neighbor’ (cf. Sp. vecino) Fab. f. 5v, M. Paulo ff. 5v, 23r, MS Val. Fern. 28, 34 into vizinho Eufr. f. 46v, M. Paulo ff. 18v, 38r. The shift emveja ‘envy’ (cf. Sp. envidia) Crón. João I 56 to MPtg. inveja can again be explained by dissimilation of e – é, but the prefix may also have played an important part in the process (cf. enperador ‘emperor’ S. Nic. as against imperador Miscel. 6, 16, and the like). Cilada ‘ambush’ is a cross of OPtg. ciada and the Castilianism celada MS Val. Fern. 93. The development of inimīcus ‘foe’ on Portuguese soil requires special study.
124 See Ernout-Meillet, Dictionnaire étymologique de la langue latine2 472 (Paris, 1939); Walde-Hofmann, LEW5 1.673 (Heidelberg, 1938). But see Schuchardt ZRPh 28.448 on gésier.
125 A. Zauner, Die romanischen Namen der Körperteile, RF 14.339–530, specifically 504–7. W. Meyer-Lübke, Neubenennungen von Körperteilen im Romanischen, WS 12 (1929), is not available to me at the present moment.
126 G. Paris, Ficatum en roman, Miscellanea Ascoli 41–63 (Turin, 1901); Rom. 30.568–9. Both items are reprinted in Mélanges linguistiques 532–53 (Paris, 1909).
127 For bibliographical data, see FEW 3.490–3 (1932) and REW3 §8494 (s.v. sykoton). Important is the opinion of M. L. Wagner, La stratificazione del lessico sardo, RLiR 4.14–7, who (in accordance with Meyer-Lübke, Einführung3 §168, but against Wartburg) regards fícatu as the oldest reflex. The same author's Studien über den sardischen Wortschatz 115 (Geneva, 1930) contains only the correction of a minor detail. REW3 accepts Wartburg's theory with one reservation concerning the development in Eastern Sicily.
128 The nucleus of Paris—Wartburg's theory is as follows. Early appreciation of the Greek cuisine stimulated the Romans to coin, with the means of their own language, an equivalent of (
παρ) συκωτό. This first attempt at an imitation was *fīcātum, preserved in archaic peripheral dialevts, such as Roumanian, Dalmatian, Raeto-Romance. In the meantime the Greek worc radiated directly westward, reaching Rome as *sǽcotum, *sẹ́cotum. The following three stages of the process reflect the gradual assimilation of native *fīcātum to imported *sécotum: (1) *fícatu, only the stress is affected; reflected in Spanish, Portuguese, and some Italian and Sardinian dialects; (2) *fécatu: both stress and tonic vowel are affected; reflexes in Astur-Leonese, Gallo-Romance, and Italian; (3) *fécotu: stress, tonic vowel, and post-tonic vowel are affected: traces only in Eastern Sicily, most vulnerable of all Romance provinces to Greek influence. Metathesis, suffix change, and other accidents complicate the development in individual dialectal areas.
In elaborating this theory, the following writings may be taken into consideration:
E. G. Wahlgren, Un problème de phonétique romane 163–71 (distinguishes between OFr. feie < *fécata and OFr. fie < *fídicu, considers firie Chanson de Roland ed. Bédier line 1278 a ἄπαξ λεγόμενον patterned after empirie ‘empire’, martirie ‘martyrdom’); G. Reichenkron, Französisch moustache, Spanisch mostacho, Italienisch mostaccio, mustacchio, ZFSL 58.48–55 (example of continued radiation of a Greek word toward the West; dialectal μυστάκων penetrates into Vulgar Latin in the 3rd century and is associated with -āceum; subsequently Byzantine μουστάκι is assimilated to -āculum). On changes undergone by Greek words incorporated into Latin, see V. Väänänen, Mots grecs changeant de déclinaison en latin, NM 39.305–14 (1938). Instances of fígado ‘liver’ in Old Portuguese include: P. Men. ff. 53v, 55r; L. Mont. 19, 63; MS Val. Fern. 43; Mestre Ginaldo, RL 13.298.
129 See Ficătum, fecătum < fic⊖tum + hepăte, ZRPh 25.615–6; Lat. (h)ēpar im Romanischen, ZRPh 28.435–49; also, because of the relationship between ‘milt’ and ‘liver’, Zu den romanischen Benennungen der Milz, Sitz.-ber. d. Preuss. Akad. d. Wiss., 1917, pp. 156–70, exemplifying blends of splēn and milza and lexical distinction between the milt of man and animal.
130 Iecur belongs to an archaic heteroclite class of IE nouns with stems ending alternately in -r and -n; of this class there are only a few vestiges in Latin, e.g. iter -ineris ‘way’, femur -inis (beside -oris) ‘thigh’, cruor ‘blood’ beside cruentus 'bloody, as(s)er ‘blood’ (LEW3 1.72). On the declension of this type, see M. Leumann in Stolz-Schmalz, Lat. Gr.5 238–9; F. Sommer, Handbuch der lateinischen Laut- und Formenlehre3 355 (Heidelberg, 1914); A. Ernout, Morphologie historique du latin2 73–4 (Paris, 1927); LEW3 1.673 (with bibliography up to the year 1936). It will be noticed that most of the Latin words belonging to this group refer to the human body, to which an old lexical stock corresponds in most languages, and that they have left few if any traces in Romance. (Special conditions that may have favored the disappearance of femur have been suggested by W. von Wartburg, Die Ausdrücke für die Fehler des Gesichtsorgans in den romanischen Sprachen und Dialekten 2 [Diss. Zurich, 1912]).
131 To the bibliography on hēpar given by REW3 §4108 add D.S. Blondheim, Les parlers judéo-romans 53–4 (Paris, 1925). On hēpar in Latin, see the excellent material offered by ThLL 6.2609–10 (1938), which invalidates some of Schuchardt's surmises (e.g. the evidence of the use of hēpar as connoting the liver of animals); compare also the exhaustive data on such derivatives as hēpatia, hēpatiārius, hēpatiās, hēpatītēs (-tītis), hēpatizon, hēpatus, showing that the word was propagated by naturalists rather than by physicians exclusively. On hepático > Ptg. pátigo see RFH 2.420–1.
132 For reasons not yet transparent, rēn ‘kidney’ did persist in Romance, see Zauner, Rom. Nam. d. Körpert. 517–9; REW3 §§7206, 7209, 7209a, 7210, 7213. For lien and splēn, see LEW3 1.799 : splēn is recorded since Vitruvius, splēnicus < σπληνικός since Pliny (echoed by liēnicus in Caelius Aurelianus beside genuine Latin liēnōsus ‘splenetic‘), splēnīticus and splēniacus are more recent. Stomachus does not seem to have inherited the primary connotation of στόμαχος, see E. Boisacq, Dictionnaire étymologique de la langue grecque3 915 (Heidelberg-Paris, 1938).
133 Except in Roumanian, Macedo-Roumanian, and the dialects of Southern Italy, which absorbed the Hellenisms in the Middle Ages; see Zauner, op. cit. 346, 394, 401, 418, 456, 512–6. On gnathus ‘jaw, cheek’, which from Italy spread to France, see H. R. Kahane, Bezeichnungen der Kinnbacke im Galloromanischen 60–3 (Jena-Leipzig, 1932), and more recently Designations of the Cheek in the Italian Dialects, Lang. 17.212–22. Even for physical defects the words traceable to Greek are few; Wartburg, Ausdrücke für die Fehler des Gesichtsorgans 45, 51–2, records only strabō < στραβός ‘squint-eyed’.
134 For full documentation of the following statements, see ThLL 7.244–6 (1935).
135 See W. A. Bährens, Sprachlicher Kommentar zur vulgärlateinischen Appendix Probi 38 (Halle, 1922).
136 For instance, aequor ‘plain, (surface of the) sea’; angor ‘anguish’, decus ‘ornament’, ebur ‘ivory’, faenus ‘interest’, foedus ‘treaty’, iubar ‘morning-star’, murmur ‘roaring, rustling’, scelus ‘crime’; cf. shift acer > *acrus ‘sour’ (REW3 §92), preference given to gibbus over gibber ‘hunchback’ (REW3 §3755), and the sweeping elimination of genus ‘kind’ (REW3 §3738) and later ‘brick’ (REW3 §4924).
137 For instance, arbor ‘tree’ (REW3 §606), carcer ‘jail’ (REW3 §1679), cicer ‘chick-pea’ (REW3 §1900), cinis ‘ashes’ (REW3 §1929), femur ‘thigh’ (REW3 §3240), fulgur ‘lightning’ (REW3 §3555), fundus ‘ground’ (REW3 §3585), furfur ‘bran’ (REW3 §3595), guttur ‘throat’ (REW3 §3930), latus ‘side’ (REW3 §4934), lepus ‘hare’ (REW3 §4991), marmor ‘marble’ (REW3 §5638), mormyr ‘species of fish’ (REW3 §5686), opus ‘work’ (REW3 §6079), papāver ‘poppy’ (REW3 §6210), passer ‘sparrow’ (REW3 §6268), pignus ‘pawn’ (REW3 §6490), rōbur ‘strength’ (REW3 §7354), stercus ‘manure’ (REW3 §8245), sulphur ‘brimstone’ (REW3 §8443), tempus ‘time’ (REW3 §8634), tūber ‘bulb’ (REW3 §8966), turtur ‘turtle-dove’ (REW3 §9009), ūber ‘udder’ (REW3 §9026), vōmer ‘ploughshare’ (REW3 §9448), vultur ‘vulture’ (REW3 §9466), and others.
138 Cf. ebur beside ebor, rōbur beside rōbor.
139 Another gloss in the same codex (C. G. L. 2.325) has preserved iecorālis 'ήπατικός'. Patterned after Hellenisms like arthriticus, nephriticus is the medical term iecoriticus ‘qui iecoris vitio laborat’ (Marcellus Medicus). Its synonym iecorōsus has been encountered in Caelius Aurelianus, Sidonius Apollinaris, and the Latin version of Dioscorides. Iecunānum has been amended by Heraeus into iecunārium ‘victimārium’. Iecūsculum, a diminutive shaped after corpūsculum ‘small body’, enjoyed considerable currency. Iocinerōsus ‘suffering from a liver disease’ belonged to the Imperial Latinity. It is significant that most of the terms pertaining to medicine could not withstand the pressure of hēpar, whereas iequāria, a culinary term, persisted in the vernacular.
140 This has been particularly stressed by H. R. Kahane, Italo-Byzantinische Etymologien: Scala, Byz.-Ngrch. Jb. 16.33–58, and by Renée (Toole-)Kahane, Italienische Marinewörter im Neugriechischen, AR 22.510–82.
141 On Greek words in Catalan, especially maritime terms, see J. Coromines, Les relacions amb Grècia reflectides en el nostre vocabulari, Homen. Rubió y Lluch 3.283–316 (Barcelona, 1936). On Greek elements in the variety of Medieval Latin used by Spanish clerics see L. Spitzer's review of A. Castro, Glosarios latino-españoles de la edad media, MLN 53.122–46. As has recently been observed (C. C. Rice, Lang. 19.154–6), the number of Romance words traced back to Greek, both ancient and medieval, increases constantly, but attempts to classify them systematically, according to provinces or to semantic groups (as has been done for the inverse infiltration by H. Zilliacus, Zum Kampf der Weltsprachen im oströmischen Reich, [Helsinki, 1935]), remain to be made. Attention has recently been given by E. Löfstedt to the penetration of Greek syntactic constructions into Romance vernaculars, e.g. in Lateinisch-Griechisches und Lateinisch-Romanisches, St. Neoph. 11.173–85.
142 The rules of modern spelling as used in Portugal have been applied uniformly to titles of books throughout the article, but quotations have been reproduced faithfully from their respective sources, without eliminating orthographic inconsistencies.
The author wishes to express his gratitude to the Library of the University of California for repeated help in procuring the necessary material.