Annotated bibliography:
syllable-final /S/ in Spanish

Random audio examples (broadcast news)

Lloyd, Paul M. From Latin to Spanish. American Philosophical Society, 1987.

The preponderance of open syllables in Spanish from earliest times, and the drive to make all syllables as open as possible, has had a continuous effect on syllable-final consonants. [...]

Related, in part at least, to the preceding phenomenon is the weakening of syllable-final /-s / into an aspiration which may eventually become so weak that it disappears. The earliest written examples so far are a number of words from the late fifteenth century which omit the final -s of the plurals, e.g., mandamo (1467), juego vedados, todas la otras (1499) (Frago Gracia 1985, 298). A frequently cited form is from an autograph note by Fernando Colon, the son of the admiral, in which the name Sophonisba is written Sofonifa (Lapesa 1980, 387). The F written for SB may reflect both the aspiration of the /-s/ > [-h] and its subsequent unvoicing of the voiced bilabial fricative, to [ɸ]. Increasing evidence of a weakening of the articulation of /-s / is found in documents from the end of the sixteenth and the beginning of the seventeenth centuries . In 1550 is found para que vo lo digan with vo for vos (Frago Gracia 1985, 298). In 1575 a letter written by a musician from Toledo spells muestra twice as muetra. Twenty years later another manuscript omits final /-s / in a number sof words: la puertas, los maestrasgo, la casas, etc. Letters written by persons from Seville in America show a loss of final / -s / from the middle of the sixteenth century onward: los quale (1556), soy for sois (1560), vos enbiaste (1560) for vos embiastes, decanso, decisey, quedavadi, etc. for descanso, dieciséis, and quedábades (1568-69). Such examples indicate that the weakening and loss of final /-s / so characteristic today of Andalusia and much of Spanish America, must have been well established at least among the lowest social classes by the end of the sixteenth century . Undoubtedly more examples can be found by further examination of early documents.

Politzer, Robert, "Final -S in the Romania."

Book chapter published in 2018, but seems to date from 1947:
Politzer, Robert L. "Final-s in the Romania." Romanic Review 38, no. 2 (1947): 159.

Hamp, Eric P. "Final-s in Latin." Classical Philology 54, no. 3 (1959): 165-172.

Marotta, Giovanna, and Lucia Tamponi. "Omission of Final‐S In Latin Inscriptions: Time and Space." Transactions of the Philological Society (2018).

The paper focuses on the omission of final -s in Latin inscriptions found in the corpus CLaSSES (Corpus for Latin Sociolinguistic Studies on Epigraphic textS, http://classeslatin-linguistics.fileli.unipi.it) from 350 BC till 50 AD.

The distribution of final -s has been examined in the endings of the second declension with respect to several variables, such as the dating and the place of provenance of the inscription.

The results show that the restoration of the grapheme spread from Rome starting from the mid-third century BC, when the city was the political and economic centre of the conquered territories. Therefore, the Latin of Rome was perceived as prestigious by Roman citizens. The process of -s restoration was completed by the mid-first century BC and appears to be strictly related with the raising o > u in final syllable. Therefore, the omissions of final -s found in inscriptions outside Rome in the second and first century BC can be considered to be residuals of the ancient pronunciation in areas not yet completely reached by the innovation, whereas in Rome they can be considered as forms of a low sociolinguistic level.

Alonso, A. "Una ley fonológica del español". Hispanic Review XIII. pp. 91-101, 1945.

Navarro, Tomás. El español en Puerto Rico; contribución a la geografía lingüística hispanoamericana. Río Piedras, Univ. de Puerto Rico, 1948.

La s final de sílaba se aspira de manera general en Puerto Rico, como en la mayor parte de Hispanoamerica, en Andalucía. y en otras provincias del sur de Espana. La aspiración ocurre en Puerto Rico en el lenguaje de toda clase de personas y lugares, lo mismo en las tierras altas que en las bajas y de igual modo en los centros importantes de población que en los más retirados barrios rurales. Las personas instruídas dan a la aspiración de la s final de sílaba una forma relativamente regular, dentro de la inconsistencia de este sonido. La pronunciación popular lo somete a todas las transformaciones dadas a conocer en otros estudios de esta especie. Entre los cambios considerados como deformaciones de la ortologia normal, la aspiración de la s es tenida en concepto de inferioridad. Personas instruídas que aspiran la s en la conversación ordinaria, la pronuncian con su sonido propio en conferencias y actos académicos.

Delante de p, t, c (k), la aspiración de la s reduce y atenúa su sonido: respeto, pestaña, pescar. En las palabras de esta clase con es- inicial en principio de grupo, no solo la s sino toda la silaba se apaga hasta un grado casi imperceptible. Cuando la s forma parte de la silaba acentuada, la aspiración, en el habla
popular, se asimila más o menos al punto y modo de articulación de la oclusiva que la sigue. Seda exagerado decir que vocablos como abispa, cresta, casco, se convierten en abippa, cretta, cacco. En realidad se. percibe siempre cierto resto de la aspiración entre la vocal acentuada y la oclusion siguiente. En posición inacentuada se atenúa y debilita el efecto de la mencionada asimilación.

Ante las consonantes b, d, g, la aspiración de la s se sonoriza y abrevia sin comunicar a la consonante inmediata su sordez originaria. Son desconocidas, fuera de ejemplos de caracter individual, las combinaciones que resultan en algunas comarcas españolas de la fusión de las consonantes s-b, s-d, s-g, en enlaces como los que ofrecen, por ejemplo, las bueyes, tres dedos, sus gallinas. En algunos sujetos la aspiración sonora aparecia asimilada a la b fricativa siguiente formando una suave consonante doble: lob-braso, los brazos (Trujillo Alto, San Sebastian, Rosario Alto). El grupo s-g apareció bajo la forma de una atenuada velar sorda, laj-jayina, las gallinas, en Barros y Vieques.

Delante de l la aspiración de la s suele oirse con timbre sordo, pero más frecuentemente se sonoriza o se asimila en mayor o menor grado a la l siguiente. En la palabra islita, los ejemplos registrados con aspiración sorda fueron 6; los que presentaron aspiración sonora, 16; los intermedios entre aspiración y asimilación a la l, 6, y los que aparecieron con clara asimilación, il-lita, solamente 3. Hubo ademas 8 sujetos en los dos extremos del pais, Naguabo, Pefia Pobre, Humacao, Maunabo, Salinas, Guama, La Parguera y Caho Rojo, los cuales, en lugar de islita prefirieron invariablemente el indigenismo cayo.

etc. ...

Alonso, D. "Sobre la -s final de sílaba en el mundo hispánico". Enciclopedia de Lingüística Hispánica,, Suplemento: La fragmentación fonética peninsular. Madrid: Consejo Superior de Investiagciones Científicas pp. 47-53 (1962).

Beym, R. "Porteño /s/ and [h] [ɦ] [s] [x] 0̸ as variants", Lingua 12. 1963

Roxana Ma and Elenor Herasimchuk, "The Linguistic Dimensions of a Bilingual Neighborhood," in Bilingualism in the Barrio, eds. J. A. Fishman, R. L. Cooper, and R. Ma (Bloomington, Indiana: Indiana University Press, 1972):

Cedergren, Henrietta Cecilia Jonas. The interplay of social and linguistic factors in Panama. PhD Dissertation, Cornell University, 1973.

p. 40:

p.41:

Terrell, Tracy D. "Final /s/ in Cuban Spanish." Hispania (1979): 599-612.

The phonological processes of aspiration (s → h) and deletion (s → 0) of / s / in syllable final (mosca 'fly') and word final (los niños 'the children') position affect so many segments that the speech of Spanish speakers of areas in which these processes are used produces an overall phonetic impression quite different when compared with dialects in which these processes are not normally used. [...]

Certain methodological observations are in order. It was my intention to distinguish a variety of phonetic manifestations of /s1/.5 However, it became quickly apparent that such a task, theoretically so simple, on a practical basis was impossible. It is imperative in any science to demand that others be able to replicate the results of any investigation. Replicability of the results would have been very difficult to achieve with a fine transcription. For this reason, the following system was selected.

s: all phones with some sibilance.
0: complete absence of a phone representing /s/.
h: normally aspirated, sometimes very weak, often voiced or nasalized and possibly assimilated resulting in a geminate consonant cluster.

It should be noted that this is essentially the same system used by others who have done quantitative studies of Spanish phonology.6

__________________________

5Very detailed studies of the phonetic output of the rule of aspiration are found in Tomás Navarro-Tomás, El español en Puerto Rico (Rio Piedras, 1948)

6See also Ma and Herasimchuk, op.cit. and Cedergren, op. cit. Cedergren included glottal stop as a manifestation of the 0 category whereas I included it in the h group. This must be kept in mind when making direct comparisons of deletion rates of the two studies.

 

Poplack, Shana. "The notion of the plural in Puerto Rican Spanish: Competing constraints on (s) deletio." Locating language in time and space, 1980:

Puerto Rican Spanish (s) is variably subject to two weakening processes, aspiration and deletion, so that a phrase such as las cosas bonitas. 'the pretty things' can also be realized [lah 'kosah bo' nitah] or [Ia 'kosa bo' nita].

Lipski, John M. "Reduction of Spanish word-final /s/ and /n/." Canadian Journal of Linguistics 31, no. 2 (1986): 139-156.

Widdison, Kirk. "Two models of Spanish s-aspiration", Language Sciences 17(4) 1995.

Sayahi, Lofti. "Final /s/ retention and deletion in Spanish: The role of the speaker's type of competence". Language Sciences 2005.

Lynch, Andrew. "A sociolinguistic analysis of final /s/ in Miami Cuban Spanish". Language Sciences 31(6) 2009.

This study analyzes the variation of syllable- and word-final /s/ among two generations of Cubans in Miami, Florida (USA): older, early exile immigrants who arrived in Miami as adults in the 1960s and 1970s, and young Miami-born Cubans whose maternal and paternal grandparents immigrated to Miamifrom Cuba prior to 1980. Since sibilant weakening is generally considered to be an ongoing language change in Caribbean Spanish, it was hypothesized that the young generation of English-dominant bilinguals would present with much higher rates of aspiration and deletion, in keeping with [Silva-Corvalán, C., 1994. Language Contact and Change. Spanish in Los Angeles. Clarendon Press, Oxford] hypothesis that linguistic changes are accelerated in situations of language contact. However, the data instead revealed significantly higher rates of sibilant retention among young Miami-born speakers, indicative of a ‘reversed’ language change. This finding is attributed principally to the social need of the Miami-born grandchildren of early exile Cubans to differentiate their speech from that of later Cuban immigrant groups, mostly for political and ideological reasons.

[...]

The present study [examines] some of the principal social and linguistic factors implicated in the production of syllable- and word-final /s/. This particular phonological segment presents a high degree of variability in contemporary Spanish which many scholars (Terrell, 1979, Terrell, 1981, Lafford, 1986 and Calero Fernández, 1993, among others) attribute to an ongoing process of diachronic change involving the sibilant (cf. Penny, 2000).

Lipski (1995, p. 291) observed that “in many Spanish dialects, including those of southern and western Spain, the Canary Islands, the Caribbean region and much of Central and South America, syllable- and word-final /s/ suffers frequent weakening.” Canfield (1981) initially relied upon this feature to distinguish grosso modo two major dialect regions of the Spanish-speaking Americas: highlands and lowlands. This apparent geographical division has a sociopolitical and economic basis, according to Penny (2000). He affirmed that, historically, the division “corresponds to the degree of closeness of contact between central Spain and the specific American area concerned: those areas which, because of their political and economic importance in the Empire, attracted prestigious speakers of central Castilian varieties are the ones which retain /-s/ most frequently” ( Penny, 2000, p. 148). In the Caribbean, the influence of speakers of Andalusian varieties of Spanish—in which aspiration and deletion frequently occur—was predominant, while contact with the /s/-retaining central Castilian norm was more distant ( Penny, 2000, p. 149).

With regards to Caribbean Spanish in particular, Ma and Herasimchuk’s (1971) study of Puerto Ricans in New York was the first large-scale, contemporary speech-based project to document the correlation of /s/ weakening and stylistic variation. The first important variationist4 study of /s/ in Caribbean Spanish was that of Cedergren (1973), who analyzed the speech of 79 individuals in Panama City according to a series of linguistic and social factors. With regards to linguistic factors, she found that aspiration was most strongly conditioned by the presence of a following consonant and deletion was most likely to occur before a pause. In social terms, Cedergren observed that aspiration was much more frequent among speakers of the lower socioeconomic classes, and more common among younger speakers in general. The findings of these studies confirmed, from a variationist perspective, the impressionistic observations made by numerous other scholars of Caribbean Spanish earlier in the 20th century, Navarro-Tomás (1948) among them.5 Observing the Spanish of Puerto Rico during the 1920s, Navarro-Tomás remarked that “final /s/ is generally aspirated in Puerto Rico over all social classes and regions. Educated people give this aspirated /s/ a fairly regular form, while the uneducated submit it to variants of pronunciation” (1948, p. 69).

Since the 1970s, there has been a proliferation of synchronic studies of sibilant reduction in contemporary Spanish.

File-Muriel, Richard J., and Earl K. Brown. "The gradient nature of s-lenition in Caleño Spanish." Language Variation and Change 23, no. 02 (2011): 223-243.

Whereas previous studies of Spanish s-weakening have relied on impressionistic coding, the present study examines temporal and gradient acoustic details in the production of /s/ by eight females from Cali, Colombia, during sociolinguistic interviews. We propose a metric for quantifying s-realization by employing three scalar-dependent variables: s-duration, centroid, and voicelessness. The results of linear regressions indicate that the dependent variables are significantly conditioned by local speaking rate, word position, following and preceding phonological context, stress, and lexical frequency. This study sheds light on how each independent variable influences s-realization acoustically. For example, as local speaking rate increases, duration, centroid, and voicelessness decrease, which is indicative of lenition, and the same weakening tendency is observed when /s/ occurs in word-final position or is followed by a nonhigh vowel, whereas frequency contributes only to s-duration. We discuss the advantages of opting for instrumental measurements over symbolic representation.

Brown, Earl K., Michael S. Gradoville, and Richard J. File-Muriel. "The variable effect of form and lemma frequencies on phonetic variation: Evidence from/s/realization in two varieties of Colombian Spanish." Corpus Linguistics and Linguistic Theory 10, no. 2 (2014): 213-241.

Bullock, Barbara E., Almeida Jacqueline Toribio, and Mark Amengual. "The status of s in Dominican Spanish." Lingua 143 (2014): 20-35.

Torreira, Francisco, and Mirjam Ernestus. "Weakening of intervocalic/s/in the Nijmegen Corpus of Casual Spanish." Phonetica 69 (2012): 124-148.

Bradley, Travis G. "Optimality Theory and Spanish Phonology." Language and Linguistics Compass 8, no. 2 (2014): 65-88.

Parrell, Benjamin. "The role of gestural phasing in Western Andalusian Spanish aspiration." Journal of phonetics 40, no. 1 (2012): 37-45.

Mack, Sara. "A sociophonetic analysis of/s/variation in Puerto Rican Spanish." In Selected Proceesings of the 11th Hispanic Linguistics Symposium. Somerville, MA: Cascadilla. 2011.

There are three traditionally recognized variants of /s/ syllable- and word-finally in Caribbean Spanish: the sibilant [s], a fully retained variant; the [h], commonly known as the aspirated variant; and the phonetic zero, also known as the deleted or elided variant (Alba, 1990; Cedergren, 1978; Lafford, 1986; López Morales, 1980 & 1983). 2 In some speech communities, /s/ weakening has advanced to the stage where /s/ epenthesis occurs; in a discussion of standard and local linguistic norms in Dominican Spanish, Zentella (2003) cites the comic, often mocking phrase “hablar fisno,” a play on the expression hablar fino (‘to speak fine’), in which hypercorrective /s/-epenthesis occurs at the end of the first syllable in the word fino. The phrase hablar fisno is known in several dialects, but is seen as an especially apt commentary on Dominican Spanish. Although the weakening process is very advanced in the dialect, speakers realize that the absence of [s] is stigmatized, and insert [s] in the syllable rhyme in places where there is no underlying /s/, as in the case of abogado ‘lawyer’ > asbogado or abosgado (Nuñez-Cedeño, 1994, p. 30). While hablar fisno is often equated with formality, higher socioeconomic status, or “posh speech” (Roca, 2005, p. 38), it has come to be associated with male effeminacy in some contexts (Zentella, 2003, p. 60).

Martínez-Gil, Fernando, and Sonia Colina, eds. Optimality-theoretic studies in Spanish phonology. Vol. 99. John Benjamins Publishing, 2007.

Lynch, Andrew. "A sociolinguistic analysis of final/s/in Miami Cuban Spanish."Language Sciences 31, no. 6 (2009): 766-790.

Penny, Ralph John. A history of the Spanish language. Cambridge University Press, 2002.

Carreira, María. "Validating and promoting Spanish in the United States: Lessons from linguistic science." Bilingual Research Journal 24, no. 4 (2000): 423-442.

Castilian has its humble origins in the north-central part of the Iberian Peninsula in what is now the northern part of the Province of Burgos. Because of its remote location, this version of vulgar Latin departed more from the prestigious Roman variant than the Iberian variants that had more direct access to the capital of the Roman empire. In its state of isolation, Castilian created new lexical items and retained older linguistic itemsthat fell out of use in Rome and parts of the Empire in closer contact with Rome. Many of these "incorrect" features of speech were perpetuated as the Fall of the Roman Empire all but eliminated any linguistic corrections that may have come from Roman grammarians. Penny (1991) notes:

Varieties of Hispano-Romance speech which were hitherto peripheral (in both geographical and linguistic terms) are extended southwards at the expense of those varieties which one can presume were previously the most prestigious and the most in keeping with the Romance spoken outside the Peninsula. And among these peripheral varieties of Hispano-Romance, it was one of the most "abnormal," namely Castilian, which was to have the greatest territorial and cultural success. (p.13)

Fox, Michelle Minnick. "Usage-based Effects in Latin American Spanish: Syllable-final /s/ Lenition."

ABSTRACT: Previous studies have identified factors that contribute to the weakening and deletion of syllable-final /s/ in Latin American Spanish, including the dialect and sex of the speaker, the phonetic environment and grammatical status of the /s/, and functional considerations. Proponents of a usage-based model of language claim that the structure of language is shaped by how it is used, and therefore factors such as word frequency, word predictability, and the context that words appear in have an effect on the form of language. This dissertation investigates the extent to which these usage-based factors contribute to syllable-final /s/ lenition in addition to the previously identified factors. Automated speech recognition methods were used to code three dependent variables for a corpus of over 50,000 tokens of syllable-final /s/: deletion or retention of /s/, duration of retained /s/, and the spectral center of gravity of retained /s/. Multiple regression was performed for each of the dependent variables, on all of the data combined and on several subsets of the data. For each multiple regression, usage-based factors were added to a base model to determine which of them improve the model. Word frequency and word predictability based on the following word both have an effect in the expected direction, with more frequent and more predictable words having higher levels of lenition. Word predictability based on the preceding word has the opposite effect, with more predictable words having lower levels of lenition. The phonetic context that words appear in most frequently also has an effect, with words that are more often followed by a consonant having more advanced lenition, even after taking into consideration the actual phonetic context. These usage-based factors contribute both to the categorical process of deletion and the gradient processes of shortening and weakening of articulation. This data supports the claim that these usage-based variables form part of the speaker's knowledge and that speakers have knowledge of low-level phonetic detail. This study suggests that the extent of lenition may be determined both within the lexical entry as in an exemplar model and by processing during production.

Ryant, Neville, and Mark Liberman, "Large-scale analysis of Spanish /s/-lenition using audiobooks", ICA 2016.