Author:
Luiz Carlos Schwindt Federal University of Rio Grande do Sul, Brazil

Search for other papers by Luiz Carlos Schwindt in
Current site
Google Scholar
PubMed
Close
https://orcid.org/0000-0003-0533-589X
Free access

Abstract

This paper deals with the underlying representation of [w]-final words in Brazilian Portuguese, usually spelled with <l> and pronounced as [w] and occasionally as [ɫ] (e.g., papel / [paˈpɛw] ~ [paˈpɛɫ] ‘paper’). It focuses on non-verbs derived by a vowel-initial suffix preceded by [l] (e.g., papel+eiro ‘paper+suffix’ / [papeˈlejɾʊ] ‘papermaker’; cliente+ismo ‘client+suffix’ / [kliẽnteˈlizmʊ] ‘patronage’). The results from a pseudoword task answered by 219 participants contrasted to lexicon data from Corpus Brasileiro show that native speakers associate such derived forms with bases already containing [l] in the last syllable, either in the onset or the coda position. This observation is interpreted in a constraint-based approach, with the assumption that a demand for alignment between vowel-initial suffixes and roots closed by /l/, along with the requirement for phonological correspondence between base and derivative, is highly ranked in a grammar that accounts for learning morphophonological representations in the language.

Abstract

This paper deals with the underlying representation of [w]-final words in Brazilian Portuguese, usually spelled with <l> and pronounced as [w] and occasionally as [ɫ] (e.g., papel / [paˈpɛw] ~ [paˈpɛɫ] ‘paper’). It focuses on non-verbs derived by a vowel-initial suffix preceded by [l] (e.g., papel+eiro ‘paper+suffix’ / [papeˈlejɾʊ] ‘papermaker’; cliente+ismo ‘client+suffix’ / [kliẽnteˈlizmʊ] ‘patronage’). The results from a pseudoword task answered by 219 participants contrasted to lexicon data from Corpus Brasileiro show that native speakers associate such derived forms with bases already containing [l] in the last syllable, either in the onset or the coda position. This observation is interpreted in a constraint-based approach, with the assumption that a demand for alignment between vowel-initial suffixes and roots closed by /l/, along with the requirement for phonological correspondence between base and derivative, is highly ranked in a grammar that accounts for learning morphophonological representations in the language.

1 Introduction

This paper addresses a controversy regarding the underlying representation (UR) of the final segment of nouns and adjectives that are mostly spelled as <l> and usually produced as [w] in Brazilian Portuguese (BP), such as papel [paˈpɛw] ‘paper’.

The phenomenon, apparently strictly orthographic or phonetic, has important morphological and phonological implications and can be approached from a less or more abstract perspective. From the first perspective, the UR (when some kind of UR is admitted) is the segment commonly attested in speech, namely, the labiovelar approximant /w/, whereas from the second perspective, the UR is the alveolar lateral consonant /l/.

The argument to sustain the first perspective is quite obvious even in a formalist approach; it concerns the principle of Occam’s razor: entities are not to be multiplied beyond necessity. However, there are at least three types of evidence in favor of a more complex approach to this problem, which are described as follows:

  1. (i)Non-vocalized alternants are also attested in the coda position, word-finally or medially, in some BP dialects, particularly in southern Brazil, as shown in Fig. 1.1 The preferred non-vocalized alternant is a velarized lateral consonant, but alveolar laterals and deletion are also attested in this context.

a.pa[w]co ~ pa[ɫ]co ~ pa[l]co ~ pa∅co‘stage’
b.aventa[w] ~ aventa[ɫ] ~ aventa[l] ~ aventa∅‘apron’
Fig. 1.
Fig. 1.

Non-vocalized /l/ codas in southern Brazil dialects (Collischonn & Quednau 2010, 142)

Citation: Acta Linguistica Academica 68, 1-2; 10.1556/2062.2021.00482

It should be noted that the data from the corpus VARSUL analyzed by Collischonn & Quednau (2010) are from the 1990s. The authors also found the greater use of vocalization among informants under 50 years of age, suggesting a possible change in progress. Despite this, the fact that the process is not yet categorical in BP cannot be neglected, especially when the debate about its UR is at stake.2 Nevertheless, for the sake of simplicity, the surface forms analyzed in this text will be generically referred to as [w]-final, as [w] is the segment found in the majority of the data.

  1. (ii)Words closed by underlying /w/ (e.g., pau ‘stick’, céu ‘sky’, degrau ‘step’, chapéu ‘hat’, mausoléu ‘mausoleum’), although few in number in the BP lexicon, never exhibit such a segment as [ɫ]; they also systematically select the regular allomorph -s as the plural marker. Words supposedly closed by underlying /l/, which realize the final segment as [w] ~ [ɫ], form the plural by adding [js] instead.

URsingularplural
a.mausolé/w/mausolé[w] ~ ∗mausole[ɫ]mausolé[ws] ~ ∗mausolé[js]‘mausoleum/s’
b.hote/l/hote[w] ~ hote[ɫ]hote[js] ~ ∗hote[w]s‘hotel/s’

There is no consensus in the literature about the morphological status of the palatal approximant in [js]. It is considered by some authors as a product of base allomorphy triggered by plural affixation (Camara 1969; 1970; Morales-Front & Holt 1997; Mateus & D'Andrade 2000; Schwindt & Wetzels 2016; among others) or as part of a suppletive suffix by others (Cristófaro-Silva 2012; Becker et al. 2018; among others). Regardless of this interpretation, the alternation shown in (2b) remains almost categorical in all BP dialects.3

  1. (iii)An alveolar lateral consonant emerges before vowel-initial derivational suffixes. Less frequently, this consonant can also emerge before inflectional suffixes and in external sandhi contexts.

a.suffix derivation:pape[w] / pape[l]aria‘paper’ / stationary store’
b.plural inflection:ma[w] / ma[l]es‘evil/s’
c.external sandhi:ma[w] educado~ma[l]educado‘ill mannered’

In the domain of suffix derivation, this process is categorical with [w]-final bases, although an epenthetic [l] may also emerge in a few bases not containing any synchronic representation for /l/ or /w/.

a.chá ‘tea’ / cha[l]eira‘kettle’
b.cliente ‘client’ / cliente[l]ismo‘patronage’

Basing especially on evidence (iii), this paper argues that the ideal UR for these words in BP is a form closed by an abstract root-final /l/. This is interpreted in a ranking of constraints within the Optimality Theory framework. The assumption is that an alignment constraint imposes that vowel-initial suffixes attach to roots, not words, and a paradigmatic constraint requires phonological correspondence between roots and derivatives, which accounting for the emergence of [l] in the vast majority of this type of derived word in the language. The empirical support for this thesis comes from the results of an experiment in which native speakers had to relate the derived pseudowords to pluralized forms. The main objective is to characterize the ideal base patterns behind the predominant plural patterns.4 In addition, data from a large BP corpus have been prepared to function as a reference sample whenever a comparison with a usage lexicon is necessary.

The rest of the paper is organized as follows. A brief overview of some of the most relevant approaches to the phenomenon discussed here is presented in Section 2. The method for constituting and analyzing data from the experiment, as well as the lexicon sample, is presented in Section 3. The main quantitative results are presented and discussed in Section 4, and the constraints involved in the selection of the singular bases corresponding to the plural forms are shown in Section 5. In the last section, the main findings of the research and its limitations are summarized.

2 Background

As mentioned above, the analysis of the [w]-final base representation in BP is related to the discussion about the morphological process of irregular plural inflection and the phonological phenomenon of consonant epenthesis. In other words, the admission of [w] as the final segment of these bases implies assuming -js as a plural allomorph of -s and imposes the assumption that [l] is epenthetic or part of the affix in words derived by suffixes frequently interpreted as initiated by vowels, such as -eiro/a or -ismo. On the other hand, the consequence of assuming an abstract /l/ as the UR for most of these bases is the need to account for the emergence of [j] in the plural and the /l/-vocalization process in the singular, as this segment frequently surfaces as [w]. This section provides a summary of some referential studies on these topics, with the sole purpose of contextualizing the problem.

2.1 Irregular plurals in BP

In BP, plurals are mostly formed by adding -s at the right edge of the word (e.g., carro / carros ‘car/s’). There are two main sets of irregularities in plural formation: words ending in a nasal diphthong (e.g., ladrão / ladrões ‘thief / thieves’) and words ending in an oral diphthong or [Vɫ] (e.g., lençol / lençóis ‘sheet/s’); the latter is the phenomenon addressed in this paper. The productivity of irregular plurals in BP has been described in different papers, including those of Huback (2010a; 2010b), Rizzato (2018), Schwindt et al. (2021).

Representing the irregularity of these plurals includes assessing whether -ões and -js are allomorphs of -s (Cristófaro-Silva 2012; Becker et al., 2018; and others) or whether the allomorphy affects the bases instead, with the plural suffix being unchanged (Camara 1969; 1970; Morales-Front & Holt 1997; Schwindt & Wetzels 2016; and others).

Cristófaro-Silva (2012) argues from the perspective of usage-based models that the representation of irregular plurals in BP is directly affected by lexical frequency. The author states that the plural morpheme |is| is attached, among others, to [w]-final bases, which suggests no allomorphy, strictly speaking. Moreover, the final segment of these bases is considered a vocalized lateral consonant in the paper, suggesting some level of abstraction in the representation of singular forms. In an analysis from the perspective of language acquisition, Becker et al. (2018) also assume that -is is a plural allomorph. Their study investigates the alternation [w ~ j] in plural inflection, taking /w/ as the UR’s final segment. The data are from a pseudoword task answered by children and a control group consisting of adults. From the results, the authors observed some avoidance of the alternation in monosyllables and in new diphthongs with low dispersion between vowel and glide (e.g., [ej, ew, oj, ow]), although these forms are present in the lexicon. Data are analyzed from the perspective of the MaxEnt model. A positional faithfulness constraint guarantees the preservation of monosyllables, interpreted as initial syllables. The distinction between diphthongs with lax and tense vowels is ensured by comparative markedness: diphthongs [ej, oj] are tolerated, and they surface faithfully when they already exist in underlying forms, whereas new instances are blocked.

Camara (1969; 1970) assumes only one plural marker in BP, -s, and attributes the irregularity discussed here to morphophonemic rules, which convert a theoretical theme vowel into a palatal glide in the plural context (e.g., ∗animalesanimaesanima[j]s ‘animals’). The final segment of the base is /l/, although the author admits that in dialects in which [w] is used /l/ becomes /w/ instead of being deleted, and the alternation required by the plural is therefore w/j. Likewise, Morales-Front & Holt (1997), in a constraint-based approach, assume /l/-final bases and -s as the plural marker in the input representation. They argue that the alveolar lateral consonant is forced into the nucleus leftward by a syllable structure constraint when it is followed by a tautosyllabic consonant, which is -s in this case. The change of value of the feature [lateral] is the least marked option, ranked above [l]-deletion or vowel insertion. For Schwindt & Wetzels (2016), the problem with this analysis is that the creation of a nucleus demanded by the plural marker seems to be unmotivated in dialects that already vocalize this segment in final position of singular forms.5 Instead, the most appropriate analysis seems to be to derive the plural by a general rule of coda vocalization and assimilation of the glide to the coronal place of the plural suffix (e.g., ane/l/ → ané[w] ~ ané[ɫ] → ané[js] ‘ring/s’). /l/-vocalization, a process attested in different languages of the world, is described in a relatively simple way in autosegmental terms. Costa (2004), based on Walsh Dickey (1997), argues that [ɫ] is a product of adding a [dorsal] feature to the dorsal node, whereas [w] is a product of the loss of the feature [coronal] and, consequently, the loss of the laterality of the segment. However, the sonorant element created by this process cannot be licensed because, in principle, it is not part of the Portuguese system, causing the labial node to be added to the consonant node point. As it is a segment with two points of articulation, the theory predicts that the delinking of [coronal] produces a dorsal segment, whereas the delinking of [dorsal] produces a coronal segment.

2.2 Consonant epenthesis

Admitting surface-based singular forms to the irregular plurals discussed in this paper imposes accounting for the emerging [l] in derived forms. In other words, the analysis should explain how from a base such as jorna[w] ‘newspaper’, a form such as jorna[l]ismo ‘journalism’ is derived. For most approaches starting from less abstract representations, derived forms are not a good argument to support that these bases are closed by /l/, as this consonant can emerge even in words that do not contain any [l] in the final coda (e.g., chá ‘tea’ / cha[l]eira ‘kettle’). In these approaches, the alternatives for dealing with the emerging [l] are to consider it part of the suffix or epenthetic.

The first alternative involves lexicalizing forms, such as -leiro and -lismo (and many other vowel-initial suffixes), in parallel to the canonic forms of these suffixes without [l]. Although there are good reasons to imagine that speakers can store forms based on their frequency in languages, justifying the split of lexical entries requires, at the least, that these alternative forms are productive, and no other mechanism in grammar better describes the emergence of [l].

Epenthesis is also a controversial option to deal with this problem. Phonetic segments without any morphological affiliation inserted in words are often interpreted as strategies to repair violations of syllable structure constraints. Cagliari & Massini-Cagliari (2000), in an Optimality Theory analysis, state that consonant epenthesis in BP is mainly provoked by the violation of Onset and *Hiatus, in addition to being restricted to sonorant segments. With respect to vowel-initial suffixes, however, epenthesis is not categorical in BP, and both hiatus (e.g., rato ‘mouse’ / ratoeira ‘mousetrap’) and the syncope of the final vowel of the base (e.g., atleta ‘athlet’ / atletismo ‘athletics’) can arise. Canfield (2018), analyzing words suffixed by -al, -ada, and -eiro, as well as Pires (2016), analyzing words closed by -ada, both using data from the Houaiss Dictionary, find that [z] and [ɾ] are the preferred epenthetic consonants in BP. They also note that the stress of the last vowel of the base can determine the choice between these two consonants: athematic words, closed by stressed vowels, usually select [z] (e.g., +ada ‘shovel+suffix’ / pazada ‘hit with a shovel’), whereas thematic words, closed by unstressed vowels, usually select [ɾ], with the thematic vowel converging to -a in the latter case (e.g., bicho+ada ‘animal+suffix’ / bicharada ‘many animals together’).

By contrast, in the proposal of an abstract representation that departs from /l/ in the base-final position, epenthesis is restricted to a few phonotactically marked contexts, as proposed in Section 5.

3 Method

This analysis starts from the premise that knowledge about word formation in a language must be seen, in addition to the perspective of the lexicon in use, from the perspective of the potential lexicon, that is, considering those words that satisfy all structural requirements to be formed, although they do not exist yet. In view of this, based on a sample extracted from Corpus Brasileiro (CBras),6 a large database available online containing spoken and written language, a decision task experiment using pseudowords was performed. Non-verbal bases derived by the vowel-initial suffixes -eiro/a and -ismo (henceforth -eirV and -ismV) preceded by the consonant [l] (e.g., brasileiro ‘Brazilian’, populismo ‘populism’) are analyzed. The research is limited to only two suffixes in order not to make the experiment too long, ensuring nevertheless the maximum representativeness of the contexts under analysis. For the sake of uniformity, aiming at the comparison between the two samples, the same affixes were examined in the lexicon sample. The particular choice of -eirV and -ismV is justified by the fact that such vowel-initial suffixes are quite productive in BP (Grodt 2009) in addition to having the same number of syllables and forming nouns and adjectives.

Words suffixed by -eirV and -ismV represent 0.7% of all CBras tokens. These suffixes may be preceded by any consonant that can be syllabified as their initial onsets. In most cases, these consonants are part of the root (e.g., papel+eiro ‘stationery’, bol(a)+eiro ‘expert soccer player’) and, occasionally, epenthetic (e.g., chá+l+eira ‘kettle’). Fig. 2 below shows the distribution of these consonants preceding the analyzed suffixes in CBras, initially ignoring their morphological affiliation, only to explore possible patterns of co-occurrence in these cases.

Fig. 2.
Fig. 2.

Previous segment of suffixes -eirV and -ismV – general sample CBras

Citation: Acta Linguistica Academica 68, 1-2; 10.1556/2062.2021.00482

In fact, [l] is a quite frequent consonant before these suffixes, although not evidently distinguished from other alveolar consonants, which are ordinary root-final onsets in BP. Interestingly, the alleged default epenthetic consonant, [z], as mentioned in Section 2, is less frequent in these contexts. What these data show is that the mere frequency distribution is far from constituting sufficient evidence to support the thesis that [l] should be understood as an epenthetic consonant in words derived by -eirV or -ismV, hence the need for a more accurate analysis.

The two samples considered in this study are detailed below. Focus is given to the treatment of the experiment results, with the lexicon sample – which is smaller for having lemmatized data – serving as a support for the exploitation of new forms.

3.1 Experiment

A pseudoword test built on Google Forms and publicly shared for 72 hours on the social network Facebook was answered by 219 BP native speakers between October 22 and 25, 2019. Given the nature of the phenomenon under analysis, informants were not socially stratified, although they initially provided information on their gender and education, including, in cases of higher education, their specific course.7

The task consisted of 24 logatomes created from actual patterns of BP syllable and stress, eight of which were distractors. The pseudowords were presented in their derived form, always with a mono or disyllabic basis preceding the structure [l]+eirV/ismV, in sentences containing a blank space to be filled in by the respective plural form, as exemplified below.

Sentence of the pseudoword test
Acabo de guardar o VALEIRO e agora me dei conta de que três caixas de …………… ficaram em cima da mesa.
‘I just put the VALEIRO away and now I realize that three boxes of …………… have remained on the table.’
(A) vás
(B) váus
(C) váis
(D) váles

As the test was a written one, the informants were warned in the introduction of the form that the stressed syllable in each alternative would be indicated by the diacritics ′ (acute) in case of an open vowel, and ^ (circumflex) or ~ (tilde), in case of closed and nasalized vowels, respectively.8 Nevertheless, differences in the aperture of the stressed mid vowels were not explored in the alternatives; in this case, the option was to use only candidates with open-mid vowels because the strings [ews], [ejs], [ows], and [ojs] are scarce or morphology specific in the Portuguese lexicon.9

The alternatives were presented to the informants in the plural, although the final objective was to analyze their bases in the singular. The motivation was to guarantee the distinction between the alternatives ending in <u> and <l> in the spelling because, as mentioned, their corresponding phonemes are generally neutralized in speech. With distractor items excluded, the sample reached 3,504 tokens.

In view of the purpose of this paper, I focus on the role of the variable plural pattern and the predictive potential of the variables number of syllables and suffix. These variables are exemplified below, accompanied by examples of respective lexicon words that served as matrices for the creation of the respective pseudowords.

Variables (experiment and lexicon samples)
CategoriesSubcategoriesAlternatives/ pseudowordsLexicon matrices
Plural patternCVsváschá/chás‘tea-sing/pl’
CVwsváwspau/paus‘stick-sing/pl
CVjsvájssal/sais‘salt-sing/pl
lVsválessolo/solos‘soil-sing/pl
Number of syllablesthreebe.lei.ragoleira‘goal’
fourgor.te.lis.moparalelismo‘paralelism’
Suffix-eirVsinil+eirobrasileiro‘Brazilian’
-ismVgogal+ismomoralismo‘moralism’
(C = any consonant other than [l]; V = any vowel)

Statistical analysis was conducted in R (R Core Team 2014) with the multinom function to perform multinominal regression models, and figures were produced using the package ggplot2 (Wickham 2016).

3.2 Lexicon

A lexicon referential sample was constituted to support the discussion of the experimental results. All non-verbs available in CBras with the structure l+eirV and l+ismV were lemmatized by applying several filters, resulting in a sample consisting of 683 lemmas and 1,535,958 tokens. Prefixes, as well as the first part of compounds, were deleted whenever the isolated base could be identified as common to a family of words. In addition, identical items, or items showing small divergences in spelling, were combined, and their frequencies were recalculated after that. Finally, derived words whose bases end in a consonant, such as doleiro ‘money changer’ < dólar ‘dollar’, were also excluded from the sample, as they do not present a potential context for [l]-epenthesis (e.g., ∗dolarleiro). The items were categorized according to variables similar to those considered in the experiment analysis.

4 Results and discussion

In this section, the results of the experimental analysis are presented and discussed. The main objective is to analyze the distribution of the four plural patterns under analysis in the test and the CBras data, with the lexical frequency effects being considered. The variables number of syllables and suffix are also examined as potential predictors of the plural patterns.

4.1 Plural patterns

As mentioned, four possible plural alternants, all starting with a consonant, were presented to the informants. These are the most frequent syllable patterns in the Portuguese lexicon: CVs, CVws, CVjs, and lVs.

The Fig. 3 below shows that the informants of the experiment associate lVs and CVjs to l-suffixed words by a wide margin. CVs and Vws are almost tied, with the latter having a slight advantage.

Fig. 3.
Fig. 3.

Plural patterns for l+suffix words – Experiment

Citation: Acta Linguistica Academica 68, 1-2; 10.1556/2062.2021.00482

The preference for lVs meets the expectations of this research, which is the reason why this context has been intentionally separated from CVs. There seems to be a quite natural motivation for it: [l] is already part of the base and is simply syllabified with the vowel-initial suffix. Therefore, it cannot be interpreted as a favorable context for epenthesis.10 The result for CVjs is also not surprising, as it is the most productive pattern of [w]-final plurals in Portuguese. CVws, on the other hand, is lexically limited in the language. According to CBras, of the universe of [w]-final words, only 1% realize the plural with CVws; all of them, however, are spelled with a final <u>, which is probably a trace of the history of such words.

Derivation, in turn, does not tend to synchronically replace the glide for [l] in these cases (e.g., pa[w] ‘stick’ / pa[wl]eira / ∗pa[l]eira ‘whack’). In the test, the same happens to dauleira, a pseudoword for which CVws exceeds CVjs by a great advantage, second only to lVs, which reinforces the idea that there is no active process in the language capable of converting a glide into a lateral consonant. Finally, CVs, although it is generally the most productive plural pattern of the language, is much less productive in terms of the derived forms under analysis. Furthermore, the fact that this pattern does not present any occurrence for three pseudowords further reduces its general mean. The variables that can contribute to the greater or lesser frequency of each of these patterns are discussed below.

For comparison purposes only, the following Fig. 4 shows the distribution of these same patterns in the lexicon, considering data from CBras. The plural patterns are deduced in this case and are restricted to the l+suffix forms found in the data search. Furthermore, the C preceding the Vjs and Vws in the plural pattern labels is a simplification adopted only to maintain parallelism with the test data, as no onset or other onset structures may occasionally be found in this group of words in the lexicon.

Fig. 4.
Fig. 4.

Plural patterns for l+suffix words – lemmatized sample CBras

Citation: Acta Linguistica Academica 68, 1-2; 10.1556/2062.2021.00482

As observed for the test, the most prevalent plural patterns in the lexicon related to the l+suffix forms are also lVs and CVjs, but with an interesting inversion both in the most and the least numerous pairs.

Examining these patterns from the perspective of lexical frequency is interesting. Although there is little or nothing to say about the CVs and lVs patterns, as they are ordinary syllables in Portuguese, the patterns that can be interpreted as plural markers, CVjs and CVws, show a particular behavior in terms of lexical frequency. While CVjs is much more frequent than CVws in absolute rates, such variants are very similar in terms of the distribution of lexical frequency, a fact that can explain the stability of CVws in the language and justifies testing the productivity of these alternants in new items. This has been shown by Schwindt et al. (2021) when analyzing the lexical frequency distribution of these alternants in quartiles, reproduced below (Table 1).

Table 1.

Irregular plurals and lexical frequency in quartiles – general sample CBras Adapted from Schwindt et al. (2021)

AlternantnInferior limit25%50%75%Superior limit100%Mean
CVjs9,0611142458.5710,897722.2
CVws9612421.350.341,745692.3

4.2 Possible predictors

With the distribution of plural patterns in the test and the lexicon being characterized, a question that may arise is whether the properties of the derived forms can predict them, particularly the number of syllables and the type of suffix.

The Fig. 5 below shows a multinomial logistic regression in which CVs is taken as the reference among the four plural variables because it seems to be the least restricted among them, coinciding with ordinary syllables in the language, the context for a genuine [l] epenthesis. The significant odds ratios are indicated by asterisks, considering a 95% confidence interval.

Fig. 5.
Fig. 5.

Predictors of plural patterns from l-suffixed words (reference CV)

Citation: Acta Linguistica Academica 68, 1-2; 10.1556/2062.2021.00482

4.2.1 Number of syllables

The pseudowords in the test have only three or four syllables. As the suffixes -eirV and -ismV already have two syllables each, pseudowords with three syllables may correspond to plural forms with one (e.g., béis., béus., bés. ← be.lei.ra) or two syllables (e.g., bé.lesbe.lei.ra), whereas pseudowords with four syllables may correspond to forms with two (e.g., go.gáis, go.gáus, gó.gasgo.ga.lis.mo) or three syllables (e.g., go.gá.lesgo.ga.lis.mo).

The number of syllables is particularly relevant in the competition of the patterns lVs and CVjs. The results for this variable show that derived forms with three syllables favor lVs and disfavor CVjs. This means that for a derived form such as CV.lej.ɾV, the best pluralized form is CV.lVs, whereas for a form such as CV.CV.lej.ɾV, the best correspondent plural form is CV.CVjs. It should be noted that plural bases with two syllables are preferable, regardless of whether they are closed by a vowel or a diphthong, as long as they belong to one of these two patterns.11 Assuming that [js] can correspond to /l/ in singular forms, as suggested by traditional analyses, instead of considering the number of syllables as a distinctive property of these two patterns, this quantitative result can be understood as an argument to approach them together as bases containing /l/, as opposed to the two other bases that do not contain /l/, CV and CVw. This will be explained further in Section 5.

4.2.2 Suffix

The suffixes -eirV and -ismV are similarly distributed in the experiment, both in relation to the previous vowels and the number of syllables of their bases. Although there is no obvious reason why a particular suffix influences the choice of a plural pattern, the fact that the lexicon sample shows a certain preference of -eirV for CVws and lVs and -ismV for CVs and CVjs justifies examining this variable in the experiment data. The logistic regression applied to the test data shows that the suffix -eirV can predict the patterns CVws and lVs, confirming the predominance observed in the lexicon. This result is particularly intriguing, especially because it differentiates CVws from CVjs, patterns whose bases coincide in the singular surface forms, but also because it differentiates CVjs from lVs, patterns that can be claimed to contain /l/ in the singular base. A hypothesis to explore in this case is the frequency of lexical items for each suffix. However, forms derived by -eirV and -ismVrelated to plural patterns CVs and CVws are very scarce in our sample, so a reliable comparison cannot be made. Conversely, lexical frequency may explain the distinct behavior of lVs, a pattern selected as significant in the statistical analysis, compared with CVjs, which was discarded by the logistic regression. This is plotted below (Fig. 6).

Fig. 6.
Fig. 6.

Lexical frequency by suffix for the patterns lVs and CVjs – lemmatized sample CBras

Citation: Acta Linguistica Academica 68, 1-2; 10.1556/2062.2021.00482

With outliers disregarded, the comparison of the lexical frequency for each suffix related to the patterns under analysis allows the following observations: (i) both suffixes show more frequent items in the pattern CVjs; (ii) -eirV shows a higher number of frequent items than -ismV for the two patterns; (iii) the less frequent items relate to suffix -ismV in the pattern lVs; (iv) the difference in terms of lexical frequency between the suffixes for the pattern lVs is considerably greater than that for the pattern CVjs. A reasonable conclusion in this case is that the lexical frequency – especially in the sense of observation (iv) – rather than the properties of the suffixes themselves is responsible for the predictive role identified for this variable. In summary, the statistical analysis of the test involving pseudowords derived by [l] followed by the V-initial suffixes -eirV and -ismV indicates a wide preference for the pluralized patterns lVs and CVjs. Considering the delimitation of this analysis, it can be said that the variables number of syllables and suffix are potential predictors of certain plural patterns. However, lexical information, including type and token frequency, which is not directly measurable in logatomes, may play a role in the selection of such variables.

Nevertheless, the generalized preference for the patterns lVs and CVjs and the distinction between the predictors of CVjs and CVws suggest that the representation of these forms is not mainly determined by markedness constraints on surface forms. Moreover, the fact that bases with two syllables are preferable as long as they belong to the pattern lVs or CVjs implies considering the hypothesis according to what the UR of the latter pattern is, in the singular form, CV/l/, which is therefore distinct from the underlying CV/w/. Given this, the pattern CVjs can be paired with lVs in the sense that both belong to bases containing an underlying /l/. In the next section, we attempt to account for this hypothesis in a constraint-based approach.

5 On the representation of BP [w]-final words

In this section, a proposal for defining the ideal UR for [w]-final words in BP is presented. The analysis is within the broader scope of Optimality Theory. It is not committed to specific sub-theories about richness of the base, lexical optimization or paradigmatic correspondence; instead, the purpose of the analysis is to identify the constraints that can explain the competition between singular bases related to the suffixed forms under analysis. These bases are interpreted here as underlying forms from which plural and suffixed forms are derived.

Considering the quantitative differences revealed in the choice of plural patterns offered to the test participants for the same derived form, it is plausible to admit a competition between them in terms of learning. Putatively, therefore, the learner, encountering an [l]+suffix new word, would rank constraints to decide the best singular base. There are at least four alternatives: CV, lV, CVw, and CVl. If these patterns are compared with the plural ones, the new element is the latter pattern, which can be paired to CVjs, as proposed by Camara (1969; 1970), Morales-Front & Holt (1997), Mateus & D'Andrade (2000), Schwindt & Wetzels (2016), among many others.

As these four candidates are possible bases in BP, and they are all selected to some extent by the test participants, justifying their ordering in addition to identifying the constraints responsible for the emergence of each pattern is necessary.

Selecting the UR, in this case, does not seem to be something that can be solved simply by using the learning strategy of lexical optimization, considering we are faced with evidence from alternations within the paradigm, as discussed by McCarthy (2002, 78). From a learning perspective, it can be said that different rankings are available to the speakers, although some express the prevalent structures of the language better than others do.

The competition between these forms involves mainly deciding on the morphological affiliation of the consonant [l] (or its phonemic correspondent) that appears in the derived form. The first point to sustain is that words with the patterns lV and CVl, the most recurrent among the test responses and although possibly corresponding to different lexical items in BP, do not differ in most cases in terms of derivation by vowel-initial suffixes, as their roots are identical. This is the case for sol ‘sun’ /solaço ‘strong sunlight’ compared with solo ‘soil’ / soleira ‘sill’.12 The reason for this is that BP has at least two types of affixes – those attached to the root and those attached to the word or stem, with the latter being distinct from the former because they are, by themselves, independent phonological words (e.g., -zinho, -mente) according to Schwindt (2013) and Ulrich & Schwindt (2020). The suffixes covered in this paper are of the first type.

From this perspective, in line with McCarthy (2006) and Benua (1997), I suggest that a generalized alignment constraint and a paradigmatic constraint together account for the affix location and the requirement of phonological correspondence between the base and the derivative in this case. These constraints, ranked above a markedness constraint concerning to the syllabic coda, are responsible for the emergence of the most abstract pattern, CVl, as the optimal UR for [w]-final words in BP.

Align(Suffix,L;Root,R)
The left edge of a suffix must be aligned with the right edge of a root.
Ident-OO
Corresponding segments of the base and the derived form must agree.
*l]σ
[l] is forbidden in coda position.

The following tableau shows the ranking for candidates to inputs for the derived pseudoword fabeleiro. Align(Suffix,L;Root,R) bans the candidate that requires epenthesis, c, in favor of those that contain /l/, or some segment representing it, in the root. These candidates in turn violate other constraints in the hierarchy. Ident-OO, which expresses the expectation of segmental correspondence between the base and the derivative, bans b, because of undergoing syncope of the thematic vowel in the derived form, and d, because it surfaces /l/ as /w/. ∗l]s, a markedness constraint that prohibits an alveolar lateral consonant from emerging in the coda position, is violated by the candidate a, the winner.

If the analysis proposed in the Tableau 1 is correct, a vocalization or velarization process is required to account for the fact that forms closed by /l/, despite representing the best UR, do not surface in the language. As mentioned in Section 2, many studies have taken this point of view, trying to account for the relationship between this final segment and the emergence of the palatal glide in pluralized forms, represented in the test by the Cvjs pattern. In this paper, I focus only on the relevance of the constraint ∗l]σ, referring the reader to the literature on the phonetic instantiation of these forms, which is partly summarized in Section 2.

Tableau 1.

Align(Suffix,L;Root,R), Ident-OO ›› ∗l]σ

It must be made clear, however, that the vocalized form in Tableau 2 cannot be equated with forms closed by an underlying /w/ (e.g., pa[w] ’stick’ / pa[w][l]eira ‘something heavy, really hard’). As mentioned, an analogous example was attested in the experiment, characterized as a typical case in which the preferred pluralized pattern is CVws. In this context, I suggest that a phonotactic constraint rejecting a diphthong that is followed by a vowel licenses the epenthesis (Tableau 3).

*Diph-V
A diphthong followed by a vowel is forbidden.
Tableau 2.

∗l]σ ›› Ident-IO

Tableau 3.

Diph-V ›› Align(Suffix,L;Root,R), Ident-IO

In summary, the following ranking can model the cases discussed in this section:

*Diph-V ›› Align(Suffix,L;Root,R), Ident-OO ›› ∗l]σ ›› Ident-IO

Words such as chapé/w/ (‘hat’), in which the final [w] seems to interchange with [l] in the derivation, as in chape[l]aria (‘coatcheck’), need further investigation. This case seems to be as exceptional as that permitting epenthesis in cha[l]eira (‘kettle’), with the addition that chapéu originates from an Old French form containing [l], chapel. These words rarely preserve compositional meaning (chapelaria is a term rarely adopted to designate ‘a place where hats are kept/sold/manufactured,’ and chaleira is no longer used to designate a ‘tea-pot’). In addition, it should not be ruled out that constraints related to writing, which are in conflict with phonological and morphological constraints, may contribute to explaining l/w alternation in BP.

Another issue not addressed directly in this section concerns the hypothesis of suffix allomorphy, that is, although Align(Suffix,L;Root,R) ensures that suffix derivation in BP is root-based, it says nothing about the possibility of [l] integrating a variant of the vowel-initial suffix. However, the hypothesis of lexicalizing morphemes (as opposed to epenthesis), although not exactly contradicting the thesis defended in this article, seems to be weakened by the fact that the majority of words suffixed by [l] + eirV/ismV both in the test and in the lexicon data can be explained by containing [l] in the base. A more accurate study of the effect of the recurrence of these forms and their representation in memory is likely to contribute to qualifying this debate. Meanwhile, an analysis in which the lexicon does not need to store a new subcategory of contextually undefined morphemes seems to be more appropriate.

6 Final remarks

This paper has dealt with the constraints involved in learning URs of [w]-final words in BP, focusing on derived words in which an intervening [l] emerges between the base and the vowel-initial suffix (e.g., capita[l]ismo ‘capitalism’, cliente[l]ismo ‘patronage’). Data from a pseudoword task contrasted to a lexicon sample show that the speakers relate these forms preponderantly to bases already containing [l] in the last syllable, either in the onset or coda position. This finding weakens the hypothesis that [l] is part of the suffix or a product of epenthesis. At the same time, it contributes to supporting the thesis that ordinary suffixes in BP are attached to roots, rather than words, and, therefore, before [l] meets the requirement for vocalization, which is to be on the right edge of a prosodic word. In a constraint-based approach, the high ranking of a generalized alignment constraint and a paradigmatic correspondence constraint seems to account for this fact and is consistent with the proposition of abstract URs for the bases analyzed.

From the point of view of quantitative analysis, the approach of derived words has imposed some restrictions on the investigation of possible predictors for the bases related to them. For instance, the neutralization of mid vowels in BP-suffixed words does not allow any conclusion about the possible role of vowel tenseness in the selection of base patterns. Improvements in the test and statistical analysis may contribute to minimizing this and other possible limitations of the research. From a theoretical point of view, the descriptive and analytical contribution of the approach of this specific BP phenomenon points to the importance of deepening the debate about the learnability of constraints that account for UR’s within the scope of Optimality Theory.

Acknowledgments

I am grateful to the National Council for Scientific and Technological Development (CNPq) for the financial support for this research (grant PQ-310921/2018-0). I thank Camila Ulrich, Isabela Petry, Júlia Ricardo, Nathan Barcellos, Pedro Gaggiola, and Pedro Surreaux for contributing in different stages of this research. I also thank Michael Becker and the anonymous reviewers for their comments and useful advice. Needless to say, all remaining errors are my own responsibility.

References

  • Aronoff, Mark. 1976. Word formation in generative grammar. Cambridge, MA: MIT Press.

  • Becker, Michael, Andrew Nevins, Filomena Sandalo and Érick Rizzato. 2018. The acquisition path of [w]-final plurals in Brazilian Portuguese. Journal of Portuguese Linguistics 17(4). 117. http://doi.org/10.5334/jpl.189.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Benua, Laura. 1997. Transderivational identity: Phonological relations between words. Doctoral dissertation. University of Massachussets, Amherst, MA. Retrieved from https://scholarworks.umass.edu/dissertations/AAI9809307.

    • Search Google Scholar
    • Export Citation
  • Cagliari, Luiz Carlos and Gladis Massini-Cagliari. 2000. A epêntese consonantal em português e sua interpretação na Teoria da Otimalidade [The consonantal epenthesis in Portuguese and its interpretation in Optimality Theory]. Revista de Estudos da Linguagem [Journal of Language Studies] 9(1). 163192. http://dx.doi.org/10.17851/2237-2083.9.1.109-162.

    • Search Google Scholar
    • Export Citation
  • Camara, Joaquim Mattoso Jr. 1969. Problemas de linguística descritiva [Descriptive linguistic problems]. Petrópolis: Ed. Vozes.

  • Camara, Joaquim Mattoso Jr. 1970. Estrutura da língua portuguesa [Portuguese language structure]. Petrópolis: Ed. Vozes.

  • Canfield, Samanta Sá. 2018. Breve descrição da epêntese consonantal em palavras derivadas por sufixação no português brasileiro [Brief description of consonant epenthesis in words derived by suffixation in Brazilian Portuguese]. Cadernos do IL 56. 5769. Retrieved from https://seer.ufrgs.br/cadernosdoil/article/view/83495/50776.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Collischonn, Gisela and Laura Quednau. 2010. As laterais variáveis da região sul [The variable sides of the south region]. In L. Bisol and G. Collischonn (eds.) Português do sul do Brasil: variação fonológica [Portuguese from southern Brazil: phonological variation]. 1st edition. Porto Alegre: EDIPUCRS. 154176.

    • Search Google Scholar
    • Export Citation
  • Costa, Cristiane Ferreira. 2004. A vocalização da lateral pós-vocálica como fenômeno neogramático do nível pós-lexical [The vocalization of the post-vocalic lateral as a neogrammatical phenomenon of the post-lexical level]. Organon 36. 8391. https://doi.org/10.22456/2238-8915.31156.

    • Search Google Scholar
    • Export Citation
  • Cristófaro Silva, Thaïs. 2012. Organização fonológica de marcas de plural no português brasileiro: uma abordagem multirrepresentacional [Phonological organization of plural marks in Brazilian Portuguese: A multi-representational approach]. Revista da ABRALIN 11. 273305. Retrieved from https://revista.abralin.org/index.php/abralin/article/view/1141/1064.

    • Search Google Scholar
    • Export Citation
  • Espiga, Jorge. 2002. A lateral posvocálica na fronteira dos campos neutrais: estudo sociolinguístico da regra telescópica nos dialetos de Chuí e Santa Vitória do Palmar [The postvocalic side on the boundary of neutral fields: Sociolinguistic study of telescopic rule in the dialects of Chuí and Santa Vitória do Palmar]. Letras de Hoje 37(1). 4968. Retrieved from https://revistaseletronicas.pucrs.br/ojs/index.php/fale/article/view/14158/9397.

    • Search Google Scholar
    • Export Citation
  • Girelli, Carl Anthony. 1988. Brazilian Portuguese syllable structure. Doctoral dissertation. University of Connecticut, Storrs, CT. Retrieved from https://opencommons.uconn.edu/dissertations/AAI8905365.

    • Search Google Scholar
    • Export Citation
  • Grodt, Aline. 2009. Um estudo sobre produtividade derivacional no português falado no sul do Brasil [A study on derivational productivity in Portuguese spoken in southern Brazil]. MA dissertation. Federal University of Rio Grande do Sul, Porto Alegre. Retrieved from https://www.lume.ufrgs.br/bitstream/handle/10183/21581/000736124.pdf.

    • Search Google Scholar
    • Export Citation
  • Huback, Ana Paula. 2010a. Plurais irregulares do português brasileiro: efeitos de frequência [Irregular plurals of Brazilian Portuguese: frequency effects]. Revista da ABRALIN [Magazine of ABRALIN] 9(1). 1140. Retrieved from https://revista.abralin.org/index.php/abralin/article/view/1047/970.

    • Search Google Scholar
    • Export Citation
  • Huback, Ana Paula. 2010b. Plurais em -ão do português brasileiro: efeitos de frequência [Plurals in -ão of Brazilian Portuguese: frequency effects]. Revista Linguística [Linguistic Magazine] 6(1). 926. https://doi.org/10.31513/linguistica.2010.v6n1a4436.

    • Search Google Scholar
    • Export Citation
  • Mateus, Maria Helena and Ernesto d’Andrade. 2000. The phonology of Portuguese. Oxford: Oxford University Press.

  • McCarthy, John J. 2002. A thematic guide to Optamility Theory. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511613333.

    • Search Google Scholar
    • Export Citation
  • McCarthy, John J. 2006. Morphology: Optimality Theory. In K. Brown (ed.) Encyclopedia of language and linguistics. 2nd edition. 308316.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Morales-Front, Alfonso and D. Eric Holt. 1997. On the interplay of morphology, prosody and faithfulness in Portuguese pluralization. In F. Martínez-Gil and A. Morales-Front (eds.) Issues in the phonology and morphology of the major Iberian languages. Washington, DC: Georgetown University Press. 393437.

    • Search Google Scholar
    • Export Citation
  • Pires, Caroline de Castro. 2016. Epêntese consonantal em contexto de juntura morfêmica: considerações sobre o sufixo -ada [Consonant epenthesis in the context of morphemic junction: Considerations on the -ada suffix]. MA dissertation. Pontifical Catholic University of Rio Grande do Sul, Porto Alegre. Retrieved from http://tede2.pucrs.br/tede2/handle/tede/6724.

    • Search Google Scholar
    • Export Citation
  • R Core Team. 2020. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. https://www.R-project.org/.

    • Search Google Scholar
    • Export Citation
  • Rizzato, Érick. 2018. Interação do plural de -ão e do aumentativo -zão na formação de compostos no português brasileiro [Interação do plural de -ão e do aumentativo -zão na formação de compostos no português brasileiro]. MA thesis. University of Campinas, Campinas. Retrieved from http://repositorio.unicamp.br/jspui/handle/REPOSIP/331822.

    • Search Google Scholar
    • Export Citation
  • Schwindt, Luiz Carlos. 2013. Palavra fonológica e derivação em português brasileiro: considerações para a arquitetura da gramática [Phonological word and derivation in Brasilian Portuguese: considerations for the architecture of grammar]. In L. Bisol and G. Collischonn (eds.) Fonologia: teorias e perspectivas [Phonology: Theories and perspectives]. 1st edition. Porto Alegre: EDIPUCRS. 1528.

    • Search Google Scholar
    • Export Citation
  • Schwindt, Luiz Carlos, Pedro Eugênio Gaggiola and Isabela Prisco Petry. 2021. Frequência e distribuição de plurais irregulares no corpus brasileiro [Frequency and distribution of irregular plurals in the Brazilian corpus]. Revista de Estudos da Linguagem [Journal of Language Studies]. 29(2). Retrieved from http://www.periodicos.letras.ufmg.br/index.php/relin/article/view/17466.

    • Search Google Scholar
    • Export Citation
  • Schwindt, Luiz Carlos and W. Leo Wetzels. 2016. The morphology and phonology of inflection. In W. L. Wetzels, J. Costa and S. Menuzzi (eds.) The Handbook of Portuguese linguistics. Oxford: Wiley-Blackwell. 188209. https://doi.org/10.1002/9781118791844.ch11.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tasca, Maria. 2000. A preservação da lateral alveolar na coda: uma explicação possível [Preservation of the alveolar lateral in the coda: A possible explanation]. Letras de Hoje [Today’s Letters] 35(1). 331354. Retrieved from https://revistaseletronicas.pucrs.br/ojs/index.php/fale/article/view/14771.

    • Search Google Scholar
    • Export Citation
  • Ulrich, Camila Witt and Luiz Carlos Schwindt. 2020. Prosodic independence of affixes in Brazilian Portuguese: An experimental approach. Proceedings of the 2019 Annual Meeting on Phonology. https://doi.org/10.3765/amp.v8i0.4684.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Walsh Dickey, Laura. 1997. The phonology of liquids. Doctoral dissertation. University of Massachusetts, Amherst, MA. Retrieved from https://scholarworks.umass.edu/dissertations/AAI9721443.

    • Search Google Scholar
    • Export Citation
  • Wickham, Hadley. 2016. ggplot2: Elegant graphics for data analysis. New York: Springer-Verlag. Retrieved from https://ggplot2.tidyverse.org.

1

Collischonn & Quednau (2010) report rates of vocalization instead of lateral preservation. The percentages were just inverted here because of the direction assumed for the argument in this paper.

2

Tasca (2000) and Espiga (2002), in an autosegmental approach, treated vocalization as a case of phonological change in BP with properties of a telescopic rule in the following direction: [l]> [ɫ]> [w].

3

Violations of this plural alternation (e.g., chapé[js], hoté[ws]) are very rare in BP, according to the CBras database. The few cases attested in the language seem to be better explained as hypercorrection in sociolinguistic terms.

4

Although this research depends on the correspondence between derived and plural forms, the argument does not depend strictly on paradigmatic relations. Plural and suffixed forms are considered to be derived – in a general sense of the term – from the singular form, their basic or underlying representation (UR).

5

Girelli (1988) proposes that the velarization of /l/ is also a way to instantiate vocalization. In this perspective, [ɫ] and [w] would result from a similar process, not depending on the plural context to emerge.

7

Informants who graduated in language-related courses did not exhibit significantly different behavior from others regarding the selection of the examined plural patterns.

8

Such diacritics are used with a very similar function in the official Portuguese spelling, so they probably did not sound odd to the respondents.

9

The vast majority of Portuguese non-verbs whose plural forms end in [ews] or [ejs] refer to the gentile suffix -eu (e.g., europeus ‘Europeans’) and the adjective-forming suffix -vel (e.g., adoráveis ‘adorable-pl’), respectively. Items ending in [ojs] in CBras, except for the noun bois ‘oxen-pl’, are non-pluralizable words, as pois ‘then’, depois ‘after’, or dois ‘two’. There is no record of a Portuguese productive word ending in [ows].

10

In this case, epenthesis would probably be blocked by some constraint militating against sequences of similar adjacent segments, which would ban forms such as *valeleiro, an unusual type of construction in Portuguese.

11

The avoidance of monosyllables with the lV pattern is motivated by a limitation of the test itself, as there are no such bases (e.g., lalismolás) among the alternatives. However, this does not seem to affect the preference for relating trisyllables to that pattern. In relation to the CVl pattern, a speculative explanation for this result is that the constraint militating against syllables closed by [l] could be reinforced in the context of monosyllables, penalizing alternatives such as béis for beleira.

12

Despite this base neutralization, BP avoids generating homophonic derivative pairs, which is probably explained by some type of blocking, in the sense of Aronoff (1976), and not by distinction in the alignment demanded by affixation. Eventual epenthesis (e.g., solzeira ‘abundant sunshine’) should be interpreted as repairing such blocking. The intervening segment, however, is [z] in this case, defined by many authors as the default consonant for epenthesis in Portuguese.

  • Aronoff, Mark. 1976. Word formation in generative grammar. Cambridge, MA: MIT Press.

  • Becker, Michael, Andrew Nevins, Filomena Sandalo and Érick Rizzato. 2018. The acquisition path of [w]-final plurals in Brazilian Portuguese. Journal of Portuguese Linguistics 17(4). 117. http://doi.org/10.5334/jpl.189.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Benua, Laura. 1997. Transderivational identity: Phonological relations between words. Doctoral dissertation. University of Massachussets, Amherst, MA. Retrieved from https://scholarworks.umass.edu/dissertations/AAI9809307.

    • Search Google Scholar
    • Export Citation
  • Cagliari, Luiz Carlos and Gladis Massini-Cagliari. 2000. A epêntese consonantal em português e sua interpretação na Teoria da Otimalidade [The consonantal epenthesis in Portuguese and its interpretation in Optimality Theory]. Revista de Estudos da Linguagem [Journal of Language Studies] 9(1). 163192. http://dx.doi.org/10.17851/2237-2083.9.1.109-162.

    • Search Google Scholar
    • Export Citation
  • Camara, Joaquim Mattoso Jr. 1969. Problemas de linguística descritiva [Descriptive linguistic problems]. Petrópolis: Ed. Vozes.

  • Camara, Joaquim Mattoso Jr. 1970. Estrutura da língua portuguesa [Portuguese language structure]. Petrópolis: Ed. Vozes.

  • Canfield, Samanta Sá. 2018. Breve descrição da epêntese consonantal em palavras derivadas por sufixação no português brasileiro [Brief description of consonant epenthesis in words derived by suffixation in Brazilian Portuguese]. Cadernos do IL 56. 5769. Retrieved from https://seer.ufrgs.br/cadernosdoil/article/view/83495/50776.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Collischonn, Gisela and Laura Quednau. 2010. As laterais variáveis da região sul [The variable sides of the south region]. In L. Bisol and G. Collischonn (eds.) Português do sul do Brasil: variação fonológica [Portuguese from southern Brazil: phonological variation]. 1st edition. Porto Alegre: EDIPUCRS. 154176.

    • Search Google Scholar
    • Export Citation
  • Costa, Cristiane Ferreira. 2004. A vocalização da lateral pós-vocálica como fenômeno neogramático do nível pós-lexical [The vocalization of the post-vocalic lateral as a neogrammatical phenomenon of the post-lexical level]. Organon 36. 8391. https://doi.org/10.22456/2238-8915.31156.

    • Search Google Scholar
    • Export Citation
  • Cristófaro Silva, Thaïs. 2012. Organização fonológica de marcas de plural no português brasileiro: uma abordagem multirrepresentacional [Phonological organization of plural marks in Brazilian Portuguese: A multi-representational approach]. Revista da ABRALIN 11. 273305. Retrieved from https://revista.abralin.org/index.php/abralin/article/view/1141/1064.

    • Search Google Scholar
    • Export Citation
  • Espiga, Jorge. 2002. A lateral posvocálica na fronteira dos campos neutrais: estudo sociolinguístico da regra telescópica nos dialetos de Chuí e Santa Vitória do Palmar [The postvocalic side on the boundary of neutral fields: Sociolinguistic study of telescopic rule in the dialects of Chuí and Santa Vitória do Palmar]. Letras de Hoje 37(1). 4968. Retrieved from https://revistaseletronicas.pucrs.br/ojs/index.php/fale/article/view/14158/9397.

    • Search Google Scholar
    • Export Citation
  • Girelli, Carl Anthony. 1988. Brazilian Portuguese syllable structure. Doctoral dissertation. University of Connecticut, Storrs, CT. Retrieved from https://opencommons.uconn.edu/dissertations/AAI8905365.

    • Search Google Scholar
    • Export Citation
  • Grodt, Aline. 2009. Um estudo sobre produtividade derivacional no português falado no sul do Brasil [A study on derivational productivity in Portuguese spoken in southern Brazil]. MA dissertation. Federal University of Rio Grande do Sul, Porto Alegre. Retrieved from https://www.lume.ufrgs.br/bitstream/handle/10183/21581/000736124.pdf.

    • Search Google Scholar
    • Export Citation
  • Huback, Ana Paula. 2010a. Plurais irregulares do português brasileiro: efeitos de frequência [Irregular plurals of Brazilian Portuguese: frequency effects]. Revista da ABRALIN [Magazine of ABRALIN] 9(1). 1140. Retrieved from https://revista.abralin.org/index.php/abralin/article/view/1047/970.

    • Search Google Scholar
    • Export Citation
  • Huback, Ana Paula. 2010b. Plurais em -ão do português brasileiro: efeitos de frequência [Plurals in -ão of Brazilian Portuguese: frequency effects]. Revista Linguística [Linguistic Magazine] 6(1). 926. https://doi.org/10.31513/linguistica.2010.v6n1a4436.

    • Search Google Scholar
    • Export Citation
  • Mateus, Maria Helena and Ernesto d’Andrade. 2000. The phonology of Portuguese. Oxford: Oxford University Press.

  • McCarthy, John J. 2002. A thematic guide to Optamility Theory. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511613333.

    • Search Google Scholar
    • Export Citation
  • McCarthy, John J. 2006. Morphology: Optimality Theory. In K. Brown (ed.) Encyclopedia of language and linguistics. 2nd edition. 308316.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Morales-Front, Alfonso and D. Eric Holt. 1997. On the interplay of morphology, prosody and faithfulness in Portuguese pluralization. In F. Martínez-Gil and A. Morales-Front (eds.) Issues in the phonology and morphology of the major Iberian languages. Washington, DC: Georgetown University Press. 393437.

    • Search Google Scholar
    • Export Citation
  • Pires, Caroline de Castro. 2016. Epêntese consonantal em contexto de juntura morfêmica: considerações sobre o sufixo -ada [Consonant epenthesis in the context of morphemic junction: Considerations on the -ada suffix]. MA dissertation. Pontifical Catholic University of Rio Grande do Sul, Porto Alegre. Retrieved from http://tede2.pucrs.br/tede2/handle/tede/6724.

    • Search Google Scholar
    • Export Citation
  • R Core Team. 2020. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. https://www.R-project.org/.

    • Search Google Scholar
    • Export Citation
  • Rizzato, Érick. 2018. Interação do plural de -ão e do aumentativo -zão na formação de compostos no português brasileiro [Interação do plural de -ão e do aumentativo -zão na formação de compostos no português brasileiro]. MA thesis. University of Campinas, Campinas. Retrieved from http://repositorio.unicamp.br/jspui/handle/REPOSIP/331822.

    • Search Google Scholar
    • Export Citation
  • Schwindt, Luiz Carlos. 2013. Palavra fonológica e derivação em português brasileiro: considerações para a arquitetura da gramática [Phonological word and derivation in Brasilian Portuguese: considerations for the architecture of grammar]. In L. Bisol and G. Collischonn (eds.) Fonologia: teorias e perspectivas [Phonology: Theories and perspectives]. 1st edition. Porto Alegre: EDIPUCRS. 1528.

    • Search Google Scholar
    • Export Citation
  • Schwindt, Luiz Carlos, Pedro Eugênio Gaggiola and Isabela Prisco Petry. 2021. Frequência e distribuição de plurais irregulares no corpus brasileiro [Frequency and distribution of irregular plurals in the Brazilian corpus]. Revista de Estudos da Linguagem [Journal of Language Studies]. 29(2). Retrieved from http://www.periodicos.letras.ufmg.br/index.php/relin/article/view/17466.

    • Search Google Scholar
    • Export Citation
  • Schwindt, Luiz Carlos and W. Leo Wetzels. 2016. The morphology and phonology of inflection. In W. L. Wetzels, J. Costa and S. Menuzzi (eds.) The Handbook of Portuguese linguistics. Oxford: Wiley-Blackwell. 188209. https://doi.org/10.1002/9781118791844.ch11.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tasca, Maria. 2000. A preservação da lateral alveolar na coda: uma explicação possível [Preservation of the alveolar lateral in the coda: A possible explanation]. Letras de Hoje [Today’s Letters] 35(1). 331354. Retrieved from https://revistaseletronicas.pucrs.br/ojs/index.php/fale/article/view/14771.

    • Search Google Scholar
    • Export Citation
  • Ulrich, Camila Witt and Luiz Carlos Schwindt. 2020. Prosodic independence of affixes in Brazilian Portuguese: An experimental approach. Proceedings of the 2019 Annual Meeting on Phonology. https://doi.org/10.3765/amp.v8i0.4684.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Walsh Dickey, Laura. 1997. The phonology of liquids. Doctoral dissertation. University of Massachusetts, Amherst, MA. Retrieved from https://scholarworks.umass.edu/dissertations/AAI9721443.

    • Search Google Scholar
    • Export Citation
  • Wickham, Hadley. 2016. ggplot2: Elegant graphics for data analysis. New York: Springer-Verlag. Retrieved from https://ggplot2.tidyverse.org.

  • Collapse
  • Expand

Editors

Editor-in-Chief: András Cser

Editor: György Rákosi

Review Editor: Tamás Halm

Editorial Board

  • Anne Abeillé / Université Paris Diderot
  • Željko Bošković / University of Connecticut
  • Marcel den Dikken / Eötvös Loránd University; Hungarian Research Centre for Linguistics, Budapest
  • Hans-Martin Gärtner / Hungarian Research Centre for Linguistics, Budapest
  • Elly van Gelderen / Arizona State University
  • Anders Holmberg / Newcastle University
  • Katarzyna Jaszczolt / University of Cambridge
  • Dániel Z. Kádár / Hungarian Research Centre for Linguistics, Budapest
  • István Kenesei / University of Szeged; Hungarian Research Centre for Linguistics, Budapest
  • Anikó Lipták / Leiden University
  • Katalin Mády / Hungarian Research Centre for Linguistics, Budapest
  • Gereon Müller / Leipzig University
  • Csaba Pléh / Hungarian Academy of Sciences, Central European University
  • Giampaolo Salvi / Eötvös Loránd University
  • Irina Sekerina / College of Staten Island CUNY
  • Péter Siptár / Hungarian Research Centre for Linguistics, Budapest
  • Gregory Stump / University of Kentucky
  • Peter Svenonius / University of Tromsø
  • Anne Tamm / Károli Gáspár University of the Reformed Church
  • Akira Watanabe / University of Tokyo
  • Jeroen van de Weijer / Shenzhen University

 

Acta Linguistica Academica
Address: Benczúr u. 33. HU–1068 Budapest, Hungary
Phone: (+36 1) 351 0413; (+36 1) 321 4830 ext. 154
Fax: (36 1) 322 9297
E-mail: ala@nytud.mta.hu

Indexing and Abstracting Services:

  • Arts and Humanities Citation Index
  • Bibliographie Linguistique/Linguistic Bibliography
  • International Bibliographies IBZ and IBR
  • Linguistics Abstracts
  • Linguistics and Language Behaviour Abstracts
  • MLA International Bibliography
  • SCOPUS
  • Social Science Citation Index
  • LinguisList

 

2023  
Web of Science  
Journal Impact Factor 0.5
Rank by Impact Factor Q3 (Linguistics)
Journal Citation Indicator 0.37
Scopus  
CiteScore 1.0
CiteScore rank Q1 (Literature and Literary Theory)
SNIP 0.571
Scimago  
SJR index 0.344
SJR Q rank Q1

Acta Linguistica Academica
Publication Model Hybrid
Submission Fee none
Article Processing Charge 900 EUR/article
Printed Color Illustrations 40 EUR (or 10 000 HUF) + VAT / piece
Regional discounts on country of the funding agency World Bank Lower-middle-income economies: 50%
World Bank Low-income economies: 100%
Further Discounts Editorial Board / Advisory Board members: 50%
Corresponding authors, affiliated to an EISZ member institution subscribing to the journal package of Akadémiai Kiadó: 100%
Subscription fee 2025 Online subsscription: 648 EUR / 712 USD
Print + online subscription: 744 EUR / 820 USD
Subscription Information Online subscribers are entitled access to all back issues published by Akadémiai Kiadó for each title for the duration of the subscription, as well as Online First content for the subscribed content.
Purchase per Title Individual articles are sold on the displayed price.

Acta Linguistica Academica
Language English
Size B5
Year of
Foundation
2017 (1951)
Volumes
per Year
1
Issues
per Year
4
Founder Magyar Tudományos Akadémia   
Founder's
Address
H-1051 Budapest, Hungary, Széchenyi István tér 9.
Publisher Akadémiai Kiadó
Publisher's
Address
H-1117 Budapest, Hungary 1516 Budapest, PO Box 245.
Responsible
Publisher
Chief Executive Officer, Akadémiai Kiadó
ISSN 2559-8201 (Print)
ISSN 2560-1016 (Online)