The schedule and abstracts are posted below.
| Friday April 18 | ||
| 8:30 – 9:30 | — Breakfast — | |
| 9:30 – 9:45 | — PɸF Welcome — | |
| 9:45 – 10:30 | Harald Hammarström | Pronoun consonant patterns: Deep inheritance or erosion effects? |
| 10:30 – 11:15 | Steven Moran | Interindividual vocal tract diversity influences the phonetic diversification of spoken languages |
| 11:15 – 11:30 | — Coffee Break — | |
| 11:30 – 12:15 | Dmitry Idiatov & Mark Van de Velde | Nasal vowels and the “absence” of nasal consonants in Northern Sub-Saharan Africa |
| 12:15 – 1:30 | — Lunch Break — | |
| 1:30 – 2:15 | Jeff Good | Convergence and divergence across geographic scales in Bantoid and Bantu languages |
| 2:15 – 3:00 | Bonny Sands | 2000 years of contact, change, and restructuring: sound systems before and after the advent of pastoralism in southern Africa |
| 3:00 – 3:15 | — Coffee Break — | |
| 3:15 – 4:00 | Claire Bowern | Disentangling contact-induced stability |
| 4:00 – 4:45 | Shelece Easterday | Disambiguating the role of contact in the areal distribution of glottalized consonants |
| Saturday April 19 | ||
| 9:00 – 10:00 | — Breakfast — | |
| 10:00 – 10:45 | Koen Bostoen | West-Coastal Bantu diachronic phonology and deep-time population contact in the Lower Kasai region of the Democratic Republic of Congo |
| 10:45 – 11:30 | Sara Pacchiarotti | The lexicon and phonology of Fulu: insights into the population history of a small-scale Central Sudanic language spoken in northwestern DR Congo |
| 11:30 – 11:45 | — Coffee Break — | |
| 11:45 – 12:30 | Rosemary Beam de Azcona | Prominence, syllable weight and contact in Eastern Otomanguean |
| 12:30 – 1:45 | — Lunch Break — | |
| 1:45 – 2:30 | Marc Brunelle | Tone in Mainland Southeast Asia: Convergence, Inheritance and Intensity of contact |
| 2:30 – 3:15 | Laura Arnold | Word prosody as a window into the prehistory of northwest New Guinea |
| 3:15 – 3:30 | — Coffee Break — | |
| 3:30 – 4:15 | Kofi Yakpo | Tone-stress contact in the Afro-Atlantic prosodic area |
| 4:15 – 4:30 | — Concluding remarks — | |
| 4:30 – 6:30 | — Reception — | |
Abstracts
Laura Arnold
“Word prosody as a window into the prehistory of northwest New Guinea”
In this talk, I explore what the word-prosodic patterns of several closely related Austronesian languages of the Raja Ampat archipelago—Ambel, As, Batta, Biga, Matbat, and several dialects of Maˈya and Salawati—can reveal about linguistic prehistory and contact in the wider northwest New Guinea region. The Raja Ampat languages are unusual within Austronesian in that they all have lexical tone. In addition, the Maˈya dialects have a cross-linguistically unique combination of tone and lexical stress. These word-prosodic patterns pose some interesting questions about their historical development: to what extent are they the result of spontaneous innovation, inheritance, substrate influence, or areal convergence?
I first present new findings from a bottom-up reconstruction of word prosody in Raja Ampat, which provides robust evidence for subgrouping. These findings show that lexical stress developed through the retention of Austronesian penultimate stress, along with now-fossilized word-formation processes that created some words with final stress. The reconstructions also show that tone developed at least twice in Raja Ampat: once in an ancestor of Ambel, and once in proto-Maˈya-Salawati, the parent of Batta, Biga, Maˈya, and Salawati. These findings refine our understanding of Austronesian settlement and diversification in the archipelago over the last 3,500 years, and demonstrate that tone is a comparatively recent innovation in the Raja Ampat languages.
I then look at the role of internal and external factors in tonogenesis. The internal mechanisms for tonogenesis in Raja Ampat are cross-linguistically unusual, with stress and vowel height as the primary conditioning factors. I also consider the possibility that tonogenesis was contact-induced, by comparing the tone systems of Raja Ampat languages with those of non-Austronesian languages in the broader region. In particular, I examine potential parallels with the East Bird’s Head family, spoken some 300km to the east, and discuss the possibility of an earlier prosodic convergence zone across northwest New Guinea. More broadly, this talk highlights the potential of prosodic reconstruction as a tool for uncovering patterns of human migration and contact.
Rosemary Beam de Azcona
“Prominence, syllable weight and contact in Eastern Otomanguean”
Eastern Otomanguean languages, including Zapotec, Chatino, Mixtec and Popolocan languages, have a variety of stress systems reported, including quantity-sensitive (Nakamoto 2017) and stress fixed on the iamb (Campbell 2014) or the penult. Duration is the main phonetic cue for prominence in these languages, and this can be achieved by lengthening of the vowel (e.g. Arellanes 2009, 2021) or a following consonant (Beam and Gregorio 2023). This talk focuses on metrical structure in the area in and around the Zimatlán and Sola districts of Oaxaca, Mexico. Soltec data point to an earlier quantity-sensitive Zapotecan system but also borrowing from Chatino and other Zapotec languages. Totomachapam, Coyachilla and Asunción Mixtepec Zapotec have prosodic patterns that hint at contact with Mixtec and possibly Chatino. The relevant variables are the location of prominence and whether it is achieved by underlying contrasts or surface lengthening of consonants, vowels, or both. Archaeologically, a Zapotecan presence is attested in this region from 300-100 BCE whereas the earliest Mixtec sites in this region date from around 1100 CE. The Ethnohistorical research on Mixtec-Zapotecan contact has focused on relations between elites over the last millennium. Prosodic patterns of Zapotec languages in the region suggest that the social history of the common people may have involved some Mixtec speakers shifting to Zapotec. Mixtec substrate effects in Zapotec may in turn preserve prosodic patterns once found in, but now lost from, the Mixtec languages of this part of Oaxaca.
Koen Bostoen
“West-Coastal Bantu diachronic phonology and deep-time population contact in the Lower Kasai region of the Democratic Republic of Congo”
The languages spoken immediately south of the Congo rainforest today, from the Atlantic Coast until as far east as the Loange River, all belong to West-Coastal Bantu (WCB), a discrete branch of the Bantu language family (Vansina 1995; Bastin et al. 1999; de Schryver et al. 2015; Pacchiarotti et al. 2019; Pacchiarotti & Bostoen 2020; Koile et al. 2022), also known as West-Western Bantu (Grollemund et al. 2015). WCB languages cover today parts of southern Gabon, southern Congo Republic, southwestern Democratic Republic of Congo (DRC) and northern Angola. Given this modern geographical implantation, it is reasonable to assume that the earliest WCB speakers were the first Bantu speakers south of the Congo rainforest. As far as we can judge from the WCB languages that survived the ravages of time, their center of highest internal divergence is the DRC’s Lower Kasai region, more specifically between the Kamtsha and Kasai Rivers (Pacchiarotti et al. 2019). Ancestral WCB speakers diversified there into distinct language communities and expanded from there westward towards the Atlantic Coast. Taking into account new archaeological data obtained during the BantuFirst project (Coutros et al. 2025), we believe that these early westward expansions started during the Early Iron Age, more specifically toward the end of the first millennium BCE (Bostoen et al. 2025).
The Bantu languages spoken today in the wider vicinity of the WCB homeland area share several unique features in terms of phonology, morphology, and syntax which are absent elsewhere in WCB and often also elsewhere in Bantu more generally. As the geographic distribution of these shared innovations crosscuts all major genealogical WCB subgroups and occur in some neighboring Central-Western Bantu languages, we attribute them to language contact that started after westward expansion out of the homeland had started. We discuss six diachronic sound shifts whose historical spread is believed to be contact-induced: (1) voiced prenasalised consonant cluster reduction (Pacchiarotti et al. 2024); (2) diphthongization (Koni Muluwa & Bostoen 2012); (3) labial-velar stops (Maselli et al. 2021, 2023; Maselli 2024); (4) final vowel loss (Pacchiarotti & Bostoen 2021); (5) umlaut (Bostoen & Koni Muluwa 2014); (6) development of interior vowels (Pacchiarotti et al. 2021). Both labial-velar stops (Idiatov & Van de Velde 2021) and interior vowels (Rolle et al. 2017, 2020) are considered phonological features typical of the so-called ‘Macro-Sudan Belt’, a linguistic macro-area in northern Sub‑Saharan Africa (Clements & Rialland 2008; Güldemann 2008).
We argue that these shared diachronic sound changes are indicative of extensive phonological substrate interference in the Bantu languages of the Lower Kasai region, probably from language shifters belonging to populations no longer present in the study area, both indigenous hunter-gatherer groups and immigrant groups such as Ubangi speakers. After an initial period of Bantu language divergence in the Early Iron Age, the WCB homeland area was likely subject to language convergence triggered by advanced bilingualism involving non-Bantu speakers, both during the Early Iron Age and the Late Iron Age.
Claire Bowern
“Disentangling contact-induced stability”
Australian languages are known for their relatively uniform phone inventories with over 3/4 of the Indigenous languages of the continent describable as a “long, flat” inventory with a single stop series, a nasal for each place of articulation, and 3 vowels (with or without phonemic length). In this talk I examine the role of language contact in language change, in particular the role that language convergence may play in the unusual uniformity of phoneme systems throughout the languages of Australia (Busby 1980, Fletcher & Butcher 2014, Miceli & Round 2022, Bowern 2022, Round 2023). There are three ways in which language contact may have theoretically contributed to this situation. First, rapid contact induced convergence may have played a role: that is, perhaps Australian languages appear similar because, in Dixon’s (1997) words, they have “converged on a common prototype”. This is very unlikely however, because of the scale of contact required, along with the absence of other indications of continent-wide borrowings (Sutton & Koch 2005, among others).
Secondly, one of the sources of new phonemic contrast in languages is borrowing between languages with different phone systems. If all the languages in a region have similar inventories, this removes a potential source of new contrast, potentially slowing down rates at which new phonemic contrasts appear. This is a potential contributing factor although it is worth noting that loan adaptation is also frequent.
The third hypothesis is that borrowing between languages with identical phoneme systems may in fact inhibit language change (rather than simply removing a source of innovative contrasts). I show two ways in which this might play a role. The first is in the way that such contact disrupts structured allophony, which is a necessary precursor to sound change and contrast innovation (cf Blevins & Wedel 2006). The second is that the population dynamics of such multilingualism may end up reinforcing norms and impeding the spread of variants.
Marc Brunelle
“Tone in Mainland Southeast Asia: Convergence, Inheritance and Intensity of contact”
Mainland Southeast Asia is often considered a typical Sprachbund, with languages from five distinct phyla that converge on a number of typological features, like word shape and tonality (Matisoff 1973; Alieva 1984; Enfield 2005). However, this view is partly biased by an emphasis on large national languages and limited descriptions of smaller language varieties.
Using on a database of tonal features in 252 languages (updated from Brunelle and Kirby 2015), I will first illustrate the wide array of tonal diversity found in languages of the area, that ranges from atonal languages (Malayic, Aslian) to languages with large tone inventories but limited tonal rules (Tai-Kadai), as well as languages with small inventories and complex tonal processes (Kuki-Chin). I will then show that geographic distance (a proxy for contact) is, at best, a weak predictor of tonality in languages of the area, using generalized additive models with a geographic smooth, demographic and phylogenetic information, and linguistic factors like syllable structure and the presence of laryngeal contrasts in consonants. Results suggest that phylogenetic affiliation and canonical word shape are the most robust predictors of tonality.
As this seems to contradict received assumptions, I then draw from two celebrated examples of contact-induced tonogenesis, Vietnamese and Chamic languages (Alves 2001; Thurgood 1996, 1999) to illustrate the complex interactions between internal drift and population admixture likely to have led to convergence (using historical, genetic and archeological evidence). I argue that a few instances of intensive language contact, typically involving language shifts, could have disproportionately shaped the language landscape of the area.
Shelece Easterday
Disambiguating the role of contact in the areal distribution of glottalized consonants
Ejectives, implosives, and glottalized resonants, collectively known as ‘glottalized consonants’, show skewed geographical distributions. Given the strong areal patterning of these consonant types, it is natural to ask what role language contact has played and continues to play in their present-day distributions. However, this line of inquiry is complicated by other important considerations.
Ejectives have been proposed to be highly diffusible (Urban and Moran 2021), but at the same time are old and stable features in most of the families in which they occur (Easterday to appear). The restricted areal distribution of glottalized resonants, which almost always co-occur with ejectives, has been interpreted to suggest a separate overlapping areal spreading pattern for the former rather than a dependence on the distribution of the latter (Maddieson 2014), but both consonant types have been observed to arise through the same process of fusion within languages (Fallon 2002). Implosives have been noted to have a distribution which is more areal than genealogical (Maddieson 2014), but are also noted to be old inherited features in some families of Africa, and frequently innovated in others within the region (Clements and Rialland 2007).
Understanding the stability of a linguistic pattern requires a disambiguation of its propensity to be inherited from its propensity to be innovated or acquired through contact (Nichols 2003); understanding the effect of contact on the distribution of that pattern also requires this disambiguation. In this paper I attempt to address these factors in a systematic way in order to better weigh them against one another in accounting for current patterns. In order to isolate contact effects, I examine evidence for diffusion of glottalized consonants in a language sample whose construction allows for the disentangling of contact effects from genealogical inheritance (Di Garbo and Napoleão de Souza). To address the issue of inheritance, I consult reconstructions of families whose present-day languages are attested to have glottalized consonants. To address the issue of innovation, I analyze properties of synchronic and historical phonological processes leading to glottalized consonants.
Different profiles are found for the three glottalized consonant types in terms of all of the factors cited above, but a general finding is that the introduction through contact of a glottalized consonant type into a language which previously did not have it is not frequently attested. On the other hand, contact is shown to have less direct effects which allow for the proliferation of glottalized consonants in languages that already have them.
Jeff Good
“Convergence and divergence across geographic scales in Bantoid and Bantu languages”
The effects of language contact originate in social interaction among members of communities where different languages are present, and contact can result in change that leads to convergence and divergence, as well as increased stability for linguistic features (Kühl & Braunmüller 2014). The well-documented existence of large-scale areal patterns raises important questions about how low-level patterns of contact can result in shared grammatical features across vast regions and hundreds of languages. Answering such questions will require refinements to our approaches to language change that allow us to blend genealogical and areal analysis over different scales of space and time. Such analysis can help us determine whether the effects of contact differ qualitatively and quantitatively across the micro-scale, meso-scale, and macro-scale and, thereby, allow us to build models of change that are sensitive to such “scale-based” effects.
In this talk, I will focus on the use of lexical data from Bantoid and Bantu languages at different scales—including the examination of phonological patterns found within the lexical data—to explore differences between patterns of linguistic diversification found at different geographic scales. This will include an examination of data from the Lower Fungom region of North West Cameroon, a linguistic micro-area, for which an exceptional amount of fine-grained data is available (Good under review), data from languages of Tanzania based on the Tanzanian Language Survey (Nurse & Philippson 1975), as an example of the meso-scale, and Bantu-wide data from Grollemund et al. (2015), as an example of the macro-scale. A key question to be considered is how apparent effects of contact differ at the micro-scale, where language communities are in direct contact with each other, in comparison to the meso-scale and macro-scale, where direct contact would have been more limited, at least historically.
Harald Hammarström
“Pronoun consonant patterns: Deep inheritance or erosion effects?”
It has been observed many times that pronoun roots of different language families resemble each other. As the argument goes, if these resemblances are not due to chance or borrowing, they must represent retentions from macro-families. The resemblances often boil down to a simple 1PSG / 2PSG canon, yet the most ambitious proponents argue for macro-families such as Eurasiatic (m-/t-, Greenberg 1997), a wide-range Trans New Guinea family (n-/k-, Ross 1995), Amerind (n-/m-, Nichols & Peterson 1996) and many others.
In this presentation we will explore an alternative explanation. If a form for a meaning is stable, i.e., it seldom undergoes replacement, the form remains but is subject to ‘erosion’, i.e., leniting sound changes. If pronouns are stable, observations of them should then be more likely to manifest a more eroded form than that found in less stable items. Consequently, pronoun similarities across families may well reflect convergent erosive evolution rather than ancient remnants.
Given tree topologies (from glottolog.org) and wordlists (from asjp.clld.org) we can test the above hypothesis on a global scale, in two steps:
(i) estimate the stability of a given form and/or meaning via automatic cognate judgements
(ii) develop an erosion measurement to quantify how eroded a given word form is
The ASJP transcriptions and automatic cognate judgments are not of such quality that they can be used for individual studies, yet are arguably of sufficient quality for the question at hand.
Dmitry Idiatov & Mark Van de Velde
“Nasal vowels and the “absence” of nasal consonants in Northern Sub-Saharan Africa”
Contrastive nasal vowels are particularly common in the languages of Northern Sub-Saharan Africa (NSSA) when compared to the rest of the world, and have been considered as one of its defining areal features (Clements & Rialland 2008; Hajek 2013; Hyman 1972; Maddieson 2007; Rolle 2013; Williamson 1973). Another specificity of languages of NSSA with respect to nasalization that has been highlighted in the phonological literature is the possibility to analyze various languages as lacking (contrastive) nasal consonants (cf. Bearth 1992; Bole-Richard 1985; Clements & Rialland 2008; Hyman 1972; Ladefoged 1964; Schachter & Fromkin 1968). However, the degree of plausibility of such a “no-nasal” analysis varies significantly across languages (Clements & Rialland 2008:49), and in our view, it largely remains a (somewhat misleading) idealization of more complex phonological realities of the languages in question (see also Bearth 1992; Fromkin 1977).
In this paper, we begin by taking a closer look at the complexities of nasalization phenomena in NSSA, such as the so-called contextual perseverative nasalization of vowels after nasal consonants. We also consider the common pathways for the emergence of nasal vowels in NSSA, several of which appear to be typologically less common, such as *CVNV > CNV > CṼ proposed by Hyman (1972). Interestingly, different pathways appear to show a certain areal bias. Second, we apply the methodology from Idiatov & Van de Velde (2021)’s research on labial-velars to look at the lexical distributions of nasal vowels and combinations of implosives and liquids with nasal vowels in the languages of NSSA in RefLex, a very large lexical database of African languages (Segerer & Flavier 2011-2025). The main goal of this exercise is to determine whether we can detect any non-trivial geographical distribution of lexical frequencies of these features that may tell us something about the linguistic prehistory of NSSA. Our preliminary results suggest that there may be an important degree of convergence with the hotbeds of high lexical frequency of labial-velars (Idiatov & Van de Velde 2021).
Steven Moran
Interindividual vocal tract diversity influences the phonetic diversification of spoken languages
Human evolution has led to the capacity for language and to the morphological specialization of the vocal tract for speech production. However, little is known about how this variation translates into phonetic diversity between speakers, and ultimately, across languages. We demonstrate that different vocal tract shapes and associated articulatory strategies leave consistent acoustic signatures, which show congruent patterns of phonetic variation both within and across speech communities. Recalling a central tenet of evolutionary biology — that within-group variation feeds processes of between-group diversification — we propose a “neutral-like vowel evolution model”, which serves as a null hypothesis to disentangle biological from cultural influences on language change.
Our explicit non-consideration of phonology is a necessary first step in evolutionary analyses. At the heart of historical linguistics lies the question of why languages change the way they do. However, the comparative method cannot tell us how languages evolve because it lacks any assessment of inter-individual variation –- within and across populations –- which is necessary for identifying evolutionary processes in complex and dynamic systems. Rather than comparing typological (phonemic) categories, we examine overarching patterns of phonetic variation after excluding effects of speaker sex, size, and age from the acoustic signal. We analyzed 19,335 vowel utterances made by 3867 speakers who produced the full set of cardinal vowels (i, e, a, o, u) in ten languages (Basque, Bulgarian, Hindi, Italian, Portuguese, Romanian, Russian, Tamil, Turkish, Vietnamese). We then compared patterns of variation of the vowel space of each speaker within and between languages.
The congruence of phonetic patterns of variation within and between groups suggests that vowel sounds evolved -– and continue to evolve -– according to a neutral-like evolutionary process. This finding also suggests that analyzing the acoustic variation of speakers within and across languages may offer new insights into linguistic prehistory.
Sara Pacchiarotti
“The lexicon and phonology of Fulu: insights into the population history of a small-scale Central Sudanic language spoken in northwestern DR Congo”
Fulu [fuu] is the exonym for a small-scale (12,000 speakers in 1986) Sara-Bongo-Bagirmi (SBB) language (Central Sudanic) spoken in several villages to the northeast of Bosobolo and Mobaye near the Ubangi River in the North Ubangi province of the northwestern Democratic Republic of the Congo (DRC) (Fultz and Morgan 2017). Fulu call themselves and their language Gbaya (van Bulck 1954; Tucker and Bryan 1956). The same endonym (Gbaya or Gbaya-Fulu) is used by speakers of Bagiro [fuu], currently located 45 km to the northwest of Mobaye, across the Ubangi River in the Central African Republic (CAR). Boyeldieu (1990: 93) claims that Furu and Bagiro are two exonyms referring to one and the same language. Nevertheless, Boyeldieu’s (1990) evidence for Furu consists of an 80-item wordlist (Mortier 1941) and some sentences contained in Hulstaert (1950).
Based on recently collected first-hand fieldwork data from twenty-one villages on the Bosobolo-Gbadolite road, in this talk we: (i) provide new insights on the sociolinguistic situation of Fulu in the region; (ii) offer a preliminary account of the diachronic and synchronic phonology of Fulu and compare it to that of Bagiro spoken in CAR as described by Boyeldieu (2000); and (iii) assess retentions and innovations for lexicon reconstructed to Proto-Sara Bongo Bagirmi (Boyeldieu et al. 2006), the Central Sudanic subgroup to which Furu allegedly belongs.
Fulu speakers in the DRC live interspersed with speakers of several distinct “Ubangi” genealogical pools, as well as Bantu. In this area, language contact is intense and language shift allegedly happens in a multidirectional fashion among Bantu, Ubangi, and Central Sudanic language groups (Burssens 1958; Maes 1984; Boyeldieu 1990; McMaster 2005; Boyeldieu 2016). In this respect, special attention will be given to borrowings and the detection of possible contact/substrate features which might be informative of the settlement history of this group.
Bonny Sands & Anne-Maria Fehn
“2000 years of contact, change, and restructuring: sound systems before and after the advent of pastoralism in southern Africa”
The Late Stone Age migration of sheep-herders from eastern into southern Africa around 2,000 BP had notable consequences for the languages of southern Africa. Languages from the Khoe-Kwadi family spoken by food producers (particularly, those of the Khoekhoe subbranch) have heavily influenced the lexicons of languages spoken by foragers speaking Kxʼa and Tuu languages. Loans from Khoe-Kwadi languages into Kxʼa and Tuu languages far outweigh loans from Kxʼa or Tuu into Khoe-Kwadi (Fehn et al. 2024, submitted). The different societal structures (i.e. a hierarchical, wealth-accumulating pastoralists vs. a more-egalitarian, immediate-return economy of foragers) influenced the direction of loans. Reconstructing Kxʼa and Tuu and assessing contact between these groups has been difficult due to the fact that many roots (“pan-Khoisanisms”) shared by these groups are also to be found in Khoe-Kwadi; indeed, such shared lexica were the reason some earlier scholars presumed that there was a single “Khoisan” family (e.g. Greenberg 1963, Bleek 1929).
Along with loans came phonological borrowings; in this paper, we assess the structural changes in the click inventories of the three families that post-date the introduction of pastoralism into southern Africa, attempting to peel back the last 2000 years of contact to reconstruct the inventories that may have characterized (pre-)Khoe-Kwadi, (pre-)Kxʼa and (pre-)Tuu. We assess proposed reconstructions for Khoe-Kwadi (Fehn & Rocha 2023) and partial reconstructions of Kxʼa (Heine & Honken 2010, Starostin 2018) and Tuu (Starostin 2021, 2022), based on recent understanding of the types of sound changes affecting clicks (Fehn 2020a,b, Fehn & Sands 2025 & Sands & Fehn 2025, submitted). Discussion will focus on click place of articulation contrasts (particularly the retroflex and bilabial click types) and the contrast between velar and uvular places of articulation in clicks and non-clicks.
Kofi Yakpo
“Tone-stress contact in the Afro-Atlantic prosodic area”
The African-European contact zone of the Atlantic Basin has been conspicuously absent from studies of prosodic areality. This gap stems from ideas that Afro-European Creoles—central to the Atlantic zone—exhibit “simplified” prosodic structures due to creolization, including absent or reduced tone (Heine, 1978; Salmons, 1992; Trudgill, 2010). Prosodic contact outcomes across the Afro-Atlantic suggest the opposite: the imposition of tone on contact varieties appears to be a default in tone-dominant ecologies (Bordal Steien & Yakpo 2020).
Drawing on comparative data, I show that the prosodic systems of the Afro-Atlantic are distributed along an east-west areal axis. Tonal systems prevail in contact varieties spoken in Africa, stress systems in the continental Americas, and transitional hybrid systems in the Caribbean areal buffer zone. Equally, many Caribbean Creoles and European colonial varieties reveal pitch-related phenomena that suggest recent historical shifts from tone to stress and vice versa under demographic and social pressures (Yakpo 2021).
Afro-European Creoles and European colonial varieties are late arrivals in their respective linguistic ecologies. Since areal convergence takes time, typological inconsistencies may persist in the prosodic systems of contact varieties vis-à-vis those of their tonal (Africa) or stress-based (Americas) areal cohabiters. One of these is the dominance of cumulative high tone due to stress-to-tone mapping (Agostinho 2023). A second areal misalignment is the existence of residual tone in Creoles and European colonial varieties otherwise characterized by stress (Berry 1972; Yakpo 2021).
Overall, the Afro-Atlantic prosodic area constitutes a crucial element for a better understanding of the worldwide areal distribution of prosodic systems, and the social factors driving their evolution.
