1 Introduction

We investigate how Tal (West Chadic, Nigeria) grammaticalises a crosslinguistic pattern of iconicity and explore the implications of the grammaticalisation for phonological theory. Iconicity, which is the resemblance between form and meaning, is an attribute of both vocal and sign languages (Perniss et al. 2010; Dingemanse et al. 2015). In studies focusing on vocal language, iconicity is often referred to as sound symbolism (Hinton et al. 1994). To contextualise this work in a vast and growing literature on form-meaning mapping in both vocal and sign languages, we use the term iconicity throughout this work. The relevant pattern of iconic form-meaning mapping for our purpose involves associating high acoustic frequency (i.e., fundamental frequency and second formant) with proximal deixis and low acoustic frequency with distal deixis (Nichols 1971; Nuckolls 1999; Ultan 1978; Johansson & Zlatev 2013; Haynie et al. 2014). Evidence for this pattern of iconicity is stochastic. If we consider that vocal imitation and crossmodal depiction, as the basis of iconicity, is an aspect of the general cognitive system (Kuhl & Meltzoff 1996; Doupe & Kuhl 1999; Hauser et al. 2002; Lockwood & Tuomainen 2015), we should expect grammaticalised iconicity in morphophonological operations. Given that iconicity is traditionally viewed as a peripheral aspect of human language (Hockett 1960; de Saussure 1974), there is reason to believe that its grammaticalisation is underdocumented.

To present evidence for grammaticalised deictic iconicity, we document, describe, and analyse the tonal alternations of modified nouns in Tal. For example, the modified nouns bear a low tone on all their tone-bearing units (TBUs) when the modifier is a medial or distal demonstrative pronoun, third-person possessive pronoun or a noun. When the modifier is a proximal demonstrative pronoun, a first-person possessive pronoun or a second-person possessive pronoun, the final low tone of the modified noun surfaces as a mid tone. In (1), the tonal alternations are illustrated through nouns modified by the demonstrative pronouns, namely distal (far from the speaker and the addressee), medial (near the addressee) and proximal (near the speaker).

    1. (1)
    1. Demonstrative pronouns as modifiers in Tal
    1.  
    2. a.
    3. b.
    4. c.
    5. d.
    1. n
    2. dídí
    3. ɓɔ̄lɔ̄ŋ
    4. kàsɛ̀ŋ
    5. àlmákàʃī
    1. n distal
    2. dìdì ɲɛː᷄
    3. ɓɔ̀lɔ̀ŋ ɲɛː᷄
    4. kàsɛ̀ŋ ɲɛː᷄
    5. àlmàkàʃì ɲɛː᷄
    1. n medial
    2. dìdì ɲɛ̀ː
    3. ɓɔ̀lɔ̀ŋ ɲɛ̀ː
    4. kàsɛ̀ŋ ɲɛ̀ː
    5. àlmàkàʃì ɲɛ̀ː
    1. n proximal
    2. dídí sɛ̄
    3. ɓɔ̄lɔ̄ŋ sɛ̄
    4. kàsɛ̄ŋ sɛ̄
    5. àlmákàʃī sɛ̄
    1.  
    2. ‘sweat fly’
    3. ‘internal organ’
    4. ‘beer filter’
    5. ‘scissor sp.’

Based on the structure of nominal modification in Chadic languages and the deictic properties of nouns and pronominals crosslinguistically, we will consider the overwriting low and mid tones by-products of non-proximal and proximal linkers which have the tone features [–upper, –raised] and [+raised] as their respective exponents. In this case, the non-proximal and proximal linkers are non-segmental morphemes, also known as featural affixes (Akinlabi 1996). The phonological asymmetry in the realisation of the [–upper, –raised] and [+raised] featural affixes is a by-product of distinct and interacting constraints. A featural correspondence account of the featural affixation is presented within the framework of Optimality Theory (McCarthy & Prince 1993; 1995; Prince & Smolensky 2004).

We will argue that the form-meaning mapping of the proximal and non-proximal linkers is consistent with crosslinguistic patterns of distance-related iconicity (Nichols 1971; Ultan 1978; Johansson & Zlatev 2013; Haynie et al. 2014). We will also argue that iconicity motivates the long-distance and local realisations of the non-proximal and proximal featural affixes respectively, considering that the phonological locality of the featural affixes is perceptually similar to deictic gestures. As a result of this, we propose that the motivation for phonological grammar is not limited to language-internal conditions but also includes crossmodal perception of sensory imagery. We show that the grammaticalised deictic iconicity in Tal shares properties, such as phonetic grounding, naturalness and phonological asymmetry, in common with arbitrary phonological patterns. Most importantly, we contribute to the growing typology of grammaticalised iconicity (e.g., Hurch 2005 on reduplication, Alderete & Kochetov 2017 on expressive palatalisation, Akinbo 2021a on root-vowel fronting and backing). Consequently, Tal challenges the completely arbitrary view of form-meaning mapping (Hockett 1960; de Saussure 1974) and presents grammaticalised evidence in support of the view that recognises the role of both arbitrariness and iconicity in grammar (Lockwood & Dingemanse 2015; Dingemanse 2018).

As a background to the discussion in this work, we present the relevant background on the sound inventory and pronominal system of the language in §2. In §3, we present the dataset that forms the basis of this work. We present our analysis and discussion in §4. The summary and conclusion are presented in §5. For transparent and open research, the link to the database, which contains all the sound files, the accompanying textgrids of their phonetic transcription and R codes forming the basis of this work, can be found in the appendix.

2 Background

Tal is an A3 West Chadic language with about 30,000 speakers (Bulkaam 2022). The language is predominantly spoken in Pankshin local government area (LGA), Plateau State, Nigeria. Tal speakers are also found in Yashi (Shendam LGA) and along the border of Mikang and Pankshin LGAs. See Figure 1 for the main settlements. The word Tal is the name of the people, their language, and homeland (Baklit 2014). Most Tal adults are bilingual in Hausa and Tal, while most children and elders are monolingual in Tal. There are six varieties of Tal, namely Muɗak, Mbaal, Mungƙoot, Muɗong, Bongmut, and Takong. According to Bulkaam (2022), the only difference between the dialects are segmental and tonal variations in some words, such as the examples in (2). The variety of Tal that forms the basis of our investigation is Muɗak.

    1. (2)
    1. Lexical differences among varieties of Tal (Bulkaam 2022: p. 33)
    1.  
    2. eight
    3. sweet potato
    4. teeth
    5. mouth
    6. beans
    1. Mbaal
    2. dɛː᷄fɛ́j
    3. dàƙúŋ
    4. ha᷄s
    5. ƙàphɨ́
    6. ɨ̀jɨ̀m
    1. Takong
    2. pàːfɛ́j
    3. nàƙúŋ
    4. ha᷄s
    5. ƙàphɨ́
    6. ɨ̀jɨ̀m
    1. Muɗak
    2. pàːfɛ́j
    3. nàƙúŋ
    4. ha᷄s
    5. ƙàphɨ́
    6. ɨ̀jɨ̀m
    1. Muɗong
    2. ndɛ̀fɛ́j
    3. nàƙúŋ
    4. hɔ᷄ːs
    5. ƙàphɨ́
    6. ɨ̀jɨ̀m
    1. Bongmut
    2. fɛ́rɱfə̀j
    3. dàƙúŋ
    4. hɔ᷄ːs
    5. ɡɨ̀pɨ́
    6. ɨ̀ɾɨ̄m
    1. Mungƙoot
    2. dɛː᷄fɛ́j
    3. dàƙúŋ
    4. hɔ᷄ːs
    5. ɡɨ̀pɨ́
    6. ɨ̀ɾɨ̄m

The language is sparsely studied, considering that the only published work on the language is a 200 word list cited in Jungraithmayr & Holubova (2016). The only descriptive works on the language are the BA and MA theses of Bulkaam (2017; 2022).

As a background to the discussion in this work, this section focuses on aspects of Tal morphophonology that are directly relevant. The data source of our investigation is Bulkaam (2017; 2022) and the introspective knowledge of the second author who is a native speaker of the Muɗak variety. All the Tal data presented in this work are phonetically transcribed.

Figure 1
Figure 1

Tal settlements in Panskin LGA and along the border of Mikang LGA.

2.1 Consonants and vowels

There are thirty-four consonants in Tal, comprising oral stops (plosives and implosives), fricatives, nasals, trill and approximants (Bulkaam 2022), as shown in Table 1. The oral stops have four-way contrast, namely voicelessness, voicing, aspiration and implosive. The fricatives also contrast aspiration, voicing, and voicelessness.

Table 1

Phonemic consonants in Tal (Bulkaam 2022:29).

Bilabial Labio-dental Alveolar Post-alveolar Palatal Labial-palatal Velar Labial-velar Glottal
Plosive p b t d k g kp gb ʔ
Asp. ph th kh
Implosive ɓ ɗ ƙ
Fricative f v s z ʃ ʒ h
Asp. fh sh ʃh
Nasal m n ɲ ŋ
Trill r
Appr. j ɥ ɰ w
Lateral l

Tal has twelve contrastive vowels, comprising six short vowels and the corresponding long counterparts. In Table 2, we present the vowel inventory.

Table 2

Phonemic vowels in Tal (Bulkaam 2022:36).

Front Central Back
High i iː ɨ ɨː u uː
Mid ɛ ɛː ɔ ɔː
Low a aː

To show the contrastive status of all the vowels in Tal, we present the (near) minimal pairs in (3).

    1. (3)
    1. Contrast between short and long vowels in Tal
    1. a.
    2.  
    3. b.
    4.  
    5. c.
    6.  
    1. [i]
    2. [ɛ]
    3. [u]
    4. [ɔ]
    5. [ɨ]
    6. [a]
    1. lít
    2. lɛ̀
    3. ƙún
    4. nɔ᷄k
    5. ƙɨ̀n
    6. ƙáp
    1. ‘lion’
    2. ‘to put’
    3. ‘three’
    4. ‘to leave’
    5. ‘uncle’
    6. ‘to dig up crop’
    1. [i:]
    2. [ɛː]
    3. [uː]
    4. [ɔː]
    5. [ɨː]
    6. [a:]
    1. lìːt
    2. lɛ̀ː
    3. ƙūːn
    4. nɔ̀ːk
    5. ƙɨ̀ːn
    6. ƙàːp
    1. ‘shrine’
    2. ‘to harvest with sickle’
    3. ‘snake specie’
    4. ‘life’
    5. ‘buffalo’
    6. ‘baboon’

In Tal, a syllable can contain a vowel with or without an onset or coda.. A syllabic nasal can also form a syllable on its own in the language. The language permits an onset geminate or cluster, only if the first consonant is a nasal. The syllable template in Tal is represented using the moraic structure in (4).

    1. (4)
    1. Syllable structure in Tal

Under the standard moraic theory, a vowel or a syllabic nasal projects a mora (Hyman 1985; Hayes 1989). A long vowel in this case projects two morae. When a syllabic nasal forms a syllable in Tal, it bears no onset or coda.

2.2 Tone

Tal is a tone language, like all Nigerian languages, with the exception of Fulani (Williamson 1984). The language contrasts three level tones, namely H(igh), M(id) and L(ow). In addition to the level tones, the language also contrasts the contour tone HR, which is a raised H tone. As shown in (5), the tones mark both lexical and grammatical distinctions.

    1. (5)
    1. Tone contrast
    1. a.
    2. b.
    3. c.
    4. d.
    1. H
    2. M
    3. L
    4. HR
    1. thál
    2. thāl
    3. thàl
    4. tha᷄l
    1. ‘only’
    2. ‘to greet’
    3. ‘greeting’
    4. ‘a Tal person’
    1. lú
    2. lūk
    3. lùk
    4. lu᷄k
    1. ‘house’
    2. ‘to gossip’
    3. ‘to swell’
    4. ‘to burn’
    1. ɓáŋ
    2. tɔ̄ŋ
    3. tɔ̀ŋ
    4. ɗɛ᷄ːt
    1. ‘to be red’
    2. ‘sitting’
    3. ‘to sit’
    4. ‘to be bitter’

Based on the tone features in Pulleyblank (1986), we propose the feature specification in Table 3 for the tones in Tal. The tone features are modified version of the features proposed in Yip (1980). As shown in Table 3, H and HR form the natural class [+upper] and the tones L and M form the natural class [–upper]. The feature specification also indicates that the tones HR and M are [+raised].

Table 3

Tone features in Tal.

H M L HR
upper + – – +
raised – + – +

The use of tone features, instead of primitives such as H, L and M (Hyman 2010; Clements et al. 2011), is based on the tonal alternation in nominal modification (§4.1) and the acoustics of the tones. The division of the pitch range into two registers is acoustically supported with the F0(Hz) height plot in Figure 2. The plot is based on monomoraic words, which were spoken in isolation by the second author. To control for the effect of vowels and consonants on the F0(Hz) values of the tones (see Hombert 1977; Maddieson 1984; Whalen & Levitt 1995: for further discussion), the spoken words were also hummed in isolation.

Figure 2
Figure 2

F0 plots of Tal tone contrasts in speech and humming modes.

In speech mode, the pitch trajectories of the H and HR tones are as expected. However, the pitch trajectories of L and M tones in speech mode are not as expected. Specifically, the M tone has a lower F0(Hz) than the L tone at the onset but vice versa at the end. The plot of the tones in humming mode suggests that the F0(Hz) trajectories of L and M, as well as H and HR, in speech mode might be an effect of the consonants and vowels. Regardless of the segmental effect on the pitch height of the tones, the F0(Hz) plot simply shows that the four tones are divided into two pitch registers.

    1. (6)
    1. Cooccurrence of tones

The data in (6) shows that all the combinatorial possibilities are attested for the cooccurrence of level tones, but we find no instances where the HR tone occurs as the final tone of a monomorphemic word with two or more syllables. Given that all logically possible cooccurrences are attested for the level tones, we assume that tones must be associated with the tone-bearing unit (TBU) in the input. Considering that a bimoraic syllable in Tal can have two tones and in line with the proposal in Yip (2002), we assume that the TBU in Tal is a mora. Throughout this work, tones of all Tal examples are fully marked.

2.3 Possessive pronouns

The possessive pronouns in Tal are presented in Table 4. The possessive pronouns can be divided into dependent and independent.

Table 4

Possessive pronouns in Tal.

dependent independent
sg pl sg pl
1.poss nā/ná nū/nú mmāːn mmūːn
2.poss.m gā/gá gū/gú mmāk mmūk
2.poss.f gī/gí mmīk
3.poss mɨ̄ːp mɤá mù mɨ̄ːp mù mɤá

Both dependent and independent possessive pronouns are marked for person and number. In addition to these two grammatical features, the 2.sg possessive pronouns distinguish masculine and feminine gender. The possessive pronouns are labelled dependent and independent because only the independent pronouns can occur in isolation or be conjoined, as illustrated in (7).

    1. (7)
    1. Answer fragments: Independent possessive pronouns
    1. a.
    2.  
    3.  
    4.  
    5. b.
    1. mmā:n
    2. mmū:n
    3. mmāk
    4. mmīk
    5. mmāːn ƙín mmīk
    1. ‘(it is) mine’
    2. ‘(it is) ours’
    3. ‘it is yours(m)’
    4. ‘(it is) yours(f)’
    5. ‘(it is) mine and yours(f)’

Both independent and dependent pronouns can function as modifiers but we only focus on the dependent pronouns in this work. See §3 for the usage of the dependent possessive pronouns as modifiers.

2.4 Demonstrative pronouns

Tal has a three-way distinction between demonstrative pronouns, namely proximal (near the speaker), medial (near the addressee) and distal (far from the speaker and addressee), as shown in Table 5. The demonstratives are pluralised with the third-person possessive pronoun [mɤá].

Table 5

Demonstratives in Tal.

sg pl
a. sɛ̄ ‘this’ sɛ̄ mɤá ‘these’ (near the speaker)
b. ɲɛ̀ː ‘this’ ɲɛ̀ː mɤá ‘these’ (near the addressee)
b. ɲɛː᷄ ‘that’ ɲɛː᷄ mɤá ‘those’ (far from the speaker and the addressee)

The medial and distal demonstratives in the language can occur independently as the subject or object of a verb, but the proximal demonstrative cannot, as shown in (8). Consequently, the medial can also be used to refer to entities between the speaker and the addressee in certain contexts.

    1. (8)
    1. Demonstratives in Tal
    1.  
    1. a.
    1. *sɛ̄
    2. prox.dem
    1. thùːɨ̄ː
    2. kill
    1. nī
    2. goat
    1.  
    2. def.det
    1. “this (near the speaker) killed the goat” (intended meaning)
    1.  
    1. b.
    1. ɲɛ̀ː
    2. med.dem
    1. thùː
    2. kill
    1. ɨ̄ː
    2. goat
    1. nī
    2. DEF.DET
    1. “this (near the addressee) killed the goat”
    1.  
    1. c.
    1. ɲɛː᷄
    2. dist.dem
    1. thùː
    2. kill
    1. ɨ̄ː
    2. goat
    1. nī
    2. def.det
    1. “that (far from the speaker and addressee) killed the goat”

To understand the distinction between the proximal and medial demonstrative pronouns, we must look at their cooccurrence in a contrastive context.

    1. (9)
    1. Comparative use of proximal and medial demonstratives
    1.  
    1. a.
    1. phɛ̀tɛ̄j
    2. bed
    1. sɛ̄
    2. prox.dem
    1. wàk
    2. better
    1. phɛ̀tɛ̀j
    2. bed
    1. ɲɛː᷄
    2. med.dem
    1. ‘this bed (near the speaker) is better than this bed (near the addressee)’
    1.  
    1. b.
    1. phɛ̀tɛ̄j
    2. bed
    1. sɛ̄
    2. prox.dem
    1. mɤá
    2. pl
    1. wàk
    2. better
    1. phɛ̀tɛ̀j
    2. bed
    1. ɲɛː᷄
    2. med.dem
    1. mɤá
    2. pl
    1. ‘these beds (near the speaker) are better than these beds (near the addressee)’

As shown in (9), when juxtaposing two entities between the addressee and speaker, the form [sɛ̄] refers to the entity near the speaker and the form [ɲɛː᷄] refers to the entity near the addressee. In the next section, we focus on the use of the pronominal forms and nouns as modifiers.

3 Nominal modification

This section describes the modification of nouns with pronouns and nouns in Tal. The discussion will only focus on the demonstrative pronouns, possessive pronouns, and nouns as modifiers. As a point of departure, we start with demonstrative pronouns.

3.1 Demonstrative pronouns as modifiers

The demonstrative pronouns in Tal can modify a noun. In this case, the demonstrative pronouns immediately follow the nouns being modified, and the modified nouns undergo a tonal alternation, depending on the type of demonstrative modifier and the tone of the nouns in isolation.

    1. (10)
    1. H-tone nouns
    1. N
    2. lú
    3. bí
    4. áj
    5. páŋ
    6. zɔ́k
    7. fɔ́ːk
    8. ʔjáːn
    9. tíː
    10. lɛ́ː
    11. dídí
    12. ƙwáják
    1. N distal
    2. lù ɲɛː᷄
    3. bì ɲɛː᷄
    4. àj ɲɛː᷄
    5. pàŋ ɲɛː᷄
    6. zɔ̀k ɲɛː᷄
    7. fɔ̀ːk ɲɛː᷄
    8. ʔjàːn ɲɛː᷄
    9. tìː ɲɛː᷄
    10. lɛ̀ː ɲɛː᷄
    11. dìdì ɲɛː᷄
    12. ƙwàjàk ɲɛː᷄
    1. N medial
    2. lù ɲɛ̀ː
    3. bì ɲɛ̀ː
    4. àj ɲɛ̀ː
    5. pàŋ ɲɛ̀ː
    6. zɔ̀k ɲɛ̀ː
    7. fɔ̀ːk ɲɛ̀ː
    8. ʔjàːn ɲɛ̀ː
    9. tìː ɲɛ̀ː
    10. lɛ̀ː ɲɛ̀ː
    11. dìdì ɲɛ̀ː
    12. ƙwàjàk ɲɛ̀ː
    1. N proximal
    2. lú sɛ̄
    3. bí sɛ̄
    4. áj sɛ̄
    5. páŋ sɛ̄
    6. zɔ́k sɛ̄
    7. fɔ́ːk sɛ̄
    8. ʔjáːn sɛ̄
    9. tí: sɛ̄
    10. lɛ́ː sɛ̄
    11. dídí sɛ̄
    12. ƙwáják sɛ̄
    1.  
    2. ‘house’
    3. ‘something’
    4. ‘fatness’
    5. ‘stone’
    6. ‘hill’
    7. ‘bushy area’
    8. ‘sclera’
    9. ‘sand type’
    10. ‘load’
    11. ‘sweat fly’
    12. ‘twisting’
    1. (11)
    1. M-tone nouns
    1. N
    2. ɗā
    3. ūs
    4. ās
    5. bāŋ
    6. fhīn
    7. zwāːl
    8. fīːt
    9. phɛ̄ː
    10. māː
    11. kālāp
    12. bɔ̄lɔ̄ŋ
    1. N distal
    2. ɗà ɲɛː᷄
    3. ùs ɲɛː᷄
    4. às ɲɛː᷄
    5. bàŋ ɲɛː᷄
    6. fhìn ɲɛː᷄
    7. zwàːl ɲɛː᷄
    8. fìːt ɲɛː᷄
    9. phɛ̀ː ɲɛː᷄
    10. màː ɲɛː᷄
    11. ƙàlàp ɲɛː᷄
    12. bɔ̀lɔ̀ŋ ɲɛː᷄
    1. N medial
    2. ɗà ɲɛ̀ː
    3. ùs ɲɛ̀ː
    4. às ɲɛ̀ː
    5. bàŋ ɲɛ̀ː
    6. fhìn ɲɛ̀ː
    7. zwàːl ɲɛ̀ː
    8. fìːt ɲɛ̀ː
    9. phɛ̀ː ɲɛ̀ː
    10. màː ɲɛ̀ː
    11. ƙàlàp ɲɛ̀ː
    12. bɔ̀lɔ̀ŋ ɲɛ̀ː
    1. N proximal
    2. ɗā sɛ̄
    3. ūs sɛ̄
    4. ās sɛ̄
    5. bāŋ sɛ̄
    6. fhīn sɛ̄
    7. zwāɛ̀ːl sɛ̄
    8. fīːt sɛ̄
    9. phɛ̄ː sɛ̄
    10. mā sɛ̄
    11. ƙālāp sɛ̄
    12. bɔ̄lɔ̄ŋ sɛ̄
    1.  
    2. ‘calabash’
    3. ‘fire’
    4. ‘dog’
    5. ‘drum’
    6. ‘flour’
    7. ‘needle’
    8. ‘whistling’
    9. ‘place’
    10. ‘mother’
    11. ‘something used for sharing’
    12. ‘internal organ’
    1. (12)
    1. L-tone nouns
    1. N
    2. ɛ̀n
    3. ɛ̀s
    4. phàŋ
    5. phɛ̀l
    6. vɛ̀ːl
    7. swɛ̀ːp
    8. ɰàː
    9. gùː
    10. phɛ̀tɛ̀j
    11. kàsɛ̀ŋ
    1. N distal
    2. ɛ̀n ɲɛː᷄
    3. ɛ̀s ɲɛː᷄
    4. phàŋ ɲɛː᷄
    5. phɛ̀l ɲɛː᷄
    6. vɛ̀ːl ɲɛː᷄
    7. swɛ̀ːp ɲɛː᷄
    8. ɰàː ɲɛː᷄
    9. gùː ɲɛː᷄
    10. phɛ̀tɛ̀j ɲɛː᷄
    11. kàsɛ̀ŋ ɲɛː᷄
    1. N medial
    2. ɛ̀n ɲɛ̀ː
    3. ɛ̀s ɲɛ̀ː
    4. phàŋ ɲɛ̀ː
    5. phɛ̀l ɲɛ̀ː
    6. vɛ̀ːl ɲɛ̀ː
    7. swɛ̀ːp ɲɛ̀ː
    8. ɰàː ɲɛ̀ː
    9. gùː ɲɛ̀ː
    10. phɛ̀tɛ̀j ɲɛ̀ː
    11. kàsɛ̀ŋ ɲɛ̀ː
    1. N proximal
    2. ɛ̄n sɛ̄
    3. ɛ̄s sɛ̄
    4. phāŋ sɛ̄
    5. phɛ̄l sɛ̄
    6. vɛ̄ːl sɛ̄
    7. swɛ̄ːp sɛ̄
    8. ɰāː sɛ̄
    9. gūː sɛ̄
    10. phɛ̀tɛ̄j sɛ̄
    11. kàsɛ̄ŋ sɛ̄
    1.  
    2. ‘heart’
    3. ‘faeces’
    4. ‘fence stone’
    5. ‘flower’
    6. ‘bone marrow’
    7. ‘hair’
    8. ‘grains’
    9. ‘cactus’
    10. ‘bed’
    11. ‘beer filter’

As shown in (10–12), nouns that are modified with the distal [ɲɛː᷄] or medial [ɲɛ̀ː] demonstratives consistently bear an L tone on every TBU regardless of their tones in isolation. The modification of a noun with the proximal demonstrative pronoun [sɛ̄] only results in a tonal alternation when the final tone of the noun is L. As shown in (12), the final L tone of a noun surfaces as M when the noun is modified with the proximal demonstrative pronoun [sɛ̄]. This is the case regardless of the syllable, tonal combinations or tone-type, as exemplified in (13–15).

    1. (13)
    1. Bisyllabic nouns with dissimilar tones
    1.  
    2. H-M
    3.  
    4. M-H
    5.  
    6. L-H
    7.  
    8. L-M
    9.  
    10. M-L
    11.  
    1. N
    2. ʒáŋlīp
    3. ƙádāːl
    4. ƙāmɓːl
    5. ɡāːlí
    6. ǹdwáŋ
    7. ƙàƙíː
    8. tìzɛ̄ːl
    9. ƙàkɔ̄ŋ
    10. ɓāgàj
    11. vɛ̄ːkùː
    1. N distal
    2. ʒàŋlìp ɲɛː᷄
    3. ƙàdàːl ɲɛː᷄
    4. ƙàmɓɛ̀l ɲɛː᷄
    5. ɡàːlì ɲɛː᷄
    6. ǹdwàŋ ɲɛː᷄
    7. ƙàƙìː ɲɛː᷄
    8. tìzɛ̀ːl ɲɛː᷄
    9. ƙàkɔ̀ŋ ɲɛː᷄
    10. ɓàgàj ɲɛː᷄
    11. vɛ̀ːkùː ɲɛː᷄
    1. N medial
    2. ʒàŋlìp ɲɛ̀ː
    3. ƙàdàːl ɲɛ̀ː
    4. ƙàmɓɛ̀l ɲɛ̀ː
    5. ɡàːlì ɲɛ̀ː
    6. ǹdwàŋ ɲɛ̀ː
    7. ƙàƙìː ɲɛ̀ː
    8. tìzɛ̀ːl ɲɛ̀ː
    9. ƙàkɔ̀ŋ ɲɛ̀ː
    10. ɓàgàj ɲɛ̀ː
    11. vɛ̀ːkùː ɲɛ̀ː
    1. N proximal
    2. ʒáŋlīp sɛ̄
    3. ƙádāːl sɛ̄
    4. ƙāmɓːːl sɛ̄
    5. ɡāːlí sɛ̄
    6. ǹdwáŋ sɛ̄
    7. ƙàƙíː sɛ̄
    8. tìzɛ̄ːl sɛ̄
    9. ƙàkɔ̄ŋ sɛ̄
    10. ɓāgāj sɛ̄
    11. vɛ̄ːkūː sɛ̄
    1.  
    2. ‘spirogyra’
    3. ‘head’
    4. ‘tool used for digging’
    5. ‘millet sp.’
    6. ‘meningitis’
    7. ‘sacrifice’
    8. ‘sperm’
    9. ‘river area’
    10. ‘stone cave’
    11. ‘clarinet’
    1. (14)
    1. Trisyllabic nouns with dissimilar tones
    1.  
    2. L-H-L
    3. M-H-H
    4. H-L-M
    5.  
    6. L-H-H
    7.  
    8. L-M-M
    9. L-H-L-M
    1. N
    2. àsƙúwɛ̀t
    3. thāŋdájáŋ
    4. pɛ́lɛ̀mphɛ̄
    5. ƙálàʔās
    6. ƙɨ̀nkwálák
    7. nàŋʒílíŋ
    8. ƙàskɛ̄lɛ̄m
    9. àlmákàʃī
    1. N distal
    2. àsƙùwɛ̀t ɲɛː᷄
    3. thàŋdàjàŋ ɲɛː᷄
    4. pɛ̀lɛ̀mphɛ̀ ɲɛː᷄
    5. ƙàlàʔàs ɲɛː᷄
    6. ƙɨ̀nkwàlàk ɲɛː᷄
    7. nàŋʒìlìŋ ɲɛː᷄
    8. ƙàskɛ̀lɛ̀m ɲɛː᷄
    9. àlmàkàʃì ɲɛː᷄
    1. N medial
    2. àsƙùwɛ̀t ɲɛ̀ː
    3. thàŋdàjàŋ ɲɛ̀ː
    4. pɛ̀lɛ̀mphɛ̀ ɲɛ̀ː
    5. ƙàlàʔàs ɲɛ̀ː
    6. ƙɨ̀nkwàlàk ɲɛ̀ː
    7. nàŋʒìlìŋ ɲɛ̀ː
    8. ƙàskɛ̀lɛ̀m ɲɛ̀ː
    9. àlmàkàʃì ɲɛ̀ː
    1. N proximal
    2. àsƙúwɛ̄t sɛ̄
    3. thāŋdájáŋ sɛ̄
    4. pɛ́lɛ̀mphɛ̄ sɛ̄
    5. ƙálàʔās sɛ̄
    6. ƙɨ̀nkwálák sɛ̄
    7. nàŋʒílíŋ sɛ̄
    8. ƙàskɛ̄lɛ̄m sɛ̄
    9. àlmákàʃī sɛ̄
    1.  
    2. ‘type of dance’
    3. ‘type of raft’
    4. ‘dragonfly’
    5. ‘herb sp.’
    6. ‘pot cover’
    7. ‘sunbird’
    8. ‘herb sp.’
    9. ‘scissors’

The only contrastive contour tone in the language is HR. The contour tone, like all the level tones, is consistently lowered when a noun is modified with the medial or distal demonstrative but invariant in the environment of the proximal demonstrative, as shown in (15).

    1. (15)
    1. Nouns with contour tones
    1.  
    2. HR
    3.  
    4. HR
    5. HR-H
    6. L-HR
    1. N
    2. tha᷄l
    3. khu᷄m
    4. ʒɛ᷄lɓáŋ
    5. kha᷄psáŋ
    6. ƙàda᷄m
    1. N distal
    2. thàl ɲɛː᷄
    3. khùm ɲɛː᷄
    4. ʒɛ̀lɓàŋ ɲɛː᷄
    5. khàpsàŋ ɲɛː᷄
    6. ƙàdàm ɲɛː᷄
    1. N medial
    2. thàl ɲɛ̀ː
    3. khùm ɲɛ̀ː
    4. ʒɛ̀lɓ àŋ ɲɛ̀ː
    5. khàpsàŋ ɲɛ̀ː
    6. ƙàdàm ɲɛ̀ː
    1. N proximal
    2. tha᷄l sɛ̄
    3. khu᷄m sɛ̄
    4. ʒɛ᷄lɓáŋ sɛ̄
    5. kha᷄psáŋ sɛ̄
    6. ƙàda᷄m sɛ̄
    1.  
    2. ‘Tal person’
    3. ‘spirit’
    4. ‘red bean’
    5. ‘hut’
    6. ‘place for drying millet’

As mentioned earlier, the tonal alternations are not limited to nominal modification with the demonstrative pronouns. In the next section, we turn to nominal modification with possessive pronouns.

3.2 Possessive pronouns as modifiers

The nouns in Tal can also be modified with the possessive pronouns. Like the demonstrative modifiers, the nouns that are modified with the possessive pronouns also undergo tonal alternation.

    1. (16)
    1. First person possessive pronouns
    1.  
    2. L-L
    3.  
    4. H-H
    5.  
    6. M-M
    7.  
    8. L-H
    9.  
    10. M-H
    11.  
    12. M-L
    13.  
    14. L-H-L
    15. HR
    16.  
    1. N
    2. phɛ̀tɛ̀j
    3. kàsɛ̀ŋ
    4. dídí
    5. mɛ́rɛ́ŋ
    6. ƙālāp
    7. ɓɔ̄lɔ̄ŋ
    8. dàgú
    9. thàƙɔ́ːp
    10. tɔ̄nín
    11. jāŋlːːt
    12. bāgàj
    13. vɛ̄ːkùː
    14. àsƙúwɛ̀t
    15. khu᷄m
    16. jɔ᷄ŋ
    1. N my
    2. phɛ̀tɛ̄j nā
    3. kàsɛ̄ŋ nā
    4. dídí ná
    5. mɛ́rɛ́ŋ ná
    6. ƙālāp nā
    7. ɓɔ̄lɔ̄ŋ nā
    8. dàgú ná
    9. thàƙɔ́ːp ná
    10. tɔ̄nín ná
    11. jāŋlːːt ná
    12. bāgāj nā
    13. vɛ̄ːkūː nā
    14. àsƙúwɛ̄t nā
    15. khu᷄m ná
    16. jɔ᷄ŋ ná
    1. N our
    2. phɛ̀tɛ̄j nū
    3. kàsɛ̄ŋ nū
    4. dídí nú
    5. mɛ́rɛ́ŋ nú
    6. ƙālāp nū
    7. ɓɔ̄lɔ̄ŋ nū
    8. dàgú nú
    9. thàƙɔ́ːp nú
    10. tɔ̄nín nú
    11. jāŋlːːt nú
    12. bāgāj nū
    13. vɛ̄ːkū nū
    14. àsƙúwɛ̄t nū
    15. khu᷄m nú
    16. jɔ᷄ŋ nú
    1. Gloss
    2. ‘bed’
    3. ‘beer filter’
    4. ‘sweat fly’
    5. ‘being sour’
    6. ‘som. used for sharing’
    7. ‘internal organ’
    8. ‘he-goat’
    9. ‘dry okra’
    10. ‘baobab’
    11. ‘terrace’
    12. ‘cave between stones’
    13. ‘clarinet’
    14. ‘type of dance’
    15. ‘spirit’
    16. ‘invitation’
    1. (17)
    1. Second person possessive pronouns
    1.  
    2. L-L
    3.  
    4. H-H
    5.  
    6. M-M
    7.  
    8. M-H
    9.  
    10. L-H
    11.  
    12. M-L
    13.  
    14. L-H-L
    15. HR
    1. N
    2. phɛ̀tɛ̀j
    3. kàsɛ̀ŋ
    4. dídí
    5. súmáː
    6. ƙālāp
    7. ɓɔ̄lɔ̄ŋ
    8. gāːlíː
    9. fjāŋƙúm
    10. thɨ̀sɨ́j
    11. thìvúk
    12. bāgàj
    13. vɛ̄ːkù
    14. àsƙúwɛ̀t
    15. khu᷄m
    1. N your(m)
    2. phɛ̀tɛ̄j gā
    3. kàsɛ̄ŋ gā
    4. dídí gá
    5. súmáː gá
    6. ƙālāp gā
    7. ɓɔ̄lɔ̄ŋ gā
    8. gāːlíː gá
    9. fjāŋƙúm gá
    10. thɨ̀sɨ́j gá
    11. thìvúk gá
    12. bāgāj gā
    13. vɛ̄ːkū gā
    14. àsƙúwɛ̄t gā
    15. khu᷄m gá
    1. N your(f)
    2. phɛ̀tɛ̄j gī
    3. kàsɛ̄ŋ gī
    4. dídí gí
    5. súmáː gí
    6. ƙālāp gī
    7. ɓɔ̄lɔ̄ŋ gī
    8. gāːlíː gí
    9. fjāŋƙúm gí
    10. thɨ̀sɨ́j gí
    11. thìvúk gí
    12. bāgāj gī
    13. vɛ̄ːkū gī
    14. àsƙúwɛ̄t gī
    15. khu᷄m gí
    1. N your(pl)
    2. phɛ̀tɛ̄j gū
    3. kàsɛ̄ŋ gū
    4. dídí gú
    5. súmáː gú
    6. ƙālāp gū
    7. ɓɔ̄lɔ̄ŋ gū
    8. gāːlíː gú
    9. fjāŋƙúm gú
    10. thɨ̀sɨ́j gú
    11. thìvúk gú
    12. bāgāj gū
    13. vɛ̄ːkū gū
    14. àsƙúwɛ̄t gū
    15. khu᷄m gú
    1. Gloss
    2. ‘bed’
    3. ‘beer filter’
    4. ‘sweat fly’
    5. ‘fainting’
    6. ‘som. used for sharing’
    7. ‘internal organ’
    8. ‘millet specie’
    9. ‘navel’
    10. ‘bed bug’
    11. ‘foam’
    12. ‘cave between stones’
    13. ‘clarinet’
    14. ‘type of dance’
    15. ‘spirit’

As shown in (16), the final L tone of the nouns surfaces as M when they are modified by a first- or second-person possessive pronoun. However, non-final L tones or any other tones are invariant under the same condition. In this sense, the first- and second person possessive pronouns are similar to the proximal demonstrative. The examples also show that the first- and second-person possessive pronouns bear the same tone as the preceding modified noun.

The modified nouns bear an L tone on all their TBUs when the modifier is a third-person possessive pronoun, as illustrated in (18). The tone lowering of nouns modified with third-person possessive pronouns is similar to that of the distal demonstrative (see §3.1).

    1. (18)
    1. Third-person possessive pronouns
    1.  
    2. L-L
    3.  
    4. H-H
    5.  
    6. M-M
    7.  
    8. L-H
    9.  
    10. M-H
    11.  
    12. M-L
    13.  
    14. L-H-L
    15. HR
    16.  
    1. N
    2. phɛ̀tɛ̀j
    3. kàsɛ̀ŋ
    4. dídí
    5. wúrúŋ
    6. kālāp
    7. ɓɔ̄lɔ̄ŋ
    8. thìvúk
    9. thhɨ̀sɨ̀j
    10. ƙɔ̄mtɨ́ŋ
    11. gāːlíː
    12. bāgàj
    13. vɛ̄ːkù
    14. àsƙúwɛ̀t
    15. ɓa᷄mlàŋ
    16. jɔ᷄ŋ
    1. N his/her
    2. phɛ̀tɛ̀j mɨ̄ːp
    3. kàsɛ̀ŋ mɨ̄ːp
    4. dìdì mɨ̄ːp
    5. wùrùŋ mɨ̄ːp
    6. kàlàp mɨ̄ːp
    7. ɓɔ̀lɔ̀ŋ mɨ̄ːp
    8. thìvùk mɨ̄ːp
    9. thɨ̀sɨ̀j mɨ̄ːp
    10. ƙɔ̀mtɨ̀ŋ mɨ̄ːp
    11. gàːlìː mɨ̄ːp
    12. bàgàj mɨ̄ːp
    13. vɛ̀ːkù mɨ̄ːp
    14. àsƙùwɛ̀t mɨ̄ːp
    15. ɓàmlàŋ mɨ̄ːp
    16. jɔ̀ŋ mɨ̄ːp
    1. N their
    2. phɛ̀tɛ̀j mɤá
    3. kàsɛ̀ŋ mɤá
    4. dìdì mɤá
    5. wùrùŋ mɤá
    6. kàlàp mɤá
    7. ɓɔ̀lɔ̀ŋ mɤá
    8. thìvùk mɤá
    9. thɨ̀sɨ̀j mɤá
    10. ƙɔ̀mbɨ̀l mɤá
    11. gàːlìː mɤá
    12. bàgàj mɤá
    13. vɛ̀ːkù mɤá
    14. àsƙùwɛ̀t mɤá
    15. ɓàmlàŋ mɤá
    16. jɔ̀ŋ mɤá
    1. Gloss
    2. ‘bed’
    3. ‘beer filter’
    4. ‘sweat fly’
    5. ‘speeding up’
    6. ‘som. used for sharing’
    7. ‘internal organ’
    8. ‘foam’
    9. ‘bed bug’
    10. ‘masquerade type’
    11. ‘millet specie’
    12. ‘cave between stones’
    13. ‘clarinet’
    14. ‘type of dance’
    15. ‘bird sp.’
    16. ‘invitation’

The tonal alternation in nominal modification is not limited to the usage of possessive and demonstrative pronouns as modifiers but also includes associative construction. In the next section, we focus on the associative constructions.

3.3 Nouns as modifiers in associative construction

The associative construction in Tal structurally contains two nouns. The first noun (N1) is the possessum, and the second noun (N2) is the possessor. Just like the nouns with pronominal modifiers, the possessum undergoes a tonal alternation. The examples in (19) illustrate the tonal alternation.

    1. (19)
    1. Associative constructions
    1.  
    2. L-L
    3.  
    4. H-H
    5.  
    6. M-M
    7.  
    8. L-H
    9.  
    10. M-L
    11.  
    12. L-H-L
    13. M-H
    1. N1
    2. phɛ̀tɛ̀j
    3. ɨ̀jɨ̀m
    4. dídí
    5. ƙwáják
    6. ƙālāp
    7. ɓɔ̄lɔ̄ŋ
    8. tìvúk
    9. tɨ̀sɨ́j
    10. ɓāɡàj
    11. vɛ̄:kù
    12. àsƙúwɛ̀t
    13. dāŋɓáŋ
    1. N2 →
    2. tɨ̀sɨ́j
    3. dídí
    4. ƙàkí:
    5. ɡwīm
    6. tìvúk
    7. ɡwīm
    8. ɡwīm
    9. tɔ̄nín
    10. ƙɔ̄mtɨ́ŋ
    11. ɡwīm
    12. ɡwīm
    13. ɡwīm
    1. N1 N2
    2. phɛ̀tɛ̀j tɨ̀sɨ́j
    3. ɨ̀jɨ̀m dídí
    4. dìdì ƙàkí:
    5. ƙwàjàk ɡwīm
    6. ƙàlàp tìvúk
    7. ɓɔ̀lɔ̀ŋ ɡwīm
    8. tìvùk ɡwīm
    9. tɨ̀sɨ̀j tɔ̄nín
    10. ɓàɡàj ƙɔ̄mtɨ́ŋ
    11. vɛ̀:kù ɡwīm
    12. àsƙùwɛ̀t ɡwīm
    13. dàŋɓàŋ ɡwīm
    1. ‘N1 of N2’
    2. ‘bed of bed bug’
    3. ‘beans of sweat fly’
    4. ‘sweat fly of sacrifice’
    5. “twisting of man’
    6. ‘something for sharing of foam’
    7. ‘internal organ of man’
    8. ‘foam of man’
    9. ‘bed bug of baobab’
    10. ‘cave stones of masquerade type’
    11. ‘clarinet of man’
    12. ‘dance-type of man’
    13. ‘animal sp. of man’

The nouns surface with an L tone on all their TBUs when they occur as the possessum, regardless of their tones in isolation. In the next section, we present the descriptive generalisations of all the nominal modification.

3.4 Descriptive generalisations

Based on the description above, the following generalisations can be made about the modification of nouns with the (pro)nominals in Tal: (i) nouns bear an L tone on all their TBUs when they are the possessum in an associative construction or modified by a medial demonstrative pronoun, a distal demonstrative pronoun or a third-person possessive pronoun, and (ii) the final L tone of a noun surfaces as M when the noun is modified by a proximal demonstrative pronoun, a first-person possessive pronoun or a second-person possessive pronoun. In Table 6, we schematically present the generalisations.

Table 6

Schematic representation of the generalisations.

M tone on the final TBU L tone on all TBUs
possessive pronouns first and second person third person
demonstrative pronouns proximal distal and medial
associative constructions possessum (N1)

In the next section, we account for the source of the tonal overwrite, the features of the overwriting tones and the realisation of the overwriting tones. Our account of the tonal overwrite will mostly draw insights from similar constructions in other Chadic languages.

4 Analysis and discussion

4.1 Tonal overwrite as effects of deictic linkers

This section focuses on the source of the tonal overwrites in the modification of nouns with possessive pronouns, demonstrative pronouns, and the possessor in an associative construction. To account for the tonal overwrites, we must refer to similar constructions in other West Chadic languages, as schematically represented in (20). In many Chadic languages, when a noun is modified by a pronominal or nominal modifier, a linker occurs between the noun and its modifier (e.g., Hoskison 1983 on Gude, Viljoen 1983 on Buwal, Cosper & Gital 2004 on Zul, Zimmermann 2008 on Hausa, and Shay 2021 on Giziga). In some Chadic languages, the linker is a floating tone which overwrites the inherent tone of the modified noun (e.g., Hellwig 2011 on Goemai and Fwangwar 2018 on Mwaghavul).

    1. (20)
    1. Noun and (pro)nominal modifiers in West Chadic languages
    2. N1 Linker N2

Based on insight from other Chadic languages, we assume that the source of the tonal alternation on Tal nouns with (pro)nominal modifiers is a linker. To distinguish between M- and L-tone overwrites, we start with the usage of the demonstrative pronouns as modifiers. Given that the modified nouns bear different tones depending on whether the modifier is a proximal or non-proximal demonstrative pronoun, we assume that Tal has two classes of linkers, namely proximal and non-proximal linkers. In this case, the choice of linker is determined by the deictic property of the nominal modifier. Following the account of spatial deixis by Fillmore (1982), we assume that the proximal and non-proximal linkers have the semantic features [+proximal] and [–proximal] respectively. The account of the demonstrative pronouns can be extended to the possessive pronouns and nouns, given that they trigger the same tonal overwrites on their modified nouns. This means that the first- and second-person possessive pronouns have proximal deictic properties and that the third-person possessive pronouns and nouns have non-proximal deictic properties, as presented in Table 7. Deixis, as a property of pronouns and nouns, is not unique to Tal but found across languages (Fillmore 1971; Levinson 2008; Ekasriadi et al. 2021).

Table 7

Deictic properties of (pro)nominal modifiers.

[+proximal] [–proximal]
possessive pronouns first and second person third person
demonstrative pronouns proximal distal and medial
associative constructions possessum (i.e., noun)

The deictic classification of the demonstrative pronouns is fairly intuitive. To support our deictic categorisation of the possessive pronouns and nouns, we refer to the deictic classification of personal pronouns and nouns crosslinguistically. Studies suggest that the distinction between the deictic roles of English personal pronouns are drawn between “primary participants”, in this case the speaker and addressee, and “others” (Lyons 1977; Thrane 1980). Consequently, our classification of first- and second-person possessive pronouns in Tal as proximal, and the third-person possessive pronoun as non-proximal is similar to the classification of English personal pronouns. If we take into account that nouns usually have third-persons as their anaphora in many languages (Mohanan 1981; Wiese 1983; Reinhart 2016), the morphophonemic grouping of nouns and third-person pronouns as non-proximal in Tal is a crosslinguistic feature. Another argument in support of our classification of the nominal modifiers in Tal as non-proximal comes from Pomo (Pomoan, Northern California USA). In Pomo, the nouns generally refer to entities outside “focal consciousness”, just like third-person pronouns and distal demonstrative pronouns (Mithun 1990). Similar to Pomo, the impersonal pronouns in other Chadic languages, which mostly emerge from nouns, refer to entities outside the focal consciousness of the speaker (Buba 1997; Pawlak 2009). Considering that the deictic usage of pronouns and nouns is context-dependent in most languages, a unique property of proximal and non-proximal forms in Tal is that they are morphophonologically marked.

Based on the distributions of the deictic linkers in Tal, the speaker is considered the deictic centre of the demonstrative pronouns, but for the possessive pronouns and nouns, the speaker and addressee are the deictic centre. This deictic classification in Tal is comparable to languages like English (Lyons 1977) and Karbi (Konnerth 2014; 2015).

We now turn to the phonological exponents of the proximal and non-proximal linkers. As shown earlier, the tonal alternation on nouns with proximal modifiers only involves realising the final L tone of the modified nouns as M. In other words, the final TBUs of nouns with proximal modifiers bear an M, H or HR tone. To capture that the final TBUs of nouns with proximal modifiers bear either of these tones, the phonological exponent of the [+proximal] linker is considered the tone feature [+raised]. Thus, the final L tone of a noun with a proximal modifier surfacing as M is as a result of overwriting the [–raised] feature of a lexical L tone with the [+raised] exponent of the proximal linker. We assume that the [+raised] exponent of the proximal linker also overwrites the [+raised] feature of the root-final lexical M, H and HR tones, but the overwrite is not auditorily apparent. Given that nouns with non-proximal modifiers bear an L tone on all their TBUs, the phonological exponent of the [–proximal] linker is considered the tone features [–upper, –raised]. The phonological exponent of the non-proximal linker could have been analysed as the primitive L tone, but we consider it to be tone features in order to unify it with the exponent of the proximal linker. In (21), we present the deictic linkers and their exponents. As we will see in §4.3, the sound-meaning association of the deictic linkers is also in line with crosslinguistic patterns of deictic iconicity.

    1. (21)
    1. Linkers as featural affixes in Tal
    1. a.
    2. b.
    1. –proximal:
    2. +proximal:
    1. [–upper, –raised][–prox]
    2. [+raised][+prox]

Figure 3 shows the pitch contours of monosyllabic words with their tones in isolation and with the tonal overwrite of the [–proximal] and [+proximal] linkers. The F0(Hz) height of the lexical tones, namely H, L, M, and HR, are distinct in isolation. In the [+proximal] position, the lexical L and M tones are raised such that they are close to the F0(Hz) height of the lexical H tone in both speech and humming modes. The F0(Hz) heights of the lexical H and HR are also raised in the [+proximal] position for the speech mode but slightly for the humming mode. In the [–proximal] position, the F0(Hz) heights of the lexical H and HR tones are as low as the F0(Hz) height of the lexical M tone. The lexical L and M tones have nearly identical F0(Hz) height in the [–proximal] position. The results are in line with our account that the [+raised] exponent of the [+proximal] linker overwrites the lexical tones in all cases, contrary to the auditory impression that the lexical M, H and HR tones are invariant.

Figure 3
Figure 3

F0 plot of lexical tones and their realisation in nominal modification.

To investigate whether the differences are significant, we extracted F0(Hz) values at 25%, 50% and 75% intervals for each of the tones in isolation, and in both [+proximal] and [–proximal] positions. Using the lme4 package (Bates et al. 2015) in R (Team 2023), the F0(Hz) values at the three intervals were fitted to a linear mixed effect model as the dependent variable. The fixed effects were the positions (isolation, [+proximal] and [–proximal]) with the isolation as the intercept and the word-types as the random effect. The results show that the [+proximal] and [–proximal] positions have a statistically significant effect (p < 0.05) on the F0(Hz) height at a minimum of two intervals for the H, L, M and HR tones. See the appendix for the results and the supplementary material for the R code. In the next section, we formally account for the phonological realisation of the deictic linkers which have the tone features [–upper, –raised][–prox] and [+raised][+prox] as their exponents respectively.

4.2 Realisation of the featural affixes

The discussion in the previous subsection indicates that the proximal and non-proximal linkers are featural affixes (Akinlabi 1996). As featural affixes, the phonological exponents of the proximal and non-proximal linkers could have been realised on the modifier or the modified noun. The underlying assumption in this work is that the linkers are affixes which take the modified nouns as their morphological stem. In this case, the domain of morphological and phonological realisation coincide. For the realisation of the featural affixes on the stem, there are four possible options: the featural affixes could have been realised on (i) the rightmost TBU of the nominal root, (ii) the leftmost TBU of the nominal root, (iii) the medial TBU of the nominal root or (iv) all the TBUs of the nominal root. The lowering of all TBUs in the context of the non-proximal modifiers indicates that the featural affix of the non-proximal linker prefers option (iv). On the other hand, the proximal feature prefers option (i), given that the tonal alternation in the context of proximal modifiers only targets the rightmost TBU of the modified nouns.

To account for the realisation of the proximal and non-proximal featural affixes, we adopt the morpheme-specific correspondence constraints, as proposed in Finley (2009). These constraints are morpheme-specific versions of Edge-Anchor and Contiguity in McCarthy & Prince (1993; 1995). The constraints require correspondence between features and edges of a relevant domain. The formulation of the constraints in (22) is specific to the [–upper, –high] features of the non-proximal linker.

    1. (22)
    1. Constraints on the [–Upper, –High] features of non-proximal linker
    1.  
    1. a.
    1. L-Anchor-[–Prox]
    2. The [–Upper, –Raised] features of the non-proximal must be in correspondence with the leftmost TBU of the root.
    1.  
    1. b.
    1. R-Anchor-[–Prox]
    2. The [–Upper, –High] features of the non-proximal must be in correspondence with the rightmost TBU of the root.
    1.  
    1. c.
    1. O-Contiguity-[–Prox]
    2. The tone feature of a TBU in correspondence with the featural affix [–Upper, –Raised] must form a contiguous string.
    1.  
    1. d.
    1. Integrity-[–Prox]
    2. No [–Upper, –High] of the non-proximal linker has multiple correspondents in the output.

The constraint L-Anchor-[–Prox] assigns no violation if the exponent of the non-proximal linker is realised on the leftmost TBU. Similarly, the constraint R-Anchor-[–Prox] also assigns no violation if the exponent of the non-proximal linker is realised on the rightmost TBU. By realising the exponent of the non-proximal linker on both edges of the root, it will automatically be realised on the medial TBU(s) as a result of adjacency and precedence relations (see Archangeli & Pulleyblank 1994; Pulleyblank 1996; Ní Chiosáin & Padgett 2001). In a situation where the exponent of the non-proximal linker is realised but not on the rightmost TBU of the root, the constraint R-Anchor-[–Prox] assigns a violation for each root TBU to the right of the realised exponent. Similarly, if the exponent of the non-proximal linker is realised but not on the leftmost TBU, the constraint L-Anchor-[–Prox] assigns a violation for each root TBU to the left of the realised exponent. As illustrated in (23a), another possible way to satisfy the constraints is to realise one copy of the exponent on the leftmost TBU and another copy on the rightmost TBU. This solution can result in the violation of the constraint Integrity-[–Prox], which prohibits the exponent of the non-proximal linker from having multiple correspondents in the output. L-Anchor-[–Prox] and R-Anchor-[–Prox] assign violation for each TBU of the stem if the exponent of the non-proximal linker is not realised in the output.

    1. (23)
    1. / thāŋdájáŋ +[–Upper, –Raised][–Prox] /… → [thàŋdàjàŋ] ‘type of raft’
    1.  
    1. a.
    1. *[thāŋdájáŋ]
    1.  
    1. b.
    1. [thàŋdàjàŋ]

Under the account of morphemic harmony in Finley (2009), the realisation of the featural affix on the medial TBU will be considered an effect of the constraint o-contiguity-[–prox], which prohibits a gapped representation. Considering that a gapped representation can create a phonetically inviable structure with contradictory precedence relations (see Pulleyblank 1996; Ní Chiosáin & Padgett 2001), Akinbo (2021a) suggests the replacement of Contiguity with Integrity for morphemic harmony. Following Akinbo (2021a), we replace the constraint o-contiguity-[–prox] with the morpheme-specific version of the constraint integrity, as in (22d).

    1. (24)
    1. *Float-T
    2. No floating tone features

Realising the featural affix can result in deleting or floating the lexical-tone features of the nominal root. For analytical reasons, we assume that the features of the lexical tones float. The floating tone features will violate the constraint *Float-T, which prohibits floating tone features (Pulleyblank 1997; Zoll 2003). For the constraint *Float-T, one violation is assigned to all the floating features in each tone.

We can account for the realisation of the non-proximal morphemic feature on all TBUs by ranking the constraints l-anchor-[–prox], r-anchor-[–prox] and integrity-[–prox] above the constraint *Float-T. The account is illustrated in (25). The lexical tone and the exponent of the non-proximal linker are indicated with numeral indexation. Featural affixation is indicated by parentheses, comparable to an autosegmental association line.

    1. (25)
    1. Non-proximal linker

For not realising the exponent of the non-proximal linker on the rightmost or leftmost edge of the root, the candidates in (25a-c,f) are ruled out. For having multiple correspondents of the phonetic exponent, the candidate in (25d) is ruled out. The candidate in (25e) wins for spreading the exponent of the non-proximal linker to all root TBUs. The candidate in (25f) and the optimal candidate are similar in terms of their surface tones, but the featural correspondence constraints are able to distinguish lexical and grammatical tones.

We now turn to the phonological realisation of the proximal linker. We assume that the proximal instantiation of the anchor constraints in (22) exists in the system. The the exponent of the proximal linker is only realised on the rightmost TBU of the nominal root is a kind of asymmetry between the exponent of the proximal and non-proximal linkers. We can account for the asymmetry with the proximal-specific instantiation of the constraints. In this case, all the proximal-specific instantiation of the constraints, except the constraint L-Anchor-[+prox], have to be ranked above *Float-T. With the ranking, as shown in (26), we can account for the phonological realisation of the proximal feature.

    1. (26)
    1. Proximal linker

We have shown that the phonological realisation of the proximal and non-proximal feature involves morphemic-featural correspondence constraints. While the realisation of the non-proximal features involves maximal extension, the phonological feature of the proximal linker is only realised on the rightmost TBU.

Realising the exponent of the proximal linker on the rightmost TBU of the word can be perceptually motivated, considering that contour tones across languages are often restricted to perceptually prominent positions, such as the initial and final TBU of a domain (Yip 1989; Zoll 1996; Gordon 2001). In the same way, the contour tone in Tal, as shown in Table 6, is restricted to word-initial position. Thus, realising the exponent of the proximal linker on the rightmost TBU of the nouns with at least two syllables can also be considered a strategy for preserving the lexical tone of the perceptually prominent position (i.e., the word-initial position). Under this account, the exponent of the proximal linker could have associated with all the TBUs, excluding the leftmost one. The realisation of the proximal feature suggests that there is an additional motivation in the language, beyond preserving the lexical tone of the perceptually prominent position. Our account is that the local and long-distance realisations of the proximal and non-proximal features, respectively, are also motivated by iconicity, which is form-meaning resemblance. To this end, we examine the form-meaning association of the deictic linkers in the next section.

4.3 Iconic status of the deictic linkers

The traditional view in modern linguistics is that form-meaning mapping in grammar is completely arbitrary (Hockett 1960; de Saussure 1974), but this view is giving way for the emerging view that both arbitrariness and iconicity play crucial roles in grammar (Lockwood & Dingemanse 2015; Dingemanse 2018). Iconic sound-meaning mapping covers various semantic categories and there is compelling evidence from distance-related words (Nichols 1971; Ultan 1978; Johansson & Zlatev 2013; Haynie et al. 2014; Wikström et al. 2023). The crosslinguistic pattern of distance-related iconicity involves associating front vowels with proximal deixes and back vowels with non-proximal deixes (Tanz 1971; Ultan 1978; Woodworth 1991; Traunmüller 1994; Johansson & Zlatev 2013; Joo et al. 2022). This pattern of distance iconicity is not limited to vowels but also includes consonants. For instance, a survey of 120 Australian languages shows that palatal and velar consonants tend to be associated with near and far words respectively (Haynie et al. 2014). However, the iconic association of distance-related words with morphophonological alternations and tone patterns is underdocumented. To this end, it is important to ask whether the exponents of the proximal and non-proximal linkers in Tal are consistent with crosslinguistic patterns of distance iconicity.

To establish whether the sound-meaning association of the deictic linkers in Tal is a kind of iconicity, we must refer to the phonetic basis of deixis-specific iconicity. Articulatory and acoustic explanations have been put forward for iconically associating front vowels (and analogously front consonants) with low magnitude and back vowels (and analogously back consonants) with high magnitude across languages. On the articulatory end, the prediction is that the magnitude of a referent directly correlates with the felt and seen degree of articulatory aperture in the segment’s production (Sapir 1929; Johansson & Zlatev 2013; Kawahara 2021). To investigate the acoustic cues that underlie magnitude iconicity, Knoeferle et al. (2017) conducted acoustic and perceptual experiments. The results of their experiments indicate that the magnitude of a referent positively correlates with the values of first formant (F1) and intensity but negatively correlates with fundamental frequency (F0) and second formant (F2): (i) high F1 (Hz) and low intensity (dB) tend to be associated with low magnitude and vice versa; (ii) low F2 (Hz) and F0 (Hz) tend to be associated with high magnitude and vice versa.

The articulatory hypothesis cannot explain the form-meaning association of the deictic linkers in Tal, given that the exponents of the deictic linkers in the language are tonal not segmental. More so, there is no evidence to suggest that the everyday user of a language is aware of the biomechanics involved in tone production. However, the prediction of the acoustic explanation holds, considering that tone lowering is associated with the non-proximal linker and tone raising with the proximal linker. Consequently, the form-meaning association of the proximal and non-proximal linkers in Tal is consistent with crosslinguistic patterns of distance-related iconicity. By this, we mean that the inverse relationship between the size of a vibrating body and its natural acoustic frequency is depicted through the form-meaning association of the deictic linkers. This indicates that the intent to create perceptuomotor analogies drives the iconic form-meaning of the deictic linkers (Dingemanse et al. 2015; Thompson et al. 2021). Ohala (1994) refers to this pattern of perceptuomotor analogies as the frequency code. The form-meaning mapping of the deictic linkers in Tal indicates that distance-related iconicity is not limited to segmental features but includes tone features.

We now turn to the long-distance and local realisations of the non-proximal and proximal features respectively. Spreading the exponent of the non-proximal linker over a long phonological distance and docking the exponent of the proximal linkers on the final TBU seem to depict the deictic properties of the referents (i.e., farness and nearness respectively). As illustrated in (27), the phonological realisation of the proximal and non-proximal features can be considered an iconic grammaticalisation of the deictic gestures, and respectively. Similar to the gestures, the locality of the deictic featural affixes in Tal calls attention to the personal and spatial properties of the referents (Hoffmeister 1978; Bangerter & Louwerse 2005; Ciciliani & Wilbur 2006; Tkachman 2022). This indicates that iconicity is not limited to the exponents of morphosyntactic elements but also includes the phonological realisation of such exponents. Associating the exponent of the proximal linker with the leftmost TBU instead of the rightmost one could still have resulted in an iconic mapping. That the exponent of the proximal linker is not linked to the the left edge (the prominent position in Tal) suggests an interplay between iconicity and language-internal condition. The exponents of the deictic linkers and their phonological realisation contribute to the typology of iconicity, considering that these patterns of iconicity are rare are undocumented in spoken language until now.

    1. (27)
    1. Iconic realisation of the deictic linkers for [àsƙúwɛ̀t] ‘type of dance’
    1.  
    1. a.
    1.  
    1. b.

The deictic linkers in Tal are similar to crosslinguistic patterns of iconicity, which also involve featural affixation. For example, the vowel alternations of Korean (Lee 1992) and Japanese (Hamano 1986) ideophones have been shown to be cases of featural affixation. Just like the iconic tonal alternations in Tal, the featural affixations of Japanese and Korean ideophones have also received theoretical attention (Akinlabi 1996; Chung 2000; Finley 2009). The expression of diminutive and augmentative in Fungwa through root-vowel fronting and backing respectively is another pattern of featural affixation with iconic motivation (Akinbo 2021a; 2023a). Our account of Tal is crucially influenced by that of the root-vowel mutation in Fungwa. Compared to discussions about other levels of grammar, the implications of iconicity for phonological grammar are one of the least explored issues in linguistic theory. In the next section, we explore what the form-meaning mapping of the deictic linkers in Tal means for phonological theories.

4.4 Theoretical implications of deictic linkers in Tal

This section discusses the implications of form-meaning mapping of the deictic linkers in Tal for phonological theory. Ohala (1994) proposes the frequency code hypothesis as the biological motivation for the iconic association of high acoustic frequency with low magnitude referent and vice versa. The frequency code hypothesis is in line with the acoustic principle that there is an inverse relationship between the size of a vibrating body (e.g., vocal cords) and the natural frequency at which the body vibrates (Titze 1989; Lee et al. 1999; Titze 2008). In this case, the association of high pitch with proximal and low pitch with non-proximal relatively evokes the natural acoustic principles. According to Kawahara (2020), this kind of natural connection between form and meaning in iconicity is a naturalness argument for the inclusion of iconicity in phonological theory. This suggests that the form-meaning mapping of the deictic linkers in Tal is phonetically natural.

Evidence for iconicity are mostly from probabilistic tendencies in lexicon across diverse samples of world languages (Hinton et al. 1994; Haynie et al. 2014; Thompson et al. 2021; Winter et al. 2022) and experimental conditions (e.g. Sapir 1929; Ramachandran & Hubbard 2001; Dingemanse et al. 2016; Rabaglia et al. 2016; Laing et al. 2017; Perry et al. 2018; Vainio 2021; Van Hoey et al. 2023). Outside reduplication and repetition (Hurch 2005), iconicity in morphophonological processes is under-represented (Alderete & Kochetov 2017; Akinbo 2021a; b). Being that iconicity has long been considered a peripheral aspect of human language, there is a reason to believe that morphophonological processes motivated by iconicity, such as the crosslinguistic patterns of expressive palatisalisation (Alderete & Kochetov 2017) and the iconic root-vowel mutation in Fungwa (Akinbo 2021a; 2023a), are underdocumented. The form-meaning associations of the proximal and non-proximal linkers in Tal are crucial evidence for iconicity in core grammar, given that the morphophonemic properties of the deictic linkers in Tal are comparable to arbitrary featural affixes (Akinlabi 1996; Ettlinger 2004). For instance, the realisation of the non-proximal linker results in morphemic harmony (Finley 2009). Similar to various patterns of arbitrary and iconic phonological processes (Knoeferle et al. 2017; Hayes 1999; Hayes et al. 2004), the deictic iconicity in Tal is phonetically grounded. Most importantly, we are able to account for the realisation of the morphemic features of the proximal and non-proximal linkers with lexically indexed constraints that have been applied to arbitrary pattern of featural affixation.

We now turn to the issue of phonological asymmetry, which is widely attested in arbitrary phonological patterns. One of the widely attested patterns of phonological asymmetry is the restriction of contour tones to long vowels, word-initial position or word-final position (Gordon 2001; Zoll 2003; Zhang 2004). Similar to contour tones, level tones also show phonological asymmetry. For example, relative to L tone (and M tone, depending on the number of tones in a language), H tone is rarely targeted for deletion and epenthesis (Pulleyblank 1986; 2004; Zoll 2003) and is often restricted to a specific position in their post-lexical distribution (Meyers 1997; Zoll 2003; Jenks & Rose 2011). Compared to arbitrary phonological patterns, phonological asymmetry is underdocumented in iconicity (Haynie et al. 2014; Akinbo 2021a). One of the few documented cases of phonological asymmetry in iconicity is the fronting and backing of non-high vowels that mark diminutive and augmentative respectively in Fungwa. The iconicity in Fungwa is consistent with crosslinguistic patterns of phonological asymmetry between high and non-high vowels, considering that the root-vowel mutations of the evaluative formation do not target high vowels in most cases (Akinbo 2021a; b; 2023a). The tonal alternation of modified nouns in Tal presents another example of phonological asymmetry in iconicity, given that the [+raised] exponent of the proximal linker is only realised on the final TBU not all root TBUs like [–upper, –raised] exponent of the non-proximal linker.

As shown earlier, the motivation for the phonological asymmetry of the deictic features in Tal is iconicity. However, in phonological theory, perceptual advantage or prominence is traditionally considered the functional source of various phonological asymmetries (Zoll 1996; Beckman 1998; Walker 2001; 2005). As argued in §4.2, the iconic realisation of the proximal feature at the right edge, rather than the left edge, is also a result of preserving the input-output mapping of tone at the perceptually prominent position, which happens to be the left edge in the language. Being limited to perceptual prominence and distinctive identification (Boersma 1998; Frisch 2004), the traditional notion of perceptual motivation in linguistic theory cannot account for phonological asymmetries that are motivated by iconicity. The basis of the iconic phonological asymmetry lies in depiction, which is the perceived or sensory resemblance between a form and its referent (Abell 2009; Hyman 2012; Dingemanse 2015). Crossmodal correspondence or similarity, as a characteristic of depictive representation, also suggests that an intention to create perceptuomotor analogies of language-external experience facilitates iconic form-meaning mapping (Dingemanse et al. 2015; Thompson et al. 2021). If we take into consideration that crossmodal depiction is the basis of iconic patterns, we see that the perceptual motivation for phonological patterns is not limited to phonetic prominence and distinctive identification but includes crossmodal perception of sensory imagery. Consequently, the asymmetry in the phonological realisation of the proximal and non-proximal features in Tal is both natural and perceptually motivated. The idea that visual sensory inputs can motivate phonological condition is not new; the visual representation of phonological features in an orthographic system has been shown to influence loanword adaptation (Vendelin & Peperkamp 2006; Detey & Nespoulous 2008; Daland et al. 2015).

Iconicity is crucial to feature theory, given that segments in both arbitrary and iconic sound-meaning mapping have distinctive features as the basis of their representation (Alderete & Kochetov 2017; Kawahara 2020; Akinbo 2021a). In addition to the featural basis of segments in arbitrary phonological patterns, tones can also have features as their basic unit (Yip 1980; Pulleyblank 1986; Clements et al. 2011; McPherson 2017), but some researchers argue that tones are basic units of phonological representation like features (Hyman 2010; Clements et al. 2011). Evidence for both views about phonological representation of tones exists in arbitrary phonological processes, but such evidence is sparse for iconicity. The [+raised] feature as the exponent of the proximal linker in Tal is an evidence for tone features in iconicity.

We argue that iconicity does not only operate at the level of features but can also have privative and binary features like arbitrary phonological contrasts (Chomsky & Halle 1968; Clements & Hume 1995). For the present purpose, we only discuss the binary aspect. By grouping some semantic categories found in crosslinguistic patterns of magnitude iconicity into low magnitude vs. high magnitude, as in size small vs. big (Sapir 1929; Ultan 1978; Shinohara & Kawahara 2010), distance near vs. far (Johansson & Zlatev 2013; Haynie et al. 2014), gender female vs. male (Pitcher et al. 2013), weight light vs. heavy (Ibarretxe-Antuñano 2017), age young vs. old (Kawahara et al. 2018), height short vs. tall (Awoyale 1989; Egbokhare 2001) and color brightness vs. darkness (Moos et al. 2014), as in Table 8, we present phonological typologies of magnitude iconicity. We do not consider the form-meaning mapping in the table to be an exhaustive list.

Table 8

Phonological typology of magnitude iconicity.

Magnitude Iconicity
Low Magnitude High Magnitude
Segmental features [–back] [+back]
[+high] [–high]
[–voiced] [+voiced]
[–son] [+son]
Tone features H(/M) L
H L(/M)
[+raised] [–raised]
[+upper] [–upper]
Phonological processes local long distance

The morphophonological properties of the deictic featural affixes in Tal do not only challenge the traditional view that form-meaning association in core grammar is completely arbitrary but also presents a grammaticalised evidence for the alternative view that form-meaning mapping involves both arbitrariness and iconicity (Dingemanse et al. 2015; Lockwood & Dingemanse 2015). While various patterns of iconicity are known for being phonologically marked (Nuckolls 1996; Dingemanse 2012; Andrason 2017; Thompson 2018), patterns of iconicity like the proximal and non-proximal linkers in Tal indicate that they are also subject to similar constraints as arbitrary phonological patterns. Phonetic and natural bases of iconicity suggest its communicative potential rests on (relative) vocal imitation (Ohala 1994; Lockwood & Dingemanse 2015). The deictic iconicity in the core grammar of Tal is in line with the hypothesis that imitation is an integral aspect of language faculty (Hauser et al. 2002).

The discussion in §4.3 indicates that the basis of various iconic patterns involves paying attention to the details of a language-external event and linguistically depicting (aspects of) the event in a perceptually similar manner (Dingemanse 2015; Hawkins et al. 2023). The perceptual similarity between the event and phonological element calls attention to the property of the event (Ohala 1994; Dingemanse 2011; Erben Johansson et al. 2020; Thompson & Do 2019). That said, iconic mapping between form and meaning might be conventional and governed by language-internal constraints (Dingemanse et al. 2015; Akinbo 2021a; Thompson et al. 2022). The depictive property of iconicity is in line with the proposal of Emergent Grammar (Hopper 1987; Bybee 2010; Mielke 2008; Archangeli & Pulleyblank 2022; 2017: etc), that grammar emerges from general cognitive properties, such as paying attention to details and similarities. That imitation also facilitates sound-meaning mapping in music is more evidence for language involving general cognitive properties (Patel & Iversen 2003; Mercado III et al. 2014; Akinbo 2023b). For example, Akinbo (2023b) finds that Yorùbá gamers map meaning to videogame music by imitating the pitch contour of music motifs with the tone melody of their verbal interpretations. This suggests that iconicity can provide insight into the longstanding debate about the cognitive basis and relation between language and music (Koelsch et al. 2004; Patel 2008; 2012; Peretz et al. 2015).

5 Summary and conclusion

We have described and analysed the tonal alternation in three patterns of nominal modifications in Tal. Nouns bear an L tone on every TBU when they are modified with non-proximal modifiers. The final tone of nouns with a proximal modifier is raised. The L-tone overwrite of nouns with a non-proximal modifier is considered an effect of a non-proximal linker which has the tone features [–upper, –raised] as its phonological exponent. The tone raising of the final TBU of a noun with a proximal modifier is considered an effect of a proximal linker which has the tone feature [+raised] as its exponent. In this case, the deictic linkers in Tal are featural affixes. We account for the phonological realisation of the deictic linkers using morpheme-specific featural correspondence constraints.

The association of tone lowering with the non-proximal linker and tone raising with the proximal linker is consistent with the crosslinguistic pattern of distance-related iconicity. Notably, realising the exponents of the non-proximal linker over a long phonological distance and the proximal linker over a short distance is a previously unreported pattern of iconicity. The form-meaning association of the deictic linkers and their realisations suggests that iconicity is an integral aspect of grammar and can motivate the locality of morphophonological features. Considering that previous studies on deictic iconicity are only based on segmental features, the grammaticalised deictic iconicity in Tal contributes to the typology of grammaticalised iconicity.

The pattern discussed in this work has only been reported in A3 West Chadic languages such as Goemai Hellwig 2011:100) and Mwaghavul (Fwangwar 2018). Even in those cases, the tonal alternation in nominal modification is only limited to the L-tone overwrite of the associative construction. It would be important to investigates whether the pattern of tone overwrite in Tal is an areal feature of the A3 West Chadic languages. As mentioned earlier, most A3 West Chadic languages are understudied; only a few of them have been formally analysed. This work is an initial attempt at describing and formally analysing these languages. For our future direction, we hope to extend our research on Tal beyond iconicity.

Appendix

The results of the linear mixed effect model are presented in this appendix. The significance codes for the model are {≤ 0.001} ***, {≤ 0.01} **, {≤ 0.05} * and {> 0.05} ns (not significant).

Table 9

The distinction between H tone in isolation, [–proximal], and [+proximal] positions.

Isolation [–proximal] [+proximal]
Humming F025% *** *** ***
F050% *** *** ***
F075% *** *** ***
Speech F025% *** *** ***
F050% *** *** ***
F075% *** *** ***
Table 10

The distinction between L tone in isolation, [–proximal], and [+proximal] positions.

Isolation [–proximal] [+proximal]
Humming F025% *** ns ***
F050% *** * ***
F075% *** *** ***
Speech F025% *** *** ***
F050% *** *** ***
F075% *** *** ***
Table 11

The distinction between M tone in isolation, [–proximal], and [+proximal] positions.

Isolation [–proximal] [+proximal]
Humming F025% *** ** ***
F050% *** * ***
F075% *** * ***
Speech F025% *** * ***
F050% *** * ***
F075% *** ns ***
Table 12

The distinction between HR tone in isolation, [–proximal], and [+proximal] positions.

Isolation [–proximal] [+proximal]
Humming F025% *** *** ns
F050% *** *** **
F075% *** * ***
Speech F025% *** ** ns
F050% *** *** *
F075% *** *** ***

Abbreviations

1 = first person, 2 = second person, 3 = third person, dem = demonstrative, dist = distal, med = medial, pl = plural, prox = proximal, sg = singular, tbu = tone-bearing unit.

Data availability

The data and other resources that form the basis of this study can be found at the following link: https://osf.io/m8apg/. The link contains the sound files of all the examples cited in this paper, the sound files for the acoustic measurements, an R script for the data visualisation and statistical model, and the source files for creating the map of Tal settlements.

Acknowledgements

We are grateful to Claire Halpert, Douglas Pulleyblank, Nathan Sanders, Nkechi Mgbodichinma Ukaegbu, and the audience at the poster session of the Annual Meeting on Phonology in 2023 (AMP2023) for their comments and suggestions that shaped aspects of this paper. We also want to thank two anonymous reviewers and Sara Finley for their reviews and comments. We express our gratitude to Jack Mahlmann for his assistance in proofreading the final draft of this paper. Any errors of fact or explanation in this paper are our own responsibility.

Competing interests

The authors have no competing interests to declare.

References

Abell, Catharine. 2009. Canny resemblance. Philosophical Review 118(2). 183–223. DOI:  http://doi.org/10.1215/00318108-2008-041

Akinbo, Samuel. 2021a. Featural affixation and sound symbolism in Fungwa. Phonology 38(4). 537–569. DOI:  http://doi.org/10.1017/S0952675721000324

Akinbo, Samuel Kayode. 2021b. Vowel harmony and some related processes in Fungwa: University of British Columbia Dissertation.

Akinbo, Samuel Kayode. 2023a. Reduplication, repetition and sound symbolism in Fungwa. Phonological Data and Analysis 5(2). 1–24. DOI:  http://doi.org/10.3765/pda.v5art2.72

Akinbo, Samuel Kayode. 2023b. Sound-meaning mapping: Verbal imitation of Super Mario music by Yorùbá gamers. Language Sciences 98. 101553. DOI:  http://doi.org/10.1016/j.langsci.2023.101553

Akinlabi, Akinbiyi. 1996. Featural affixation. Journal of Linguistics 32(2). 239–289. DOI:  http://doi.org/10.1017/S0022226700015899

Alderete, John & Kochetov, Alexei. 2017. Integrating sound symbolism with core grammar: The case of expressive palatalization. Language 93(4). 731–766. DOI:  http://doi.org/10.1353/lan.2017.0056

Andrason, Alexander. 2017. The “exotic” nature of ideophones–from Khoekhoe to Xhosa. Stellenbosch Papers in Linguistics 48(1). 139–150. DOI:  http://doi.org/10.5774/48-0-286

Archangeli, Diana & Pulleyblank, Douglas. 1994. Grounded phonology. Cambridge, MA: MIT Press.

Archangeli, Diana & Pulleyblank, Douglas. 2017. Phonology as an emergent system. In Hannahs, Stephen J. & Bosch, Anna R. K. (eds.), The routledge handbook of phonological theory, 476–503. London: Routledge. DOI:  http://doi.org/10.4324/9781315675428-17

Archangeli, Diana & Pulleyblank, Douglas. 2022. Emergent phonology, vol. 7. Berlin: Language Science Press.

Awoyale, Yiwola. 1989. Reduplication and the status of ideophones in Yoruba. Journal of West African Languages 19(1). 15–34.

Baklit, Gideon. 2014. Tal: History and location, regaining lost ground. Bloomington: Author House.

Bangerter, Adrian & Louwerse, Max M. 2005. Focusing attention with deictic gestures and linguistic expressions. In Proceedings of the Annual Meeting of the Cognitive Science Society, vol. 27.

Bates, Douglas & Maechler, Martin & Bolker, Ben & Walker, Steven. 2015. Fitting linear mixed-effects models using lme4. Journal of Statistical Software 67(1). 1–48. DOI:  http://doi.org/10.18637/jss.v067.i01

Beckman, Jill N. 1998. Positional faithfulness: University of Massachusetts Dissertation.

Boersma, Paul. 1998. Functional phonology: University of Amsterdam Dissertation.

Buba, Malami. 1997. The deictic particle di-n in Hausa. African Languages and Cultures 10(1). 29–45. DOI:  http://doi.org/10.1080/09544169708717811

Bulkaam, Michael S. 2017. Tense and aspect in Tal: Theological College of Northern Nigeria, Bukuru-Jos BA Thesis.

Bulkaam, Michael S. 2022. Plural verbs in Tal in the context of West Chadic languages: Theological College of Northern Nigeria, Bukuru-Jos MA Thesis.

Bybee, Joan. 2010. Language, usage and cognition. Cambridge: Cambridge University Press. DOI:  http://doi.org/10.1017/CBO9780511750526

Chomsky, Noam & Halle, Morris. 1968. The sound pattern of English. New York: Harper and Row.

Chung, Chin-Wan. 2000. An optimality-theoretic account of vowel harmony in Korean ideophones. Studies in Phonetics, Phonology and Morphology 6. 431–450.

Ciciliani, Tamara Alibaić & Wilbur, Ronnie B. 2006. Pronominal system in Croatian Sign Language. Sign Language & Linguistics 9(1–2). 95–132. DOI:  http://doi.org/10.1075/sll.9.1-2.07ali

Clements, George N. & Hume, Elizabeth V. 1995. The internal organization of speech sounds. In Goldmsith, John & Yu, Alan & Riggle, Jason (eds.), The handbook of phonological theory, 245–317. Malden, MA: Wiley-Blackwell 2nd edn.

Clements, George N. & Michaud, Alexis & Patin, Cédric. 2011. Do we need tone features. In Goldsmith, John A. & Hume, Elizabeth & Wetzels, Leo (eds.), Tones and features: phonetic and phonological perspectives, 3–24. Berlin: De Gruyter Mouton. DOI:  http://doi.org/10.1515/9783110246223.3

Cosper, Ronald & Gital, Garba Mohammed. 2004. Genitive constructions in South Bauchi (West Chadic) languages, Zul and Polchi, with comparisons to ancient Egyptian. In Takács, Gábor (ed.), Egyptian and Semito-Hamitic (Afro-Asiatic) studies in Memoriam Werner Vycichl, 286–295. Leiden/Boston: Brill. DOI:  http://doi.org/10.1163/9789047412236_027

Daland, Robert & Oh, Mira & Kim, Syejeong. 2015. When in doubt, read the instructions: Orthographic effects in loanword adaptation. Lingua 159. 70–92. DOI:  http://doi.org/10.1016/j.lingua.2015.03.002

de Saussure, Ferdinand. 1974. Course in general linguistics. London: Fontana.

Detey, Sylvain & Nespoulous, Jean-Luc. 2008. Can orthography influence second language syllabic segmentation?: Japanese epenthetic vowels and french consonantal clusters. Lingua 118(1). 66–81. DOI:  http://doi.org/10.1016/j.lingua.2007.04.003

Dingemanse, Mark. 2011. Ezra pound among the Mawu: Ideophones and iconicity in Siwu. In Michelucci, Pascal & Ljungberg, Christina & Fischer, Olga (eds.), Semblance and signification, 39–54. Amsterdam: John Benjamins. DOI:  http://doi.org/10.1075/ill.10.03din

Dingemanse, Mark. 2012. Advances in the cross-linguistic study of ideophones. Language and Linguistics Compass 6(10). 654–672. DOI:  http://doi.org/10.1002/lnc3.361

Dingemanse, Mark. 2015. Ideophones and reduplication: Depiction, description, and the interpretation of repeated talk in discourse. Studies in Language 39(4). 946–970. DOI:  http://doi.org/10.1075/sl.39.4.05din

Dingemanse, Mark. 2018. Redrawing the margins of language: Lessons from research on ideophones. Glossa: a journal of general linguistics 3(1). DOI:  http://doi.org/10.5334/gjgl.444

Dingemanse, Mark. 2019. Ideophone as a comparative concept. In Akita, Kimi & Pardeshi, Prashant (eds.), Ideophones, mimetics, and expressives, vol. 16, 13–33. Amsterdam: John Benjamins. DOI:  http://doi.org/10.1075/ill.16.02din

Dingemanse, Mark & Blasi, Damián E. & Lupyan, Gary & Christiansen, Morten H. & Monaghan, Padraic. 2015. Arbitrariness, iconicity, and systematicity in language. Trends in Cognitive Sciences 19(10). 603–615. DOI:  http://doi.org/10.1016/j.tics.2015.07.013

Dingemanse, Mark & Schuerman, Will & Reinisch, Eva & Tufvesson, Sylvia & Mitterer, Holger. 2016. What sound symbolism can and cannot do: Testing the iconicity of ideophones from five languages. Language 92(2). e117–e133. DOI:  http://doi.org/10.1353/lan.2016.0034

Doupe, Allison J. & Kuhl, Patricia K. 1999. Birdsong and human speech: common themes and mechanisms. Annual Review of Neuroscience 22(1). 567–631. DOI:  http://doi.org/10.1146/annurev.neuro.22.1.567

Egbokhare, Francis O. 2001. Phonosemantic correspondences in Emai attributive ideophones. In Voeltz, FK Erhard & Kilian-Hatz, Christa (eds.), Ideophones, vol. 44, 87–96. Amsterdam/Philadelphia: John Benjamins Publishing. DOI:  http://doi.org/10.1075/tsl.44.08egb

Ekasriadi, Ida Ayu Agung & Artawa, Ketut & Dhanawaty, Ni Made & Sutama, Putu. 2021. Personal deixis reversal of Balinese language. The International Journal of Language and Cultural (TIJOLAC) 3(2). 72–85.

Erben Johansson, Niklas & Anikin, Andrey & Carling, Gerd & Holmer, Arthur. 2020. The typology of sound symbolism: Defining macro-concepts via their semantic and phonetic features. Linguistic Typology 24(2). 253–310. DOI:  http://doi.org/10.1515/lingty-2020-2034

Ettlinger, Marc. 2004. Aspect in Mafa: an intriguing case of featural affixation. In Adams, Nikki & Cooper, Adam & Parrill, Fey & Wier, Thomas (eds.), Proceedings of the Annual Meeting of the Chicago Linguistic Society, vol. 40, 73–86. Chicago Linguistic Society.

Fillmore, Charles J. 1971. Lectures on deixis. Stanford, CA: CSLI Publications.

Fillmore, Charles J. 1982. Towards a descriptive framework for spatial deixis. In Jarvella, Robert J. & Klein, Wolfgang (eds.), Speech, place, and action: studies in deixis and related topics, 31–59. Chichester, New York: Wiley.

Finley, Sara. 2009. Morphemic harmony as featural correspondence. Lingua 119(3). 478–501. DOI:  http://doi.org/10.1016/j.lingua.2008.09.009

Frisch, Stefan A. 2004. Language processing and segmental ocp effects. In Hayes, Bruce & Kirchner, Robbert & Steriade, Donca (eds.), Phonetically based phonology, vol. 346, 346––371. Cambridge, UK: Cambridge University Press. DOI:  http://doi.org/10.1017/CBO9780511486401.011

Fwangwar, Tongpan Rabo. 2018. Aspects of the phonology of Mwaghavul: University of Ilorin MA Thesis.

Gordon, Matthew. 2001. A typology of contour tone restrictions. Studies in Language 25(3). 423–462. DOI:  http://doi.org/10.1075/sl.25.3.03gor

Hamano, Shoko Saito. 1986. The sound-symbolic system of Japanese (ideophones, onomatopoeia, expressives, iconicity): University of Florida Dissertation.

Hauser, Marc D. & Chomsky, Noam & Fitch, W. Tecumseh. 2002. The faculty of language: what is it, who has it, and how did it evolve? Science 298(5598). 1569–1579. DOI:  http://doi.org/10.1126/science.298.5598.1569

Hawkins, Robert D. & Sano, Megumi & Goodman, Noah D. & Fan, Judith E. 2023. Visual resemblance and interaction history jointly constrain pictorial meaning. Nature Communications 14(1). 2199. DOI:  http://doi.org/10.1038/s41467-023-37737-w

Hayes, Bruce. 1989. Compensatory lengthening in moraic phonology. Linguistic Inquiry 20(2). 253–306.

Hayes, Bruce & Kirchner, Robert & Steriade, Donca. 2004. Phonetically based phonology. Cambridge: Cambrige University Press. DOI:  http://doi.org/10.1017/CBO9780511486401

Hayes, Bruce P. 1999. Phonetically driven phonology: The role of optimality theory and inductive grounding. In Darnell, Michael & Moravcsik, Edith A. & Noonan, Michael & Newmeyer, Frederick J. & Wheatley, Kathleen (eds.), Functionalism and formalism in linguistics, 243–285. Amsterdam/Philadelphia: John Benjamins Publishing Co.

Haynie, Hannah & Bowern, Claire & LaPalombara, Hannah. 2014. Sound symbolism in the languages of Australia. PLoS One 9(4). e92852. DOI:  http://doi.org/10.1371/journal.pone.0092852

Hellwig, Birgit. 2011. A grammar of Goemai. Berlin/Boston: De Gruyter Mouton. DOI:  http://doi.org/10.1515/9783110238297

Hinton, Leanne & Nichols, Johanna & Ohala, John J. 1994. Introduction: Sound-symbolic processes. In Hinton, Leanne & Nichols, Johanna & Ohala, John J. (eds.), Sound symbolism, 1–14. Cambridge: Cambridge University Press. DOI:  http://doi.org/10.1017/CBO9780511751806.001

Hockett, Charles F. 1960. The origin of speech. Scientific American 203(3). 88–97. DOI:  http://doi.org/10.1038/scientificamerican0960-88

Hoffmeister, Robert James. 1978. The development of demonstrative pronouns, locatives and personal pronouns in the acquisition of American Sign Language by deaf children of deaf parents. University of Minnesota.

Hombert, Jean-Marie. 1977. Consonant types, vowel height and tone in Yoruba. Studies in African Linguistics 33. 173–190.

Hopper, Paul. 1987. Emergent grammar. In Aske, Jon & Beery, Natasha & Michaelis, Laura & Hana, Filip (eds.), Annual Meeting of the Berkeley Linguistics Society, vol. 13. 139–157. Berkley, California: Berkeley Linguistics Society. DOI:  http://doi.org/10.3765/bls.v13i0.1834

Hoskison, James Taylor. 1983. A grammar and dictionary of the Gude language (Nigeria, Cameroon, Chadic): The Ohio State University Dissertation.

Hurch, Bernhard. 2005. Studies on reduplication. Berlin: Mouton de Gruyter. DOI:  http://doi.org/10.1515/9783110911466

Hyman, John. 2012. Depiction. Royal Institute of Philosophy Supplements 71. 129–150. DOI:  http://doi.org/10.1017/S1358246112000276

Hyman, Larry M. 1985. A theory of phonological weight. Dordrecht: Foris. DOI:  http://doi.org/10.1515/9783110854794

Hyman, Larry M. 2010. Do tones have features. In Goldsmith, John A. & Hume, Elizabeth & Wetzels, Leo (eds.), Tones and features: Phonetic and phonological perspectives, vol. 107, 50–80. Berlin: De Gruyter Mouton. DOI:  http://doi.org/10.1515/9783110246223.50

Ibarretxe-Antuñano, Iraide. 2017. Basque ideophones from a typological perspective. Canadian Journal of Linguistics/Revue Canadienne de Linguistique 62(2). 196–220. DOI:  http://doi.org/10.1017/cnj.2017.8

Jenks, Peter & Rose, Sharon. 2011. High tone in Moro: Effects of prosodic categories and morphological domains. Natural Language & Linguistic Theory 29(1). 211–250. DOI:  http://doi.org/10.1007/s11049-011-9120-x

Johansson, Niklas & Zlatev, Jordan. 2013. Motivations for sound symbolism in spatial deixis: A typological study of 101 languages. The Public Journal of Semiotics 5(1). 1–20. DOI:  http://doi.org/10.37693/pjos.2013.5.9668

Joo, Ian & Hsu, Yu-Yin & Chersoni, Emmanuele. 2022. Iconic bias in Italian spatial demonstratives. Australian Journal of Linguistics 1–18. DOI:  http://doi.org/10.1080/07268602.2022.2056410

Jungraithmayr, Hermann & Holubova, Miroslava. 2016. The Ngas language (Shik Ngas): Fundamentals of grammar - texts - dictionary. Berlin: Dietrich Reimer Verlag.

Kawahara, Shigeto. 2020. Sound symbolism and theoretical phonology. Language and Linguistics Compass 14(8). e12372. DOI:  http://doi.org/10.1111/lnc3.12376

Kawahara, Shigeto. 2021. Phonetic bases of sound symbolism: A review. Manuscript. https://psyarxiv.com/fzvsu. DOI:  http://doi.org/10.31234/osf.io/fzvsu

Kawahara, Shigeto & Noto, Atsushi & Kumagai, Gakuji. 2018. Sound symbolic patterns in Pokémon names. Phonetica 75(3). 219–244. DOI:  http://doi.org/10.1159/000484938

Knoeferle, Klemens & Li, Jixing & Maggioni, Emanuela & Spence, Charles. 2017. What drives sound symbolism? different acoustic cues underlie sound-size and sound-shape mappings. Scientific Reports 7(1). 1–11. DOI:  http://doi.org/10.1038/s41598-017-05965-y

Koelsch, Stefan & Kasper, Elisabeth & Sammler, Daniela & Schulze, Katrin & Gunter, Thomas & Friederici, Angela D. 2004. Music, language and meaning: brain signatures of semantic processing. Nature Neuroscience 7(3). 302–307. DOI:  http://doi.org/10.1038/nn1197

Konnerth, Linda Anna. 2014. A grammar of Karbi: University of Oregon Dissertation.

Konnerth, Linda Anna. 2015. A new type of convergence at the deictic center: Second person and cislocative in Karbi (Tibeto-Burman). Studies in Language 39(1). 24–45. DOI:  http://doi.org/10.1075/sl.39.1.02kon

Kuhl, Patricia K. & Meltzoff, Andrew N. 1996. Infant vocalizations in response to speech: Vocal imitation and developmental change. The Journal of the Acoustical Society of America 100(4). 2425–2438. DOI:  http://doi.org/10.1121/1.417951

Laing, Catherine E. & Vihman, Marilyn & Keren-Portnoy, Tamar. 2017. How salient are onomatopoeia in the early input? a prosodic analysis of infant-directed speech. Journal of Child Language 44(5). 1117–1139. DOI:  http://doi.org/10.1017/S0305000916000428

Lee, Jin-Seong. 1992. Phonology and sound symbolism of Korean ideophones: Dissertation.

Lee, Sungbok & Potamianos, Alexandros & Narayanan, Shrikanth. 1999. Acoustics of children’s speech: Developmental changes of temporal and spectral parameters. The Journal of the Acoustical Society of America 105(3). 1455–1468. DOI:  http://doi.org/10.1121/1.426686

Levinson, Stephen C. 2008. The handbook of pragmatics. Cambridge: Cambridge University Press.

Lockwood, Gwilym & Dingemanse, Mark. 2015. Iconicity in the lab: A review of behavioral, developmental, and neuroimaging research into sound-symbolism. Frontiers in Psychology 6. 1246. DOI:  http://doi.org/10.3389/fpsyg.2015.01246

Lockwood, Gwilym & Tuomainen, Jyrki. 2015. Ideophones in Japanese modulate the p2 and late positive complex responses. Frontiers in Psychology 6. 933. DOI:  http://doi.org/10.3389/fpsyg.2015.00933

Lyons, John. 1977. Semantics, vol. 2. Cambridge: Cambridge University Press.

Maddieson, Ian. 1984. The effects on f0 of a voicing distinction in sonorants and their implications for a theory of tonogenesis. Journal of Phonetics 12(1). 9–15. DOI:  http://doi.org/10.1016/S0095-4470(19)30845-9

McCarthy, John J. & Prince, Alan. 1993. Prosodic morphology: Constraint interaction and satisfaction. Manuscript University of Massachusetts.

McCarthy, John J. & Prince, Alan. 1995. Faithfulness and reduplicative identity. In Beckman, Jill N. & Dickey, Laura Walsh & Urbanczyk, Suzanne (eds.), Papers in Optimality Theory, 249–384. GLSA.

McPherson, Laura. 2017. Tone features revisited: Evidence from Seenku. In Payne, Doris L. & Pacchiarotti, Sara & Bosire, Mokaya (eds.), Diversity in african languages: Selected papers from the 46th Annual Conference on African Linguistics, 5–21. Berlin: Language Science Press.

Mercado III, Eduardo & Mantell, James T. & Pfordresher, Peter Q. 2014. Imitating sounds: A cognitive approach to understanding vocal imitation. Comparative Cognition & Behavior Reviews 9. DOI:  http://doi.org/10.3819/ccbr.2014.90002

Meyers, Scott. 1997. OCP effects in optimality theory. Natural Language & Linguistic Theory 15(4). 847–892. DOI:  http://doi.org/10.1023/A:1005875608905

Mielke, Jeff. 2008. The emergence of distinctive features. Oxford: Oxford University Press. DOI:  http://doi.org/10.1093/oso/9780199207916.001.0001

Mithun, Marianne. 1990. Third-person reference and the function of pronouns in Central Pomo natural speech. International Journal of American Linguistics 56(3). 361–376. DOI:  http://doi.org/10.1086/466163

Mohanan, Karuvannur Puthanveettil. 1981. Grammatical relations and anaphora in Malayalam: Massachusetts Institute of Technology Dissertation.

Moos, Anja & Smith, Rachel & Miller, Sam R. & Simmons, David R. 2014. Cross-modal associations in synaesthesia: Vowel colours in the ear of the beholder. i-Perception 5(2). 132–142. DOI:  http://doi.org/10.1068/i0626

Ní Chiosáin, Máire & Padgett, Jaye. 2001. Markedness, segment realization, and locality in spreading, 118–156. Cambridge: Cambridge University Press Cambridge. DOI:  http://doi.org/10.1017/CBO9780511570582.005

Nichols, Johanna. 1971. Diminutive consonant symbolism in Western North America. Language 47(4). 826–848. DOI:  http://doi.org/10.2307/412159

Nuckolls, Janis B. 1996. Sounds like life: Sound-symbolic grammar, performance, and cognition in Pastaza Quechua, vol. 2. Oxford: Oxford University Press. DOI:  http://doi.org/10.1093/oso/9780195089851.001.0001

Nuckolls, Janis B. 1999. The case for sound symbolism. Annual Review of Anthropology 28(1). 225–252. DOI:  http://doi.org/10.1146/annurev.anthro.28.1.225

Ohala, John J. 1994. The frequency code underlies the sound-symbolic use of voice pitch. In Hinton, Leanne & Nichols, Johanna & Ohala, John J. (eds.), Sound symbolism, 325–347. Cambridge: Cambridge University Press. DOI:  http://doi.org/10.1017/CBO9780511751806.022

Patel, Aniruddh D. 2008. Music, language, and the brain. New York: Oxford University Press.

Patel, Aniruddh D. 2012. Language, music, and the brain: a resource-sharing framework. In Rebuschat, Patrick & Rohrmeier, Martin & Hawkins, John A. & Cross, Ian (eds.), Language and music as cognitive systems, 204–223. Oxford: Oxford University Press. DOI:  http://doi.org/10.1093/acprof:oso/9780199553426.003.0022

Patel, Aniruddh D. & Iversen, John R. 2003. Acoustic and perceptual comparison of speech and drum sounds in the North Indian tabla tradition: An empirical study of sound symbolism. In Solé, Maria-Josep & Recasens, Daniel & Romero, Joaquin (eds.), Proceedings of the 15th International Congress of Phonetic Sciences (icphs), 925–928. Adelaide: Causal Productions Pty Ltd.

Pawlak, Nina. 2009. Impersonal subject in Hausa and other Chadic languages. In Baldi, Sergio (ed.), VIII Afro-Asiatic Congress Napoli, vol. 6, 195–210. Neapol: Universita degli Studi di Napoli.

Peretz, Isabelle & Vuvan, Dominique & Lagrois, Marie-Élaine & Armony, Jorge L. 2015. Neural overlap in processing music and speech. Philosophical Transactions of the Royal Society B: Biological Sciences 370(1664). 20140090. DOI:  http://doi.org/10.1098/rstb.2014.0090

Perniss, Pamela & Thompson, Robin L. & Vigliocco, Gabriella. 2010. Iconicity as a general property of language: evidence from spoken and signed languages. Frontiers in Psychology 1. 227. DOI:  http://doi.org/10.3389/fpsyg.2010.00227

Perry, Lynn K. & Perlman, Marcus & Winter, Bodo & Massaro, Dominic W. & Lupyan, Gary. 2018. Iconicity in the speech of children and adults. Developmental Science 21(3). e12572. DOI:  http://doi.org/10.1111/desc.12572

Pitcher, Benjamin J. & Mesoudi, Alex & McElligott, Alan G. 2013. Sex-biased sound symbolism in English-language first names. PLoS One 8(6). e64825. DOI:  http://doi.org/10.1371/journal.pone.0064825

Prince, Alan & Smolensky, Paul. 2004. Optimality theory: Constraint interaction in generative grammar. Malden, MA: Blackwell. DOI:  http://doi.org/10.1002/9780470759400

Pulleyblank, Douglas. 1986. Tone in lexical phonology. Dordrecht, Holland: D. Reidel Publishing Company. DOI:  http://doi.org/10.1007/978-94-009-4550-0

Pulleyblank, Douglas. 1996. Neutral vowels in optimality theory: A comparison of Yoruba and Wolof. Canadian Journal of Linguistics/Revue Canadienne de Linguistique 41(4). 295–347. DOI:  http://doi.org/10.1017/S0008413100016601

Pulleyblank, Douglas. 1997. Optimality theory and features. In Archangeli, Diana & Langendoen, D. Terence (eds.), Optimality theory: An overview, 59–101. Oxford: Blackwell.

Pulleyblank, Douglas. 2004. A note on tonal markedness in Yoruba. Phonology 21(3). 409–425. DOI:  http://doi.org/10.1017/S0952675704000326

Rabaglia, Cristina D. & Maglio, Sam J. & Krehm, Madelaine & Seok, Jin H. & Trope, Yaacov. 2016. The sound of distance. Cognition 152. 141–149. DOI:  http://doi.org/10.1016/j.cognition.2016.04.001

Ramachandran, Vilayanur S. & Hubbard, Edward M. 2001. Synaesthesia–a window into perception, thought and language. Journal of Consciousness Studies 8(12). 3–34.

Reinhart, Tanya. 2016. Anaphora and semantic interpretation. London & New York: Routledge. DOI:  http://doi.org/10.4324/9781315536965

Sapir, Edward. 1929. A study in experimental symbolism. Journal of Experimental Psychology 12. 225–239. DOI:  http://doi.org/10.1037/h0070931

Shay, Erin. 2021. A grammar of Giziga: A Chadic language of far north Cameroon. Leiden/Boston: Brill. DOI:  http://doi.org/10.1163/9789004445970

Shinohara, Kazuko & Kawahara, Shigeto. 2010. A cross-linguistic study of sound symbolism: The images of size. In Rolle, Nicholas & Steman, Jeremy & Sylak-Glassman, John (eds.), Annual Meeting of the Berkeley Linguistics Society, vol. 36, 396–410. Berkeley, CA: Berkeley Linguistics Society. DOI:  http://doi.org/10.3765/bls.v36i1.3926

Tanz, Christine. 1971. Sound symbolism in words relating to proximity and distance. Language and speech 14(3). 266–276. DOI:  http://doi.org/10.1177/002383097101400307

Team, RStudio. 2023. Rstudio: Integrated development for R. PBC, Boston, MA: RStudio. http://www.rstudio.com/.

Thompson, Arthur L. & Chan, May Pik Yu & Yeung, Ping Hei & Do, Youngah. 2022. Structural markedness and depiction: The case of lower sequential predictability in Cantonese ideophones. The Mental Lexicon 17(2). 300–324. DOI:  http://doi.org/10.1075/ml.21016.tho

Thompson, Arthur Lewis. 2018. Are tones in the expressive lexicon iconic? evidence from three Chinese languages. PloS one 13(12). e0204270. DOI:  http://doi.org/10.1371/journal.pone.0204270

Thompson, Arthur Lewis & Do, Youngah. 2019. Defining iconicity: An articulationbased methodology for explaining the phonological structure of ideophones. Glossa: a journal of general linguistics 4(1). DOI:  http://doi.org/10.5334/gjgl.872

Thompson, Arthur Lewis & Van Hoey, Thomas & Do, Youngah. 2021. Articulatory features of phonemes pattern to iconic meanings: evidence from cross-linguistic ideophones. Cognitive Linguistics 32(4). 563–608. DOI:  http://doi.org/10.1515/cog-2020-0055

Thrane, Torben. 1980. Referential-semantic analysis: aspects of a theory of linguistic reference. Cambridge: Cambridge University Press. DOI:  http://doi.org/10.1017/CBO9780511753145

Titze, Ingo R. 1989. Physiologic and acoustic differences between male and female voices. The Journal of the Acoustical Society of America 85(4). 1699–1707. DOI:  http://doi.org/10.1121/1.397959

Titze, Ingo R. 2008. The human instrument. Scientific American 298(1). 94–101. DOI:  http://doi.org/10.1038/scientificamerican0108-94

Tkachman, Oksana. 2022. Embodiment and emergent phonology in the visual-manual modality: factors enabling sublexical componentiality: University of British Columbia Dissertation.

Traunmüller, Hartmut. 1994. Sound symbolism in deictic words. In Aili, Hans & af Trampe, Peter (eds.), Tongues and texts unlimited. studies in honour of Tore Jansson on the occasion of his sixtieth anniversary, 213–234. Stockholm: Stockholm University.

Ultan, Russell. 1978. Size-sound symbolism. In Greenberg, Joseph (ed.), Universals of human language, vol. 2, 525–568. Stanford: Stanford University Press.

Vainio, Lari. 2021. Magnitude sound symbolism influences vowel production. Journal of Memory and Language 118. 104213. DOI:  http://doi.org/10.1016/j.jml.2020.104213

Van Hoey, Thomas & Thompson, Arthur L. & Do, Youngah & Dingemanse, Mark. 2023. Iconicity in ideophones: Guessing, memorizing, and reassessing. Cognitive Science 47(4). e13268. DOI:  http://doi.org/10.1111/cogs.13268

Vendelin, Inga & Peperkamp, Sharon. 2006. The influence of orthography on loanword adaptations. Lingua 116(7). 996–1007. DOI:  http://doi.org/10.1016/j.lingua.2005.07.005

Viljoen, Melanie. 1983. A grammatical description of the Buwal language: La Trobe University, Melbourne Dissertation.

Walker, Rachel. 2001. Round licensing, harmony, and bisyllabic triggers in Altaic. Natural Language & Linguistic Theory 19(4). 827–878. DOI:  http://doi.org/10.1023/A:1013349100242

Walker, Rachel. 2005. Weak triggers in vowel harmony. Natural Language & Linguistic theory 23(4). 917–989. DOI:  http://doi.org/10.1007/s11049-004-4562-z

Whalen, Douglas H. & Levitt, Andrea G. 1995. The universality of intrinsic F0 of vowels. Journal of Phonetics 23(3). 349–366. DOI:  http://doi.org/10.1016/S0095-4470(95)80165-0

Wiese, Bernd. 1983. Anaphora by pronouns. Linguistics 21(1). 373–418. DOI:  http://doi.org/10.1515/ling.1983.21.2.373

Wikström, Alexandra & Vainio, Lari & Vainio, Martti. 2023. Spatial sound symbolism: A cross-linguistic study. In Skarnitzl, Radek & Volín, Jan (eds.), Proceedings of the 20th International Congress of Phonetic Sciences, Prague: Guarant International.

Williamson, Kay. 1984. Practical orthography in Nigeria. Ibadan: Heinemann Educational Books.

Winter, Bodo & Sóskuthy, Márton & Perlman, Marcus & Dingemanse, Mark. 2022. Trilled/r/is associated with roughness, linking sound and touch across spoken languages. Scientific Reports 12(1). 1035. DOI:  http://doi.org/10.1038/s41598-021-04311-7

Woodworth, Nancy L. 1991. Sound symbolism in proximal and distal forms. DOI:  http://doi.org/10.1515/ling.1991.29.2.273

Yip, Moira. 1989. Contour tones. Phonology 6(1). 149–174. DOI:  http://doi.org/10.1017/S095267570000097X

Yip, Moira. 2002. Tone. Cambridge: Cambridge University Press.

Yip, Moira Jean. 1980. The tonal phonology of Chinese: Massachusetts Institute of Technology Dissertation. DOI:  http://doi.org/10.3406/clao.1980.1072

Zhang, Jie. 2004. The role of contrast-specific and language-specific phonetics in contour tone distribution. In Hayes, Bruce & Kirchner, Robert & Steriade, Donca (eds.), Phonetically based phonology, 157–190. Cambridge: Cambrige University Press. DOI:  http://doi.org/10.1017/CBO9780511486401.006

Zimmermann, Malte. 2008. Quantification in Hausa. In Matthewson, Lisa (ed.), Quantification: Universals and variation, vol. 64, 415–475. West Yorkshire: Emerald Group Publishing Limited.

Zoll, Cheryl Cydney. 1996. Parsing below the segment in a constraint based framework: UC Berkeley Dissertation.

Zoll, Cheryl Cydney. 2003. Optimal tone mapping. Linguistic Inquiry 34(2). 225–268. DOI:  http://doi.org/10.1162/002438903321663398