What’s in a mimetic?
On the dynamicity of its iconic stem*
Takeshi USUKI Kimi AKITA
Kyoto Sangyo University Osaka University
Abstract
This paper explores the fundamental semantic and syntactic properties of Japanese mimetic lexemes as iconic signs that depict various eventualities by means of linguistic sound. We argue how the two central features of mimetics—stem-based morphology and dynamicity—restrict their morphosyntactic and semantic realizations. The discussion on the impossible uses of mimetics, such as intrinsically static adjectival expressions and object-oriented depictives, particularly clarifies the limits of the traditional root-based analysis of mimetic morphology and sets the basis for its theoretical treatment.
1. Introduction
In this paper, we delve into the ever under-investigated lexical content of Japanese mimetics and propose an integrated account of their possible and impossible uses. Mimetics, which are also known as ideophones, are inherently iconic words that form a highly productive and creative lexical class in Japanese (Kakehi et al. 1996; Hamano 1998) and some other languages (Hinton et al. 1994; Voeltz and Kilian-Hatz 2001; Ibarretxe-Antuñano 2006; Dingemanse 2011). Japanese mimetics cover a wide range of semantic domains such as sound (e.g., buhibuhi ‘oink-oink’), motion (e.g., burari
‘strolling’), texture (e.g., kasakasa ‘rustling, dry’), bodily sensation (e.g., zukin ‘one’s head throbbing’), and emotion (e.g., unzari ‘fed up’), and mainly, but not exclusively,
* We appreciate the constructive comments we received from the audience at the 9th International Symposium on Iconicity in Language and Literature. We are also grateful to the two anonymous reviewers for their insightful suggestions. All remaining inadequacies are ours. This study was supported in part by Grant-in-Aid for Young Scientists (B) (no. 24720179) and Grant-in-Aid for Scientific Research (C) (no. 25370425) from Japan Society for the Promotion of Science.
[SEMI-FINAL VERSION] Usuki, Takeshi, and Kimi Akita. 2015. What’s in a mimetic?: On the dynamicity of its iconic stem. In Masako K. Hiraga, William J. Herlofsky, Kazuko Shinohara, and Kimi Akita, eds., Iconicity: East Meets West, 109-123. (Iconicity in Language and Literature 14.) Amsterdam/Philadelphia: John Benjamins.
function as (manner) adverbs (Hamano 1998; Tamori and Schourup 1999).
This paper points out two common features underlying the semantic and syntactic diversity of mimetics: 1) the stem-based, rather than root-based, morphology, which is tightly paired with the aspectual-semantic system of mimetics, and 2) the fundamentally dynamic nature of mimetics as iconic expressions that “mimic” referent eventualities acoustically. We present the basis of an integrated iconic model that builds upon these features, and argue that it straightforwardly captures various formal and functional phenomena that have been sporadically reported for mimetics in the literature. We keep the “model” as theory-neutral as possible for its future application to other phenomena and other languages.
The rest of the paper is organized as follows. Section 2 summarizes the basic iconic properties of Japanese mimetics that are relevant to the present discussion. Section 3 outlines the lexical representations of mimetics featuring stem-based morphology and dynamicity. Section 4 argues for the stem-based morphology of mimetics in terms of the nature of their derivation. Section 5 presents supportive evidence for the dynamic nature of mimetics with special attention to what they cannot mimic.1 Section 6 discusses the “fictive” nature of apparent counterexamples to the dynamicity hypothesis. Section 7 concludes the study.
2. Iconic links in Japanese mimetics
We start with a basic description of Japanese mimetics, focusing on their morphosyntactic and morphophonological-aspectual systematicity. First, the majority of mimetics can be used as manner adverbs, and at the most four syntactic categories are available to one mimetic stem, such as nurunuru in (1). (Hereafter, accentuation patterns are indicated only where relevant.)2
1 Kita (1997, 2008) also discusses what mimetics cannot do, pointing out the absence of hyponymy and antonymy within the mimetic lexicon (see also Bodomo 2006), as well as that of mimetics that specify the information about the causer, experiencer, and spatiotemporal settings. Furthermore, the near absence of transitive mimetic verbs is another issue yet to be explained (Kageyama 2007; Akita 2009).
2 The abbreviations and symbols used in this paper are as follows: ACC = accusative; Aff = affix; C = consonant; CONJ = conjunctive; COP = copula; DAT = dative; GEN = genitive; MIM = mimetic; NOM = nominative; NPST = nonpast; PST = past; Q = coda obstruent (only for mimetics); QUOT = quotative; TOP = topic; V = vowel; ' = accent nucleus.
(1) a. Unagi-ga núrunuru(-to) subet-ta. (adverbial) eel-NOM MIM-QUOT slip-PST
‘An eel slipped slimily.’
b. Kono unagi-wa núrunuru-su-ru. (verbal) this eel-TOP MIM-do-NPST
‘This eel feels slimy.’
c. Kono unagi-wa nurunuru-da. (nominal-adjectival) this eel-TOP MIM-COP
‘This eel is slimy.’
d. Tawasi-de nurunuru-o tot-ta. (nominal, rare) scrubbing.brush-by MIM-ACC remove-PST
‘[I] removed the sliminess by a scrubbing brush.’
As these examples show, each categorial realization of mimetics is marked morphologically and/or prosodically. Reduplicative mimetics are initially accented in their adverbial and verbal uses, but unaccented in their nominal(-adjectival) uses (Kindaichi 1978; Tamori 1980). Kageyama (2007: 30-31) interprets this as an instance of iconicity, in which an accent (pitch fall) is linked with the dynamicity of adverbs and verbs and its absence with the staticity of nominal adjectives and nouns. Moreover, as illustrated in (1a), adverbial mimetics are marked by the quotative particle -to, which is omissible under certain phonological conditions (Nasu 2002). Verbal mimetics consist of a mimetic and a dummy verb, mainly su- ‘do’ (Tsujimura 2005; Kageyama 2007), as in (1b). Nominal-adjectival mimetics are marked by a copula, as in (1c). These morphophonology-syntax correspondences suggest that mimetics themselves are
“precategorial” and underspecified with respect to their syntactic and semantic properties (cf. Tsujimura 2005).
Second, it has been widely recognized that, as is true in other languages (Hinton et al. 1994; Fischer 2011), the morphophonological shapes of mimetics are iconically associated with their aspectual properties (Izumi 1976; Akita 2009). As shown in (2), reduplicative mimetics (C1V1C2V2-C1V1C2V2) represent durative events, whereas
“suffixal” mimetics (C1V1C2V2-Aff) mostly represent punctual and sometimes telic events. Note that korokoro and koroQ/koron/korori share the rolling meaning as well as
the bimoraic mimetic root koro.
(2) a. Isi-ga korokoro-to korogat-ta. (reduplicative) stone-NOM MIM-QUOT roll-PST
‘A stone rolled continuously.’
b. Isi-ga {koroQ/koron/korori}-to korogat-ta. (suffixal)
stone-NOM MIM-QUOT roll-PST
‘A stone rolled once.’
Thus, iconic associations have been acknowledged for at least two aspects of the mimetic system of Japanese: the prosodic patterns associated with syntactic categories and the morphological shapes associated with aspectual features.3 These two sets of iconic correspondences together constitute a limited set of productive morphophonological templates, such as C1'V1C2V2-C1V1C2V2, which represents dynamic, durative events (i.e., activity or accomplishment) (Akita 2009).
3. Lexical representations of mimetics
In this section, we propose lexical representations of mimetics that feature stem-based morphology and dynamicity. As seen in the previous section, mimetic stems are categorially underspecified, and their categories are specified by the morphosyntactic context in which they appear. However, the detailed lexical content of mimetics themselves has rarely been investigated in the literature (for exceptional explorations in the aspectual semantics of mimetics, see Toratani 1999 and Akita 2009: Chapter 5; cf. Kageyama 2007 for the conceptual semantics of mimetic verbs).
(3a) is our lexical representation of the reduplicative mimetic nurunuru, which is formally and functionally further specified by the adverbial (-to-marked) and nominal(-adjectival) environments, as in (3b) and (3c), respectively. Iconic links are indicated by coindices.
3 The systematic phonosemantics (i.e., phonetic iconicity) of the segments and phonological features of mimetics have also been described (Hamano 1998). We do not discuss this here, assuming that it does not make a major grammatical contribution.
(3) a. nurunuru:
Phonology: initially accentedi
Morphology: C1V1C2V2-C1V1C2V2j
Semantics:
Dynamicity: dynamici Durativity: durativej
Referentiality: ‘slimy’ Category:
b. nurunuru-to ([MIM-QUOT]):
Phonology: initially accentedi
Morphology: C1V1C2V2-C1V1C2V2j
Semantics:
Dynamicity: dynamici Durativity: durativej
Referentiality: ‘slimily’
Category: Adv
c. nurunuru-su- ([MIM-do]):
Phonology: initially accentedi
Morphology: C1V1C2V2-C1V1C2V2j Semantics:
Dynamicity: dynamici
Durativity: durativej
Referentiality: ‘feel slimy’
Category: V
d. nurunuru-da ([MIM-COP]):
Phonology: unaccentedi
Morphology: C1V1C2V2-C1V1C2V2j Semantics:
Dynamicity: statici
Durativity: durativej
Referentiality: ‘be slimy’
Category: N(A)
Crucially, we assume that, although mimetic stems are categorially and semantically not fully determined, they are specified with respect to prosody and dynamicity, as in (3a). Put differently, (most) mimetics are assumed to be dynamic in the lexicon. This dynamicity is paired with the presence of an accent nucleus in the phonological representation. Therefore, on one hand, adverbial and verbal mimetics, which occupy a dominant portion of mimetic uses, are not different from their stem in terms of prosody and aspect ((3a) vs. (3b, c)). On the other hand, nominal(-adjectival) mimetics do have different prosodic and aspectual specifications from their stems, as in (3d). Thus, we
argue that the formal and functional specifications of the copulative construction override the prosodic and aspectual features of mimetic stems. Some supportive evidence for this argument will be presented in Section 5.
To recapitulate, the lexical representations of mimetics we posit are stem-based, rather than root-based (cf. Blevins 2006). In Section 4, we discuss why we do not take the root-based, derivational view, which has been adopted in some phonological and morphological studies of mimetics (Nasu 2002; Kadooka 2007).
4. Stem-based morphology of mimetics
The stem basis of mimetic morphology gains support from at least two facts, which previous studies have recognized but failed to discuss in an integrated fashion. First, as many studies (e.g., Hamano 1998) argue, mimetic roots cannot stand alone (i.e., without templates) in present-day Japanese (e.g., *nuru (cf. nurunuru ‘slimy’), *doki (cf. dokidoki ‘nervous’)).4 In this sense, mimetic roots are mere theoretical constructs that are induced from stems.
Second, stem-based morphology is consistent with the limited aspectuality of lexical items that are presumably derived from mimetics (Hamano 1998; Sugahara and Hamano, this volume). Mimetic roots of all these words are also found in reduplicative stems, suggesting that they are derived from reduplicative mimetic stems by truncation (cf. Tamori 1993), as shown in (4) taken from Hamano (1998: 55-61) (she does not discuss the derivational relationship between these verbs and reduplicatives). Note, in contrast, that some of these roots are not found in non-reduplicative stems (e.g.,
*iziQ/*izin/*iziri).
(4) a. Activity [durative, atelic, dynamic]: gira-tuk- ‘glare’ (< giragira ‘glaring’)
kira-mek- ‘glimmer’ (< kiarkira ‘glimmering’) izi-ke- ‘become timid’ (< iziizi ‘timid’)
4 A few old-fashioned mimetic stems are identical to their roots (e.g., biku ‘budging’, hata ‘noticing suddenly’, soyo ‘breezing’). Some demimeticized nouns also lack additional elements (e.g., boro ‘rag’ (< boroboro ‘ragged’), goro ‘grounder’ (< gorogoro ‘a heavy object rolling’)).
yur-e- ‘sway’ (< yurayura ‘swaying’) koro-gar- ‘roll’ (< korokoro ‘rolling’) yura-g- ‘sway’ (< yurayura ‘swaying’) hikar- ‘shine’ (< pikapika ‘flashing’)
b. Accomplishment [durative, telic, dynamic]: kuru-m- ‘roll up’ (< kurukuru ‘rolling’)
kuta-bar- ‘become exhausted’ (< kutakuta ‘exhausted’) c. State [durative, atelic, static]:
noro- ‘sluggish’ (< noronoro ‘sluggish’)
Importantly, all these predicates are durative (i.e., activity, accomplishment, or state) (see Van Valin 2005 for a recent explication of aspectual types of predicates). In order for these mimetic predicates to obtain durative semantics, they have to start at the stem level, where the reduplicative template is linked with durativity. This is why all these derivative predicates have their reduplicative counterparts.5
The present data reinforce the stem-based view of mimetic morphology. A root-based view would not provide a straightforward account for them. It would have to resort to mimetic roots, which are hypothetical entities, and end up concluding that the uniform aspectuality of derived mimetic predicates is a mere coincidence.
5. Dynamicity of mimetic stems: Their limits and potentials
In this section, we argue for the dynamic nature of mimetic stems on the basis of two sets of data. Section 5.1 cites a quantitative dataset that shows the restricted distribution of unaccented-static reduplicatives vis-à-vis accented-dynamic ones. Section 5.2 points out the utter absence of mimetics for genuinely static eventualities.
5.1. Accentuation of reduplicative mimetics
5 The present discussion begs the question of why only durative predicates are derived. In this respect, a close investigation into unattested punctual predicates, such as *suto-tuk- ‘fall flat’ (< suton ‘thumping’), which do not have their reduplicative counterparts (e.g., *sutosuto), would be needed.
As we summarized in Section 2, reduplicative mimetics exhibit systematic category-accentuation correspondences. Dynamic categories are linked with accented forms, whereas static categories are linked with unaccented forms. The two accentuation patterns are not available to each reduplicative mimetic. Some reduplicatives are always accented (e.g., (5)), others allow both accented and unaccented patterns (e.g., (6)), and a few are limited to the unaccented pattern (e.g., (7)).
(5) a. Kami-o tyókityoki-to kit-ta. (accented) paper-ACC MIM-QUOT cut-PST
‘[I] cut the paper snip-snip.’
b. *Kono kami-wa tyokityoki-da. (unaccented) this paper-TOP MIM-COP
‘This paper is snippy.’
(6) a. Beekon-o kárikari-to tabe-ta. (accented) bacon-ACC MIM-QUOT eat-PST
‘[I] ate the bacon with a crunching sound.’
b. Kono beekon-wa karikari-da. (unaccented) this bacon-TOP MIM-COP
‘This bacon is crunchy.’
(7) a. *Ai-wa sono aidoru-ni méromero-to akogare-ta. (accented) Ai-TOP that idol-DAT MIM-QUOT adore-PST
‘Ai adored the idol too much.’
b. Ai-wa sono aidoru-ni meromero-da. (unaccented) Ai-TOP that idol-DAT MIM-COP
‘Ai is too fond of the idol.’
Using a collection of 502 reduplicative mimetics mainly taken from Kakehi et al. (1996), Akita (2009: 165) quantitatively shows that while 87.69% (171) of the unaccented reduplicatives (195) have accented counterparts, only 35.77% (171) of the accented
reduplicatives (478) have unaccented counterparts (see also Hamano 1998: 12). This distribution suggests the secondary status of unaccented reduplicatives, which are derived from their accented counterparts.
Moreover, some unaccented-only reduplicatives, such as matimati ‘diverse’ and metyametya ‘incoherent’, have a non-mimetic flavor. These reduplicatives even behave as adjectives, exceptionally favoring the adjectival copula -na in a noun modification construction (e.g., {matimati/metyametya}-{na/?no} iken ‘{diverse/incoherent} opinions’) (see Uehara 1998). Note that nominal(-adjectival) mimetics are normally followed by the genitive -no in noun modification (e.g., {karikari/katikati}-{?na/no} beekon ‘{crunchy/stone-hard} bacon’) (see Section 5.2).
One possible interpretation of these facts is that most reduplicative mimetics are lexically registered as accented and dynamic, and many of them are “deaccented” and
“staticized” by a copulative construction, whereas a few exceptional unaccented-static reduplicatives are registered as such, as are non-mimetic adjectives (cf. Akita 2009). This interpretation is consistent with the proposed dynamicity of mimetics.
5.2. Absence of intrinsically static mimetics
The dynamicity of mimetics is further reinforced by the absence or rarity of intrinsically static items and constructions in mimetics. First, as already noted by prevous studies, such as Izumi (1976) and Kindaichi (1978), Japanese has no single mimetic for smell and taste. Apparent exceptions, such as punpun ‘reeking’, tun ‘stinging the nose’, mattari ‘smooth and rich’, and piripiri ‘pungent’, are thought to refer to cutaneous sensation (Yamanashi 1988/2007). Japanese also lacks mimetics for color and temperature. These semantic fields are covered by non-mimetic reduplicatives (e.g., aoao ‘very blue’ (< ao ‘blue’), akaaka ‘very red’ (< aka ‘red’), atuatu ‘piping hot’ (< atu- ‘hot’)) and synesthetic uses of mimetics (e.g., kinkin ‘extremely refrigerated (cold enough to make a tinkling sound if one hits on it)’ < ‘tinkling’) (Section 6). Assuming that all these are typical object properties, the absence of mimetics representing them suggests their inability to imitate pure state due to their dynamicity.6
6 Mimetics of these semantic types are sporadically reported in other languages (e.g., irundu ‘smelling good’ (Tamil; Abbi 1994: 16), llıı llıı ‘sweet’ (Didinga; de Jong 2001: 126), kpìnàkpìnà ‘dark/black’
Second, few adjectives are derived from mimetics. Japanese has two types of adjectives, -i-marked (-i = NPST) and -na-marked (-na = COP), neither of which is productive in mimetics. Exceptional instances of demimeticized -i-adjectives include boro(-i) ‘(be) ragged’ (< boroboro ‘ragged’) and noro(-i) ‘(be) sluggish’ (< noronoro
‘sluggish’).7 As we mentioned in Section 5.1, although -na-marking is not entirely impossible in informal speech, it typically gives way to genitive marking (e.g., karikari-{?na/no} beekon ‘crunchy bacon’). As adjectives describe properties, the low compatibility of mimetics with the adjectival category is clearly ascribed to their dynamicity orientation.
Third, in Japanese, NP-de can describe the property of either the subject referent or the object referent, called a “depictive” predicate (Koizumi 1994: 27). For instance, the depictive predicate hadaka-de ‘naked’ in (8a) depicts the property of Taro, who read a book, whereas nama-de ‘raw’ in (8b) depicts that of the bonito eaten by Taro.
(8) a. Taroo-ga hadaka-de hon-o yon-da. Taro-NOM naked-COP book-ACC read-PST
‘Taro read a book naked.’
b. Taroo-ga katuo-o nama-de tabe-ta. Taro-NOM bonito-ACC raw-COP eat-PST
‘Taro ate the bonito raw.’
(Koizumi 1994: 27)
In contrast, the depictive use of stative (nominal-adjectival) mimetics is limited. As Sells (2013) notes, they cannot depict object referents (cf. Shibagaki 2013: 167-168). Compare (9b) with the slightly colloquial but acceptable subject-oriented depictive in
(Siwu; Dingemanse 2011: 197)). However, this by no means directly rejects our dynamicity hypothesis. In fact, many instances that are described as color mimetics appear to be intensifiers (e.g., hak1 maa1 maa1 (black MIM MIM) ‘so dark’ (Cantonese; Bodomo 2006: 211)). Secondary expressions like the Japanese examples here may also be common in those languages. See Dingemanse (2012) for a crosslinguistic overview of the semantic coverage of mimetics.
7 An anonymous reviewer suggested adding the demimeticized adjective tyara(-i) ‘showy, flashy’ (< tyaratyara ‘showy, flashy’) to the exception list. We would classify this adjective as another type of coinage, which is peculiar to highly colloquial discourse and is not unique to mimetics (Akita, to appear). Other examples of this type of word formation include ero(-i) ‘erotic’ (< ero(tikku) ‘erotic’), mabu(-i)
‘attractive’ (< mabu ‘real’), mendo(-i) ‘tiresome’ (< mendokusa(-i) ‘tiresome’), and nau(-i) ‘trendy’ (< nau ‘now’).
(9a).
(9) a. Taroo-ga ukiuki-de hon-o yon-da. Taro-NOM MIM-COP book-ACC read-PST
lit. ‘Taro read a book buoyant.’
b. *Taroo-ga katuo-o pitipiti-de tabe-ta. Taro-NOM bonito-ACC MIM-COP eat-PST
‘Taro ate the bonito fresh.’
The contrast between (9a) and (9b) appears to come from the types of properties that are expressible by the two types of depictives. On one hand, subject-oriented depictives depict the agent’s properties, such as the reader’s psychological state in (9a). The agent’s properties are strongly associated with or responsible for the dynamic event denoted by the verb. On the other hand, object-oriented depictives can, in principle, only depict the purely static property of a non-sentient object (Miyata 2001: 151). Therefore, the low acceptability of the object-oriented depictive pitipiti-de ‘fresh’ in (9b) suggests that the non-static element involved in the encyclopedic semantics of the mimetic (e.g., the image of a jumping bonito) clashes with the semantic requirement of the construction.8, 9
The absence of intrinsically static mimetic items and constructions discussed in this section gives additional support to our argument that mimetic stems are dynamic in nature. The dynamicity of mimetics seems to stem from the semiotic fact that they imitate eventualities, either dynamic or static, by means of linguistic sound, which is dynamic in terms of both articulation and acoustics.
8 A scale-based account, which is not incompatible with the dynamicity account here, may be possible for the incompatibility of mimetics with object-oriented depictives. In general, object-oriented depictives specify the state of the object among a set of possibilities on the scale set by the verb and the object (Katsuya Fujikawa, personal communication; cf. Aarts 1992). For example, nama-de ‘raw’ in (8b) singles out the uncooked state for the bonito, excluding other types of bonito dishes, such as grilled bonito. Mimetics can be assumed underspecified with respect to the scale structure, reflecting their semantic complexity. The infelicity of object-oriented mimetic depictives appears to come from this underspecificity, which prevents a particular state from being picked up. A further pursuit of this supplementary account is far beyond the scope of this paper.
9 Some languages are reported to have a productive system of mimetic depictives (Ameka 2005; Amha and Dimmendaal 2005). Future research is hoped to take a closer look at the lexical content/status of mimetics in these languages.
6. Fictivity: Apparent counterexamples
In our dynamicity account of mimetic stems in the previous section, the static meaning of unaccented reduplicatives was attributed to the copulative construction or some individual reduplicatives. In this section, we discuss specifically how the static meaning of this type is obtained from the fundamentally dynamic semantics of mimetics, proposing two types of “fictivity”—metonymy and synesthetic metaphor—that reside in the extensive range of eventualities depicted by mimetics (see Talmy 1996 for various types of fictivity).
First, many static mimetics, including “deaccented” reduplicatives (e.g., karikari
‘crunchy’ in (6b)), involve a fictive type of metonymy. In this type of semantic extension, mimetics represent a particular state by referring to the sound that is associated with it. As Mikami (2006) argues, this association goes either toward the past, as in (10a), or toward the future, as in (10b).
(10) a. Kodomotati-wa minna ue-te garigari-ni yase, … children-TOP all starve-CONJ MIM-COP get.thin
‘All the children starved and got extremely thin…’ (vestigial)
(Mikami 2006: 209) b. … sentakumono-ga katikati-ni koot-te-simat-ta.
laundry-NOM MIM-COP freeze-CONJ-end.up-PST
‘… the laundry has frozen solid.’ (prospective)
(Mikami 2006: 211-212)
In (10a), the skinniness of the children is expressed as if it were a result of their bodies being scratched noisily. In (10b), the solid state of the laundry is expressed by the clicking sound that would be made if one hits on it. This productive type of fictivity allows the otherwise dynamic mimetics to cover static meanings.
Garigari ‘extremely thin’ and katikati ‘solid’ in (10) are analyzed as “extended” uses, which are distinct from their onomatopoeic uses (‘scratching noisily’ and
‘clinking’, respectively). However, fictive cognition goes beyond polysemous mimetics (Weilun Yu, personal communication). For example, as an anonymous reviewer pointed
out, the mimetic den ‘ostentatious’ can depict the ostentatious state of an object, as illustrated in (11a). This mimetic can also be used as a manner adverb modifying an event verb, as in (11b).
(11) a. Heya-no mannaka-ni tukue-ga den-to room-GEN center-DAT desk-NOM MIM-QUOT
oi-te ar-u. put-CONJ be-NPST
‘A desk is placed ostentatiously in the center of the room.’ b. Heya-no mannaka-ni tukue-o den-to oi-ta.
room-GEN center-DAT desk-ACCMIM-QUOT put-PST
‘[I] placed a desk ostentatiously in the center of the room.’
Unlike garigari and katikati above, this mimetic may not be analyzed as polysemous. (11a) is a vestigial sentence even without the mimetic, as it is not necessary that the speaker saw someone place the desk in the center of the room. Therefore, one can analyze the mimetic itself as specifying the (dynamic) manner of placing in both (11a) and (11b).
Second, even static mimetics without their dynamic counterparts are assumed to possess a fictive dynamic basis as iconic expressions that “mimic” eventualities by linguistic sound. For example, the unaccented-only reduplicative meromero ‘too fond’ in (7) above evokes the metaphorical image in which one is enchanted as if melting. The evocation of this fictive melting event is facilitated by the synesthetic (or cross-modal) sound symbolism of /mero/ (Hinton et al. 1994; Ahlner and Zlatev 2010) as well as the reduplicative-durative template.
Likewise, as shown by the unacceptability of (12a), the mimetic karaQ ‘nice and crisp’ is not an explicit sound expression. However, this state mimetic represents the crispness by (synesthetically) referring to the endpoint of the deep-frying event on the basis of the suffixal template (C1V1C2V2-Aff), which is linked with the punctual/telic aspect (Section 2). In this regard, (12b) can be analyzed as involving a “vestigial” type of fictive cognition in which the crisp state is conceptualized as a product of the completion of cooking that is accompanied by a hypothetical crisping sound (cf. Usuki and Akita 2013).
(12) a. *Tenpura-o karaQ-to kazit-ta. tempura-ACCMIM-QUOT bite-PST
‘[I] bit the tempura with a crisping sound.’ b. Tenpura-o karaQ-to age-ta.
tempura-ACC MIM-QUOT deep.fry-PST
‘[I] deep-fried the tempura nice and crisp.’
Thus, our dynamicity account even applies to seemingly static mimetic instances with the help of the notion of fictivity. Underlying this generalization is the very nature of mimetics as iconic expressions that depict eventualities by means of dynamic sound.10
7. Conclusion
In this paper, we have argued that, despite their “precategorial” status, mimetics are lexically specified as dynamic, and the analytical unit for them should be the aspectually burdened stem, rather than the theoretically induced root. Future investigations are hoped to place the current discussion in specific theoretical contexts, such as Distributed Morphology (Marantz 2007, cf. Tada 2010) and Construction Grammar (Fillmore and Kay 1995; Boas and Sag 2012). Such investigations will benefit from a crosslinguistic comparison of the lexical content of mimetics and its relationship to particular morphosyntactic constructions.
References
Aarts, B. 1992. Small Clauses in English: The Nonverbal Types [Topics in English Linguistics 8]. Berlin and New York: Mouton de Gruyter.
10 The interpretive (non-firsthand) nature of the “iconicity” in these cases, both sound symbolism and template-aspect correspondence, suggests the relevance of the primary vs. secondary/coerced distinction of iconicity (Ahlner and Zlatev 2010; Dingemanse 2011).
Abbi, A. 1994. Semantic Universals in Indian Languages. Shimla: Indian Institute of Advanced Study.
Ahlner, F. and Jordan, Z. 2010. “Cross-modal iconicity: A cognitive semiotic approach to sound symbolism”. Sign Systems Studies 38 (1/4): 298-348.
Akita, K. 2009. A Grammar of Sound-Symbolic Words in Japanese: Theoretical Approaches to Iconic and Lexical Properties of Mimetics. Ph.D. dissertation, Kobe University.
Akita, K. To appear. “Register-specific morphophonological constructions in Japanese”. In Proceedings of the Thirty-Eighth Annual Meeting of the Berkeley Linguistics Society. Berkeley, CA: Berkeley Linguistics Society.
Ameka, F.K. 2005. “Forms of secondary predication in serializing languages: On depictives in Ewe”. In Secondary Predication and Adverbial Modification: The Typology of Depictives, N.P. Himmelmann and E.F. Schultze-Berndt (eds), 335-378. Oxford: Oxford University Press.
Amha, A. and Dimmendaal, G.J. 2005. “Secondary predicates and adverbials in Nilotic and Omotic: A typological comparison”. In Secondary Predication and Adverbial Modification: The Typology of Depictives, N.P. Himmelmann and E.F. Schultze-Berndt (eds), 299-321. Oxford: Oxford University Press.
Blevins, J.P. 2006. “Word-based morphology”. Journal of Linguistics 42 (3): 531-573. Boas, H.C. and Sag, I.A. 2012. Sign-Based Construction Grammar [CSLI Lecture
Notes]. Stanford, CA: CSLI Publications.
Bodomo, A.B. 2006. “The structure of ideophones in African and Asian languages: The case of Dagaare and Cantonese”. In Selected Proceedings of the 35th Annual Conference on African Linguistics: African Languages and Linguistics in Broad Perspectives, J. Mugane, J.P. Hutchison and D.A. Worman (eds), 203-213. Somerville, MA: Cascadilla Proceedings Project.
de Jong, N. 2001. “The ideophone in Didinga”. In Ideophones [Typological Studies in Language], F.K.E. Voeltz and C. Kilian-Hatz (eds), 121-138. Amsterdam and Philadelphia: John Benjamins.
Dingemanse, M. 2011. The Meaning and Use of Ideophones in Siwu. Ph.D. dissertation, Max Planck Institute for Psycholinguistics, Nijmegen/Radboud University.
Dingemanse, M. 2012. “Advances in the cross-linguistic study of ideophones”. Language and Linguistics Compass 6/10: 654-672.
Fillmore, C.J. and Kay, P. 1995. Construction Grammar Coursebook. Ms., University of California, Berkeley.
Fischer, O. 2011. “Cognitive iconic grounding of reduplication in language”. In Semblance and Signification [Iconicity in Language and Literature 10], P. Michelucci, O. Fischer and C. Ljungberg (eds), 55-81. Amsterdam and Philadelphia: John Benjamins.
Hamano, S. 1998. The Sound-Symbolic System of Japanese [Studies in Japanese Linguistics]. Stanford, CA: CSLI Publications.
Hinton, L., Nichols, J. and Ohala, J.J. (eds). 1994. Sound Symbolism. Cambridge: Cambridge University Press.
Ibarretxe-Antuñano, I. 2006. Sound Symbolism and Motion in Basque [LINCOM Studies in Basque Linguistics]. München: LINCOM Europa.
Izumi, K. 1976. “Giseigo/gitaigo-no tokusitu” [Some characteristics of mimetics]. In Nihongo Koza (4) Nihongo-no goi-to hyoogen [Lecture on Japanese (4): Japanese vocabulary and expressions] [Nihongo Nihonjin sirizu [Japanese language and people series], T. Suzuki (ed.), 105-151. Tokyo: Taishukan Shoten.
Kadooka, K. 2007. Nihongo-onomatope-goi-ni okeru keitai-teki/on’in-teki taikeisei-ni tuite [On the morphological and phonological systematicity of the onomatopoeic vocabulary of Japanese]. Tokyo: Kurosio Publishers.
Kageyama, T. 2007. “Explorations in the conceptual semantics of mimetic verbs”. In Current Issues in the History and Structure of Japanese, B. Frellesvig, M. Shibatani and J.C. Smith (eds), 27-82. Tokyo: Kurosio Publishers.
Kakehi, H., Tamori, I. and Schourup, L. 1996. Dictionary of Iconic Expressions in Japanese [Trends in Linguistics Documentation]. Berlin and New York: Mouton de Gruyter.
Kindaichi, H. 1978. “Giongo/gitaigo gaisetu” [An overview of mimetics]. In Giongo/gitaigo-ziten [A dictionary of mimetics] [Kadokawa Syoziten 12 [Kadokawa concise dictionary 12]], T. Asano (ed.), 3-25. Tokyo: Kadokawa Shuppan.
Kita, S. 1997. “Two-dimensional semantic analysis of Japanese mimetics”. Linguistics 35 (2): 379-415.
Kita, S. 2008. “World-view of protolanguage speakers as inferred from semantics of sound symbolic words: A case of Japanese mimetics”. In The Origins of Language:
Unraveling Evolutionary Forces, N. Masataka (ed.), 25-38. Tokyo: Springer. Koizumi, M. 1994. “Secondary predicates”. Journal of East Asian Linguistics 3 (1):
25-79.
Marantz, A. 2007. “Phases and words”. In Phases in the Theory of Grammar, S.H. Choe (ed.), 191-222. Soeul: Dong-In Publishing.
https://files.nyu.edu/ma988/public/Phase_in_Words_Final.pdf
Mikami, K. 2006. “Nihongo-no giongo/gitaigo-ni okeru imi-no kakutyoo: konseki-teki-ninti/yoki-teki-ninti-no kanten-kara [The semantic extension of Japanese mimetics: From the perspective of vestigial and prospective cognition]”. Nihongo Nihonbungaku kenkyu [Studies on Japanese Language and Literature] 57: 199-217.
Miyata, A. 2001. “Byoosya-koobun-no seiritu-zyooken: Ninti-ron-teki-siten-kara” [Constraints on the depictive construction: From a cognitive point of view]. Proceedings of the First Annual Meeting of the Japanese Cognitive Linguistics Association, 143-153.
Nasu, A. 2002. Nihongo-onomatope-no gokeisei-to inritu-koozoo [The word formation and prosodic structure of Japanese onomatopoeia]. Ph.D. dissertation, University of Tsukuba.
Sells, P. 2013. “Overview: The grammar of mimetics”. Introductory remarks of the Grammar of Mimetics workshop, SOAS, University of London, 10 May 2013. Shibagaki, R. 2013. Analysing Secondary Predication in East Asian Languages.
Cambridge: Cambridge Scholar Publishing.
Tada, H. 2010. “Gitaigo-doosi-ni okeru <tikaku/zokusei>-kootai-no tokuisei-ni tuite” [On the peculiarity of the <perceptual/attributive> alternation in mimetic verbs]. Paper presented at the Eighth Fukuoka University Linguistics Colloquium, Fukuoka University.
Talmy, L. 1996. “Fictive motion in language and ‘ception’”. In Language and Space [Language, Speech, and Communication], P. Bloom, M.A. Peterson, L. Nadel and M.F. Garrett (eds), 211-276. Cambridge, MA: MIT Press.
Tamori, I. 1980. “Cooccurrence restrictions on onomatopoeic adverbs and particles”. Papers in Japanese Linguistics 7: 151-171.
Tamori, I. 1993. “Nihongo-onomatope-no toogo-hantyuu” [Syntactic categories of Japanese onomatopoeia]. In Onomatopia: gion-/gitaigo-no rakuen [Onomatopia: A
utopia of mimetics], H. Kakehi and I. Tamori (eds), 17-75. Tokyo: Keiso Shobo. Tamori, I. and Schourup L. 1999. Onomatope: Keitai-to imi [Onomatopoeia: Form and
meaning] [Nitieitaisyokenkyu sirizu [A Series of Comparative studies of Japanese and English]] . Tokyo: Kurosio Publishers.
Toratani, K. 1999. “Aspectual matching and mimetics in Japanese”. Proceedings of the Twenty-Eighth Western Conference on Linguistics 11: 495-507.
Tsujimura, N. 2005. “A constructional approach to mimetic verbs”. In Grammatical Constructions: Back to the Roots [Constructional Approaches to Language], M. Fried and H.C. Boas (eds), 137-154. Amsterdam and Philadelphia: John Benjamins. Uehara, S. 1998. Syntactic Categories in Japanese: A Cognitive and Typological
Introduction [Studies in Japanese Linguistics]. Tokyo: Kurosio Publishers.
Usuki, T. and Akita, K. 2013. “Fiction in an encyclopedia: A Generative Lexicon approach to fictive mimetic resultatives in Japanese”. In Gengogaku-kara-no tyooboo 2013: Hukuoka-gengo-gakkai 40-syuunen-kinen-ronbunsyuu [The view from linguistics: Papers from the 40th anniversary of Fukuoka Linguistic Circle], Fukuoka Linguistic Circle (ed.), 308-321. Fukuoka : Kyushu University Press. Van Valin, R.D., Jr. 2005. Exploring the Syntax-Semantics Interface. Cambridge:
Cambridge University Press.
Voeltz, F.K.E. and Kilian-Hatz, C. (eds). 2001. Ideophones [Typological Studies in Language 44]. Amsterdam and Philadelphia: John Benjamins.
Yamanashi, M. 1988/2007. Hiyu-to rikai [Figure of speech and understanding] [Korekusyon Nintikagaku [A Collection of Cognitive Science]]. Tokyo: Tokyo University Press.
