Lexical variation and semantic change in literary Chinese ideophones: A cognitive and computational linguistic approach

# Lexical variation and semantic change in literary Chinese ideophones: <br> A cognitive and computational linguistic approach
## 中文文言象聲詞的詞彙變異與語意變遷— <br> 以認知與計算語言學方法探討
<html>
<div style="float:left">

</div>
<hr color='#875712' size=1px width=796px>
</html>
### Ph.D candidate Thomas Van Hoey 司馬智 <br> Advisor: Lu Chiarung 呂佳蓉
### <br>Proposal<br>28 January 2019<br>National Taiwan University

---

# Roadmap

.font170[
* Introduction

* Background (literature review)
* Research questions
* Methodology
* Case study: .sc[visual] ideophones: .sc[light] ideophones
* Expected findings for the full dissertation
* Selected references
]
    
---

# Introduction: .sc[iconicity]

Iconicity has been studied in different perspectives since the 1980s:

* iconic ordering of clausal constituents in relation to their temporal order in syntax and pragmatics .ref[(e.g. Haiman 1985; Simone 1995; Radden & Panther 2004)];
* iconic relations between sounds and meaning, viz. phonaesthemes and sound symbolism in phonetics, phonology and morphology .ref[(Hinton, Nichols & Ohala 1994; Perniss & Vigliocco 2014; Lockwood 2017)]; 
* iconicity across different modalities such as in sign language .ref[(Taub 2001)] or metaphor .ref[(Hiraga 2005)]; 
* and in the .sc[lexicon] (.markeer[onomatopoeia] and .markeer[ideophones] or .markeer[mimetics], .ref[cf. Voeltz & Kilian-Hatz 2001; Dingemanse 2012; 2018; Haiman 2018)].
 
.ru-blockquote[This dissertation is about onomatopoeia and ideophones / mimetics.]

---

# Ideophones: examples

Language | Examples
--- | ---
Korean <br> .ref[(Lee 1992)] | *potïl* ‘soft and tender (surface)’, *palt’ak* ‘palpitating, jerking’, *c’onc’on* ‘woven tightly’, *ulakpulak* ‘unbalanced scary appearance’, *colcol* ‘flowing liquid’, *kalphancilphaN* ‘unable to decide’
Pastaza Quechua <br> .ref[(Nuckolls 1996)] | *dzing* ‘a sudden awareness or intuition, especially one that causes fright’, *sa* ‘expanded or random movement from or within a locus’, *tsung* ‘to absorb, cover, or drench with a liquid substance’, *palay* ‘to fall rapidly and/or peltingly, as a collectivity of entities’
Siwu <br> .ref[(Dingemanse 2011)] | *gùdùù* ‘pitch dark’, *gblogblogblo* ‘bubbling’, *fũɛ̃fũɛ̃* ‘malleable' *kpɔtɔtɔrɔ-kpɔtɔtɔrɔ* ‘walking like a  tortoise’, *ɣììì*  ‘sensation of vertigo’, *wɔ̃rã~wɔ̃rã* ‘spotted, patchy pattern’, *nyɛ̃kɛ̃nyɛ̃kɛ̃* ‘intensely sweet’, *kpiɛkpiɛ* ‘lukewarm’

Examples presented in Dingemanse's .ref[(2012:661)] overview article

---

# Ideophones: semantic domains

Domain | Example
--- | ---
.sc[sound] | *miāo* 喵 ‘mew, miaow’
.sc[movement] | *yōuyōu* 攸攸 ‘move swiftly, quickly; smoothly slipping by, rapidly rushing’
.sc[visual patterns] | *zhēngróng* 崢嶸 ‘craggy, precipitous masses of rock; loftily lifted; sheer steepness; high-piled’
.sc[other sensory perceptions] | *bìbì* 苾苾 ‘deeply fragrant’
.sc[inner feelings and cognitive states] | *tántè* 忐忑 'be perturbed, mentally disturbed'

Hierarchy presented in .ref[Dingemanse (2012:663)].

As it can be seen, .markeer[Chinese has examples for all these different domains].
Yet, it is usually left out from the cross-linguistic debate.

---

# Ideophones: terminology

Japanese-inspired | Cross-linguistic | Reduplication | Examples
--- | --- | ---
擬聲詞 <br> *giongo* 擬音語 <br> .markeer[phonomimes] | .markeer2[ideophones] <br>  onomatopoeia | non-red. or A| 哔 'beep', <br> 咻 'shew' (.sc[sound + visual])
擬態詞 <br> *gitaigo* 擬態語 <br> .markeer[phenomimes] | .markeer2[ideophones] <br> non-phonomimes  | full red. or XAA (ABB) <br> partial red. or AB |  雾-濛濛 'fog-misty' <br> 崢嶸 'precipitous'
擬情詞 <br> *gijōgo* 擬情語 <br> .markeer[psychomimes] | .markeer2[ideophones] <br> non-phonomimes   | full red. or AA | 欣欣 'joyful'

These terms each highlight a different aspect of the data (cf. section 2 of the proposal).

It is my goal to unify these aspects into a single, though multi-faceted theory.

---

# Defining ideophones

Such a theory is already in the making: Dingemanse's .ref[(2011; 2012)] definition:

* **formal component**: marked words
* **meaning component**: depiction of sensory imagery

Most studies up to this point have focused on the .sc[form].
When .sc[meaning] is taken into account, the scope usually only contains .sc[sound].

I want to focus mostly on the .sc[meaning] but not without losing .sc[form] out of sight.

Moreover: variation is what interests me.

---

# Research questions

## .markeer[Main RQ: What does the semantic structure of Chinese ideophones look like?]

### .markeer2[RQ1. Is there any variation at play, and if so, in what ways?]

### .markeer2[RQ2. How have these structures evolved over time?]

### .markeer2[RQ3. What constructions do ideophones appear in and what are their functions?]

---

# Main research question

## .markeer[What does the semantic structure of Chinese ideophones look like?]

I will address this main question by:

1. answering the .markeer[subquestions] .ref[(see below)]

1. take as a starting point Dingemanse's .ref[(2011; 2012)] definition of the cross-linguistic concept of ideophones:
> "Ideophones are .markeer[marked words] that .markeer[depict sensory imagery]."

1. .markeer[database] of Chinese ideophones, available at [https://simazhi.shinyapps.io/Chineseideophone/](https://simazhi.shinyapps.io/Chineseideophone/)   
.ref[(consultable, but still not fully functional)]

---

# RQ 1: Variation

## .markeer2[Is there any variation at play, and if so, in what ways?]

Usually, .markeer[constructions] are defined as .markeer[mappings between .sc[form] and .sc[meaning]] .ref[(Langacker 1987; 1991; 2008)] .markeer[or .sc[form] and .sc[function]] .ref[(Fillmore 1988; Goldberg 1995; 2006; Croft 2001)]

`$$form \mid meaning$$`

Chinese has a .markeer[folk model of *xíng yīn yì* 形音義],  
culminating in 'the character' *zì* 字, .ref[(Packard 1998; 2000)]:  
.markeer[
* .sc[written form], 
* .sc[phonological form],
* .sc[meaning] ]

`$$\frac{sound}{writing} \mid meaning$$`

---

# RQ 1: Variation

## .markeer2[Is there any variation at play, and if so, in what ways?]

`$$\frac{sound}{writing} \mid meaning$$`

.font90[
Pole / Modalities | Example
--- | ---
.sc[sound] vs. .sc[meaning] <br> .font80[.ref[(polysemy & homonymy)]] | /jījīzhāzhā/ referring to the sounds of 'mice' or 'people chattering'
.sc[writing] vs. .sc[meaning] | <嘰嘰喳喳\> referring to the sounds of 'mice' or 'people chattering'
.sc[sound] vs. .sc[writing] | /wēiyí/ being written as <逶迤\> or <委蛇\> <br> .font80[.ref[(there are more variations, cf. Hsu 2013)]]
.sc[writing] vs. .sc[sound]  | <丁丁\> as /dīngdīng/ or /zhēngzhēng/
.sc[meaning] vs. .sc[sound] <br> .font80[.ref[(onomasiology)]] | .sc[brightness] expressed with different ideophones (see case study)
.sc[writing] vs. .sc[sound] <br> .font80[.ref[(onomasiology)]] |.sc[brightness] expressed with different ideophones (see case study)

]
---

# RQ 1: Variation

## .markeer2[Is there any variation at play, and if so, in what ways?]

Ideophones of different modalities, e.g. .sc[sound], .sc[visual], .sc[other perceptions], .sc[inner feelings] and .sc[cognitive states].

Case study of .sc[visual] ideophones, more precisely ideophones in the domain of .sc[light].

In the future, another major case study with .sc[auditory] ideophones.

Cursory findings:

* prototypical structure of meanings
* token frequency effects
* type frequency effects

---

# RQ 2: Diachronic evolution

## .markeer2[How have these structures evolved over time?]

Using the same case study of .sc[light] ideophones:

* the semantic structure is dynamic and evolves over time
* new meanings are formed from certain meanings or clusters of meanings in similar ways for the data under inspection
* these meaning extensions are motivated (yet not predictable)

---

# RQ 3: Qualitative approaches

## .markeer2[What constructions do ideophones appear in and what are their functions?]

Some possible future lines of development:

* investigating particular constructions, <br> e.g. .sc[v-qilai + ideophone]) with particular theories (e.g. Cognitive Grammar .ref[(Langacker 1987; 1991; 2008)], collostructions .ref[(Stefanowitsch & Gries 2003)]
* multimodality: ideophone and image, <br> cf. comic book studies .ref[(Forceville & Urios-Aparisi 2009)] or clothing terms .ref[(Geeraerts, Grondelaers & Bakema 1994)]
* acquisition: the role of intersubjectivity .ref[(Verhagen 2005)] in this process

---

# Structure of the dissertation

A .markeer[holistic] study with attention to balances between:

* **synchronic** and **diachronic**
* **qualitative** and **quantitative**
* **paradigmatic** and **syntagmatic**
* **form**, **meaning**, and **usage**

---

# Methodology

The .markeer2[methodological frameworks] in this dissertation draw heavily on approaches from .markeer[Cognitive semantics], e.g.:

* Diachronic prototype semantics .ref[(Geeraerts 1997)]
* Conceptual Metaphor Theory .ref[(Lakoff & Johnson 1980; Kövecses 2017)]
* Mental Spaces .ref[(Fauconnier & Sweetser 1996; Fauconnier & Turner 2003)]
* Frequency effects .ref[(Bybee & Hopper 2001)]

Combined with .markeer2[computational approaches], such as .markeer[distributional semantics]:

* Cognitive Linguistics .ref[(Heylen, Speelman & Geeraerts 2012; Wielfaert, Heylen & Speelman 2013; Heylen et al. 2015; Peirsman, Geeraerts & Speelman 2015)]
* `word2vec`-like approaches .ref[(Mikolov, Yih & Zweig 2013; Mikolov et al. 2013; Goldberg & Levy 2014)]
* `R` .ref[(Silge 2017a; 2017b)] and `python`

Using .markeer[historical corpus material]:

* Scripta Sinica corpus 漢籍全文資料庫計畫

---

# Case study: .sc[light] ideophones

From these ideophones that are expressed by the sensory modality of .sc[vision] <br> be it dynamic (.sc[movement]) or static (.sc[vision]),
<br> I have chosen those that express .sc[light].

Similar to the English *gl-* phonaestheme:

* *glare, gleam, glim, glimmer, glint, glisten, glister, glitter, gloaming, glow*
* *glance, glaze, glimpse, glint*
* *glacé, glacier, glair, glare, glass, glaze, gloss*

---

# .sc[light] ideos: types

.font80[
<div id="htmlwidget-f4805d73f11fcfc68a20" style="width:100%;height:auto;" class="datatables html-widget"></div>
<script type="application/json" data-for="htmlwidget-f4805d73f11fcfc68a20">{"x":{"filter":"none","fillContainer":false,"data":[["1","2","3","4","5","6","7","8","9","10","11","12","13","14","15","16","17"],["熠熠","煜煜","燿燿","耀耀","爚爚","灼灼","爍爍","鑠鑠","犖犖","燁燁","爗爗","曄曄","煒煒","韡韡","煇煇","輝輝","暉暉"],[" yìyì "," yùyù "," yàoyào "," yàoyào "," yuèyuè "," zhuózhuó "," shuòshuò "," shuòshuò "," luòluò "," yèyè "," yèyè "," yèyè "," wěiwěi "," wěiwěi "," huīhuī "," huīhuī "," huīhuī "],[" 閃亮光耀的樣子。 "," 光明照耀的樣子。 "," 光明的樣子。 "," 光明的樣子。 "," 光明的樣子。 "," 花茂盛鮮明。明亮。 "," 光閃動的樣子。 "," 光明閃耀的樣子。 "," 事理分明的樣子。光明磊落的樣子。  "," 光鮮明亮的樣子。  顯赫的樣子。  "," NA "," 盛大的樣子。 "," 光彩極盛的樣子。 "," 光明盛大的樣子。 "," NA "," NA "," 晴朗的樣子。 "],[" vividly bright"," burning brightly, flamboyant"," flashing, sparkling, glittering"," flashing, sparkling, glittering"," flashing and flickering, blazingly bright"," evident, brilliant, aglow, vivid and vibrant,   brightly blazing, plain and patent"," flashing, flaring, effulgent, alight, rutilant; splendrous"," polished, gleaming; shining; glittering, flashing"," manifestly evident; conspicuous, outstanding"," brightly shining; flashing, flaring, gleaming"," NA"," brightly shining; flashing, flaring, gleaming"," swirling or globed flames; bright shimmer"," vividly dazzling, gorgeously glistering"," fire-red, blazing brightly; splendid; brilliant"," radiant illumination, glow; splendor, brilliance"," radiant, gleaming; vividly white, candent;  gloze, spread light; dazzle"]],"container":"<table class=\"display\">\n  <thead>\n    <tr>\n      <th> <\/th>\n      <th>character<\/th>\n      <th>pinyin<\/th>\n      <th>MOE<\/th>\n      <th>Kroll<\/th>\n    <\/tr>\n  <\/thead>\n<\/table>","options":{"pageLength":9,"order":[],"autoWidth":false,"orderClasses":false,"columnDefs":[{"orderable":false,"targets":0}],"lengthMenu":[9,10,25,50,100]}},"evals":[],"jsHooks":[]}</script>
]

---

background-image: url(vergrijpen.png)
background-size:contain
background-position: right

# Methodology case study: *vergrijpen*

.font70[
<br>
.markeer[A: to use physical violence against (someone)]  
.markeer[B: to oppose someone to whom one owes respect and obedience]  
.markeer2[C: to harm (someone) in a non-physical way]  
D: to oppose an abstract principle  
E: to mis-take  
F: to do something forbidden  
G: to make a mistake  
H: to adulterate  
I: to do something inadvisable, unwise, improper  
.markeer2[J: to harm (something) in a non-physicial way]  
K: to steal  
L: to violate a woman’s honour  
M: to eat or drink excessively  
N: to hurt while catching or seizing  
O: to rebel violently  
P: to catch the wrong person  
Q: to commit suicide  
R: to damage (something)
]

---

# *Huīhuī* 煇煇, *huīhuī* 輝輝,  *huīhuī* 暉暉

Let us take these three as a case study for this presentation:

Mandarin |Middle Chinese|Old Chinese|MOE|Shuowen jiezi
--- |--- |--- |--- 
暉 huī |<MC xjwɨj |<OC *qʷʰər|「 晴朗的樣子。」|「 光也。」
輝 huī |<MC xjwɨj |<OC *qʷʰər||「 光也。」
煇 huī |<MC xjwɨj |<OC *qʷʰər||「 光也。」

Based on the definitions, one could presume that their meanings are entirely the same, i.e. synonymous.

But below I will show that this is not the case.

---

# *Huīhuī* 煇煇, *huīhuī* 輝輝,  *huīhuī* 暉暉

Step 0. .markeer[Getting the data from corpus (manual copying) + 'wrangling' into concordance]

Step 1. .markeer[Identifying the collocate]

*huīhuī* 輝輝 with .markeer2[LIGHT] radical

* "Spring clouds gently drifting, .under[the sun] *blazing*" 春雲澹澹日輝輝, in collection of 御定佩文齋廣群芳譜
* "On a clear morning, .under[the lantern] *burning brightly*" 清晨輝輝燭, in collection of 御定佩文齋廣群芳譜
* "*Bright* .under[cinnabar]" 輝輝丹, in collection of 御定佩文齋廣群芳譜
* "*Bright* is the light, shining in the 5 .under[colours]" 輝輝有光曜五色, in 全後漢文

---

# *Huīhuī* 煇煇, *huīhuī* 輝輝,  *huīhuī* 暉暉

Step 1. .markeer[Identifying the collocate]

*huīhuī* 煇煇 with .markeer2[FIRE] radical

* "[...] make ill .under[the ear], and the sound *blazing*" 病耳聾煇煇, in 耳病門
* ".under[candles] *burning brightly*"燭煇煇, in 咸淳臨安志
* "The colour .under[red] is *flaming brightly*" 赤之色煇煇也, in 易緯稽覽圖

*huīhuī* 暉暉 with .markeer2[SUN] radical

* ".under[vermillion red] and *bright* the drunken face" 暉暉朱顏酡, in 全宋詩
* "*Brightly red*, watching the .under[setting sun]" 暉暉視落日, in 梁詩
* ".under[The sun] *brightly red*, under the mountain" 日暉暉兮山之下, in 全宋詩

Step 2. .markeer[Count these collocates per period]

---

# *Huīhuī* 煇煇, *huīhuī* 輝輝,  *huīhuī* 暉暉

---

# *Huīhuī* 煇煇, *huīhuī* 輝輝,  *huīhuī* 暉暉

There is a big .markeer[type frequency] difference: *huīhuī* 輝輝 with .sc[LIGHT] radical has a much higher .markeer[type frequency] in terms of different referential collocates.

Still, they each are similar in meanings, e.g. different shades of .sc[red] per ideophone.

This difference in .markeer[type frequency] inspired an inquiry into .markeer[token frequency].

The ideophone with the highest token frequency in my data was *zhuózhuó* 灼灼.

---

# *Zhuózhuó* 灼灼

---

# Computational complement

Until now, these manual methods have yielded a great amount of insight into the dynamic semantic structure of these ideophones.

However, I want to subject even more ideophones to such an analysis.

Therefore, I wanted to use computational methods to help automate the process.

.markeer[Operationalized question:] .markeer2[how can computational methods help us better understand the semantic structure of ideophones?]

As mentioned, answers were found in .markeer[distributional semantics] methodology.  
.ref[
* Cognitive linguistics: Heylen, Speelman & Geeraerts 2012; Wielfaert, Heylen & Speelman 2013; Heylen et al. 2015; Peirsman, Geeraerts & Speelman 2015
* word2vec-like approaches: Mikolov, Yih & Zweig 2013; Mikolov et al. 2013; Goldberg & Levy 2014
]

---

# Computational complement

Step 0. .markeer[Gathering data]

The material stayed the same — Scripta Sinica corpus 漢籍全文資料庫計畫 — but I wrote a `python` script (with the `selenium` library) to gather the data, instead of having to manually copy it.

Step 1. .markeer[Segmenting text]

Using the `python` library `jieba`, I segmented the text, which was relatively successful.

In order to aid in the segmentation process, two 'dictionaries' were added to `jieba`:

1. a list of ideophones (database)
1. a list of particles used in Premodern Chinese .ref[(based on Jonker, Esch & Mansvelt Beck 2011)]

Step 2. .markeer[Choice of model and units]

* word-based (as opposed to syntax-based or text-based)
* skip-grams (cf. [this blog post](https://juliasilge.com/blog/tidy-word-vectors/) and [this one](https://juliasilge.com/blog/word-vectors-take-two/))]
* functions written with `R` (a function called `sliding_windows()`)

---

# Computational complement

Step 3. .markeer[Frequencies and co-occurrence strength]

* calculating the pointwise mutual information PMI for every pair of words, using a function `tidy_pmi()`

Step 4. .markeer[Similarity]

* reducing the dimensionality with singular value decomposition SVD, with a function `widely_svd()` from the `widyr` package

Step 5. .markeer[Interpreting the results]

Now there is a long data frame ('table-like object in `R`'), with a score for each 'neighbour' ('calculated collocate')

---

# The computational *zhuózhuó* 灼灼

![](zhuozhuocomp.png)

---

# The computational *zhuózhuó* 灼灼

But: they mostly show the higher-frequency items  
— if there are 'hapaxes' or many similar scores, it can become messy.

However, they can aid in the analysis of higher levels of abstraction:

* mental spaces .ref[(Fauconnier1994; Fauconnier & Turner 2003; etc.)]
* frames .ref[(Fillmore 1975; 1988; 2003; etc.)]
* domains .ref[(Langacker 1987; 1991)] / ICMs .ref[(Lakoff 1987)]
* image schemas .ref[(Johnson 1987; Lakoff 1987)]

I categorized the most frequent collocates of the mental spaces into frames and domains, following the method proposed by Kövecses .ref[(2017)].

---

# Manual frames

---

# Computationally aided frames

---

# The frames and domains of *huīhuī*

The preceding diagram leads to two further questions, in relation to the case study of *huīhuī*:

1. .markeer[If we collapse the variable .sc[time], which frames co-occur with the three *huīhuī*s?]
1. .markeer[Given that there are three different radicals (部首/偏旁/functional components), how much do certain radicals *attract* certain frames?]

---

# Frames, domains, and *huīhuī*

---

# Interim discussion

## RQ1: Is there any variation at play, and if so, in what ways?

## RQ2: How have these structures evolved over time?

Through the case studies presented today, and those in my proposal, I have shown that:

1. .markeer[variation] is present in .sc[light] ideophones (as it is probably inherent in language throughout time) 
1. the meanings of these ideophones are .markeer[dynamic]
1. the .markeer[written form] has variation which can .markeer[influence the meaning]

## .ref[RQ3: What constructions do ideophones appear in and what are their functions?]

---

# Plan for the dissertation

## 1. More case studies
* .sc[auditory] (computational and then manual)
* other modalities .ref[(?)]

## 2. Revisiting Chinese ideophones

After these case studies, what is typical for 'the Chinese ideophone'?

## 3. Qualitative studies
* particular constructions
* multimodality
* acquisition

---

# Contributions and expected findings

## .markeer[Bridging] the literature:

* Western languages: ideophone studies 
* Japanese: mimetics studies
* Chinese:
    * onomatopoeia
    * binomes
    * reduplication studies

## Methodologically:
* approaches grounded in Cognitive Linguistics that focus on .markeer[lexical semantics] (.markeer[categorization] and .markeer[variation])
    * mental spaces
    * frames
    * domains and ICMs
    * image schemas
* computationally aided (.markeer[empirical usage-based data])

---

# Contributions and expected findings

## Expected findings

I expect to:

1.  find out ***that*** .markeer[variation] is important for Chinese ideophones, in the three poles (.markeer2[meaning], .markeer2[phonological form], and .markeer2[written form]) of the Chinese folk model.
1. find out ***how*** the interplay between variation of these three poles is structured.
1. find out ***how*** this variation evolved over time.
1. find out ***that*** the functions of Chinese ideophones becomes more clear, more specifically from the perspective of mental spaces theory.
1. find out ***how*** this reflects back on what is cross-linguistically known about ideophones. More precisely, how .markeer[cross-modality] — in this case the .markeer2[written form] — influences the meaning and usage of (Chinese) ideophones.

---

# Selected references

.font70[
Academia Sinica 中央研究院. 2015. Scripta Sinica Database (Hanji quanwen ziliaoku jihua 漢籍全文資料庫計畫). Database. Scripta Sinica Database. http://hanchi.ihp.sinica.edu.tw/ (26 June, 2016).

Bybee, Joan L. & Paul J. Hopper (eds.). 2001. Frequency and the emergence of linguistic structure (Typological Studies in Language 45). Amsterdam: Benjamins.

Croft, William. 2001. Radical construction grammar: Syntactic theory in typological perspective. Oxford: Oxford Univ. Press.

Dingemanse, Mark. 2011. The meaning and use of ideophones in Siwu. Nijmegen: Radboud University Nijmegen dissertation.

Dingemanse, Mark. 2012. Advances in the cross-linguistic study of ideophones. Language and Linguistics Compass 6(10). 654–672.

Fauconnier, Gilles. 1994. Mental spaces: aspects of meaning construction in natural language. Cambridge ; New York, NY, USA: Cambridge University Press.

Fauconnier, Gilles & Eve Sweetser (eds.). 1996. Spaces, worlds, and grammar (Cognitive Theory of Language and Culture). Chicago: University of Chicago Press.

Fauconnier, Gilles & Mark Turner. 2003. The way we think: conceptual blending and the mind’s hidden complexities. New York, NY: Basic Books.

Fillmore, Charles J., Paul Kay & Mary Catherine O’Connor. 1988. Regularity and idiomaticity in grammatical constructions: the case of Let Alone. Language 64(3). 501–538.

Forceville, Charles & Eduardo Urios-Aparisi (eds.). 2009. Multimodal metaphor (Applications of Cognitive Linguistics 11). Berlin ; New York: Mouton de Gruyter.

Geeraerts, Dirk. 1997. Diachronic prototype semantics: a contribution to historical lexicology (Oxford Studies in Lexicography and Lexicology). Oxford ; New York: Clarendon Press ; Oxford University Press.

]

---

# Selected references

.font70[
Geeraerts, Dirk, Stefan Grondelaers & Peter Bakema. 1994. The structure of lexical variation: Meaning, naming, and context (Cognitive Linguistics Research 5). Berlin: Mouton de Gruyter.

Goldberg, Adele E. 1995. Constructions: a construction grammar approach to argument structure (Cognitive Theory of Language and Culture). Chicago: University of Chicago Press.

Goldberg, Adele E. 2006. Constructions at work: the nature of generalization in language (Oxford Linguistics). Oxford ; New York: Oxford University Press.

Goldberg, Yoav & Omer Levy. 2014. word2vec explained: Deriving Mikolov et al.’s negative-sampling word-embedding method. arXiv:1402.3722 [cs, stat]. http://arxiv.org/abs/1402.3722.

Haiman, John (ed.). 1985. Iconicity in syntax: proceedings of a Symposium on Iconicity in Syntax, Stanford, June 24 - [2]6, 1983 (Typological Studies in Language 6). Amsterdam: Benjamins.

Haiman, John. 2018. Ideophones and the evolution of language. Cambridge University Press. doi:10.1017/9781107706897 (26 November, 2018).

Heylen, Kris, Dirk Speelman & Dirk Geeraerts. 2012. Looking at word meaning: An interactive visualization of semantic vector spaces for Dutch synsets. Proceedings of the EACL 2012 Joint Workshop of LINGVIS & UNCLH 16–24.

Heylen, Kris, Thomas Wielfaert, Dirk Speelman & Dirk Geeraerts. 2015. Monitoring polysemy: Word space models as a tool for large-scale lexical semantic analysis. Lingua 157. 153–172. doi:10.1016/j.lingua.2014.12.001.

Hinton, Leanne, Johanna Nichols & John J. Ohala (eds.). 1994. Sound symbolism. Cambridge [England]: Cambridge UP.

Hiraga, Masako. 2005. Metaphor and iconicity: a cognitive approach to analysing texts. Houndmills, Basingstoke, Hampshire ; New York: Palgrave Macmillan.
]

---

# Selected references

Hsu Chen-Pang 徐振邦 (ed.). 2013. Lianmianci da cidian 聯綿詞大詞典 [Great dictionary of binomes]. Beijing: Commercial Press.

Johnson, Mark. 1987. The body in the mind: the bodily basis of meaning, imagination, and reason. Chicago: University of Chicago Press.

Jonker, Dirk R., Daan J.J. van Esch & Burchard J. Mansvelt Beck. 2011. Wenyan partikels. Leiden: Stichting Shilin.

Kövecses, Zoltán. 2017. Levels of metaphor. Cognitive Linguistics 28(2). doi:10.1515/cog-2016-0052. http://www.degruyter.com/view/j/cogl.2017.28.issue-2/cog-2016-0052/cog-2016-0052.xml (14 October, 2017).

Lakoff, George. 1987. Women, fire, and dangerous things: What categories reveal about the mind. Chicago: The Univ. of Chicago Press.

Lakoff, George & Mark Johnson. 1980. Metaphors we live by. Chicago: University of Chicago Press.

Langacker, Ronald W. 1987. Foundations of Cognitive Grammar 1: Theoretical prerequisites. Stanford, California: Stanford University Press.

Langacker, Ronald W. 1991. Foundations of Cognitive Grammar 2: Descriptive application. Stanford, California: Stanford University Press.

Langacker, Ronald W. 2008. Cognitive grammar: a basic introduction. Oxford ; New York: Oxford University Press.

Lockwood, Gwilym. 2017. Talking sense: The behavioural and neural correlates of sound symbolism. Nijmegen: Radboud University Nijmegen PhD dissertation.

Mikolov, Tomas, Ilya Sutskever, Kai Chen, Greg Corrado & Jeffrey Dean. 2013. Distributed representation of words and phrases and their compositionality. In Christopher J.C. Burges, Léon Bottou, Max Welling, Zoubin Ghahramani & Kilian Q. Weinberger (eds.), Advances in neural information processing systems (Proceedings of Neual Information Processing Systems [NIPS 26]), 3111–3119.
]

---

# Selected references

Mikolov, Tomas, Wen-Tau Yih & Geoffrey Zweig. 2013. Linguistic regularities in continuous space word representations. Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational LInguistics: Human Language Technologies 746–751.

Nuckolls, Janis B. 1996. Sounds like life: sound-symbolic grammar, performance, and cognition in Pastaza Quechua (Oxford Studies in Anthropological Linguistics 2). New York: Oxford University Press.

Packard, Jerome Lee (ed.). 1998. New approaches to Chinese word formation: morphology, phonology and the lexicon in modern and ancient Chinese (Trends in Linguistics: Studies and Monographs 105). Berlin ; New York: Mouton de Gruyter.

Packard, Jerome Lee. 2000. The morphology of Chinese: A linguistic and cognitive approach. Cambridge, UK ; New York, NY, USA: Cambridge University Press.

Peirsman, Yves, Dirk Geeraerts & Dirk Speelman. 2015. The corpus-based identification of cross-lectal synonyms in pluricentric languages. International Journal of Corpus Linguistics 20(1). 54–80. doi:10.1075/ijcl.20.1.03pei.

Perniss, Pamela & Gabriella Vigliocco. 2014. The bridge of iconicity: From a world of experience to the experience of language. Philosophical transactions of The Royal Society 369. 1–13.

Radden, Günter & Klaus-Uwe Panther (eds.). 2004. Studies in linguistic motivation (Cognitive Linguistics Research 28). Berlin ; New York: Mouton de Gruyter.

Silge, Julia. 2017a. Word vectors with tidy data principles. [2017-30-10]. https://juliasilge.com/blog/tidy-word-vectors/ (4 November, 2018).

Silge, Julia. 2017b. Tidy word vectors, take 2! [2017-11-27]. https://juliasilge.com/blog/word-vectors-take-two/ (4 November, 2018).

Simone, Raffaele (ed.). 1995. Iconicity in language (Amsterdam Studies in the Theory and History of Linguistic Science v. 110). Amsterdam ; Philadelphia: J. Benjamins.
]

---

# Selected references

Stefanowitsch, Anatol & Stefan Th. Gries. 2003. Collostructions: Investigating the interaction of words and constructions. International Journal of Corpus Linguistics 8(2). 209–243. doi:10.1075/ijcl.8.2.03ste.

Taub, Sarah F. 2001. Language from the body: iconicity and metaphor in American Sign Language. Cambridge, UK New York: Cambridge University Press.

Verhagen, Arie. 2005. Constructions of intersubjectivity: discourse, syntax, and cognition. New York: Oxford University Press.

Voeltz, Erhard Friedrich Karl & Christa Kilian-Hatz (eds.). 2001. Ideophones (Typological Studies in Language v. 44). Amsterdam ; Philadelphia: J. Benjamins.

Wielfaert, Thomas, Kris Heylen & Dirk Speelman. 2013. Interactive visualization of semantic vector spaces for lexicological analysis. TALN-RÉCITAL 2013 154–166.
]