Minds, Mechanisms and Interaction in the Evolution of Language -

Rationale

The Project

Why do humans seem to have a large suite of biases to associate certain perceptual domains with one another, and what impact do these associations have on language? Since at least ancient Greece, scholars have recognized that the relationship between words and meanings does not need to be entirely arbitrary. ‘Cat’ may not be an exceptionally good word for referring to a small domestic feline, and thus languages have different names for cats. At the same time however, words for the sounds made by cats are much more similar across languages.

Image credit to James Chapman: https://soundimals.tumblr.com/

 

 

Words like meow, which are used to describe or depict sounds are known as onomatopoeia, but they are far from the only example of a broader class of iconic associations between words and meanings. For a meaning to be iconic, it must bear some form of resemblance to its meaning. In the case of onomatopoeia this resemblance is imitative and obvious, but words need not be direct imitations of meanings to be iconic. For example, it has been shown that across nearly 2/3 of the world’s languages, there are statistically significant associations between certain sounds and meanings. For example the word for ‘nose’ in many languages contain nasal consonants like /n/:

 

ASJP Tables taken from Blasi et al. 2016

 

 

The number of these types of associations that have been demonstrated in the laboratory is fairly extensive. Starting in the 1920s, researchers began to discover what were originally termed examples of phonetic symbolism. Wolfgang Kohler, for example, found that experimental participants reliably associated the nonsense words ‘takete’ and ‘baluma’ (later maluma) with jagged and curvy images, respectively. Similarly, Erwin Sapir found that participants paired words containing high front vowels with small objects and words containing low back vowels with large objects:

 

The number of these types of associations that have been demonstrated in the laboratory is fairly extensive. Starting in the 1920s, researchers began to discover what were originally termed examples of phonetic symbolism. Wolfgang Kohler, for example, found that experimental participants reliably associated the nonsense words ‘takete’ and ‘baluma’ (later maluma) with jagged and curvy images, respectively (the Bouba-Kiki or Takete-Maluma effect). Similarly, Erwin Sapir found that participants paired words containing high front vowels with small objects and words containing low back vowels with large objects:

Takete and Maluma

 

Mil and Mal

 

Subsequent to these early findings, there have been a number of experimental booms in the study of iconicity, and the number of observed sound-symbolic associations between words and meanings has seen a concomitant explosion, although research into iconicity was often treated as marginal and unimportant. Currently, we are in a modern and sustained resurgence of interest in iconicity:

 

The Iconicity Boom

 

Despite the fact that research into iconicity is on the rise, it can still sometimes be best characterized as stamp collecting. Researchers have documented dozens of crossmodal associations, but these tests are often conducted using idiosyncratic methods that mean that comparing findings and making synthetic claims about the origins, strengths, or outcomes of these associations.

The goal of the study we will be conducting as part of the Summer School is to pilot a standardized set of tests of crossmodal associations between a variety of perceptual qualities. In the auditory domain, we will explore Pitch, Amplitude, and Noise; in the visual domain we will explore Shape, Size, Color, Brightness, and Speed of Movement; and in the affective domain we will explore a number of emotions.

These 9 domains offer a number of comparisons (36 to be exact): some of these have already been observed in the literature – we know, for example, that small objects are associated with high pitch, fast movement, and brightness. Similarly, high pitched sounds are thought of as being happy (rather than sad), bright, jagged, and fast. Of the 36 associations, approximately twenty have been tested, but never in the same study and never using the same stimulus materials. In the interest of open, productive, and generalisable science, will will test all associations exhaustively.

There are several productive reasons to take this kind of approach:

  1. Stimuli can be standardized for use in further experimental designs
  2. Relatedly, stimuli can be refined for deployment using other groups of participants. Sound symbolism is often claimed to be universal, but the number of cross-linguistic and cross-cultural tests of even well-studied phenomena like the Bouba-Kiki effect is still vanishingly small.
  3. Crossmodal associations can be compared based on the relative strength of their associations
  4. We can look for mediators or moderators- are all observed biases veridical perceptuocognitive biases, or are they mediated. If, for example, we know that people associate Small objects with Brightness, Brightness with High Pitched sounds, and High Pitched sounds with small things, can we determine the relationship between those biases? Given a relationship like this one, it is possible that participants have associations between Small objects and Brightness, and Small objects and High Pitched sounds, and that the observed association between Brightness and High-Pitched sounds is a direct outcome of those two features being associated with small objects

 

Suggested Reading:

Please note that no reading is *required* for participation in the summer school, nor is any knowledge about Sound Symbolism/Iconicity research. If you find the topic interesting, or with to enhance your immersion in the summer school, here are some sources that should bring you partially up to speed.

Ramachandran, V.S., & Hubbard, E. (2001). Synethesia, a window into perception, thought, and language. Journal of Consciousness Studies, 8, 3-34. pdf

Nielsen, A.K.S. & Rendall, D. (2011). The Sound of Round: Evaluating the role of consonants in the classic Takete-Maluma phenomenon. Canadian Journal of Experimental Psychology, 65, 115-124. pdf 

Dingemanse, M., Blasi, D.E., Lupyan, G., Christiansen, M.H., & Monaghan, P. (2015). Arbitrariness, iconicity, and systematicity in language. Trends in Cognitive Science, 19, 603-615. pdf

 

 

Last checked 2017-07-03 by Alan Nielsen

Street address
Wundtlaan 1
6525 XD Nijmegen
The Netherlands


Mailing address
P.O. Box 310
6500 AH Nijmegen
The Netherlands

Phone:   +31-24-3521911
Fax:        +31-24-3521213
E-mail:   


Public Outreach Officer
Charlotte Horn