Talk:Sound symbolism

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Philosophy (Rated Start-class)
WikiProject icon This article is within the scope of the WikiProject Philosophy, which collaborates on articles related to philosophy. To participate, you can edit this article or visit the project page for more details.
Start-Class article Start  This article has been rated as Start-Class on the project's quality scale.
 ???  This article has not yet received a rating on the project's importance scale.

This whole article is a mess.

First, there are problems with organization. For example, the last paragraph begins, "/p/ has quite a similar symbolism because the process of making the sound /p/ is identical except that /b/ is voiced, and /p/ is unvoiced." /p/ has a similar symbolism to what exactly? The previous paragraph (and in fact the entire section) was about Ramachandran's ideas on metaphors. And the section is entitled "Relationship with Neuroscience".

Second, the tone is as far from encyclopedic as possible. It's more of a fan page for Margaret Magnus's pop-science book "Gods of the Word" than an article. Here's a good example: "The sounds of these words, evoke images, and images have an emotional impact, and every advertiser knows that emotions determine behaviour! Prozac was originally called Fluoxetine before the marketing guys got hold of it...."

Third, the article is extremely POV, being nothing but Magnus's ideas, and all of the support she's found for those ideas in historical and modern times.

There clearly is something to phonosemantics. Nobody disputes intra-language iconism in Japanese, Yoruba, etc. Nobody doubts that the pook/pak/peek experiments show at least some inter-language iconism.

However, this quote from the article is another story: "/b/ forms blunt bulges, whereas /p/ tends to precise points; /b/ blows up, whereas /p/ simply pops". It certainly sounds cute, but it means very little, and what it means is wrong. Think about "pan" vs. "ban" or "lap" vs. "lab". How precise is "plateau", how blunt is "bristle"? (It's also a possible copyvio from Magnus's book and/or site, but I haven't checked that.)

The real problem is that there's no established consensus out there. Linguistics as a whole is a pretty young and under-manned science, and this is at best a nascent sub-speciality of minority interest.

Do a quick search on sci.lang or LINGUIST for works like phonosemantics, iconism, ideophone, phonesthetics (there's not even a consensus on the terms...) and you'll see that almost everyone making strong claims disagrees with almost everyone else. And the quality ranges from the plausible scholarly speculations of John Lawler to the usual cranks trying to prove that their national language is descended from Sumerian. To an outsider, it's hard to tell what to believe.

Wikipedia should have an article on phonosemantics (or some other related term). But this--or anything else pulled straight from "Margo's Magical Letter Land"--is not it. —Preceding unsigned comment added by (talkcontribs)

I couldn't agree more! There are many problems with this article, and to be honest I wouldn't know where to start fixing them. It says that many of the phenomena aren't universal, yet happily goes on giving English examples. It cites Plato's Cratylus crudely out of context, without the crucial part of Socrates admitting that his theory is problematic as there are innumerable counterexamples. The tone is far too colloquial. The section 'Application of Phonosemantics' is simply original research (I will remove it in a minute, see WP:CITE). It looks like the part on Japanese phonosemantics has been copy-pasted from another article without any regard for structure or flow, and I wouldn't be surprised if there are copyvios to be found in other parts of the texts.
There are a few useful things, however, especially on the history of phonosemantics. Merging the useful material into Phonosemantics and redirecting this one might be the start of a solution. — mark 16:17, 18 February 2006 (UTC)
Certainly, the article needs work. Overall, however, it is a much better treatment of phonesthesia than we might have feared. I really think you are being a bit harsh with Magnus, dear, though I tried to tone down the plug for "Gods of the Word." The fact is that phonesthesia has long been a taboo subject in circles traveled by serious linguists, which has harmed scientific endeavor in this field, creating a situation in which an encyclopedia must make an extra effort to maintain NPOV. I have read that book and really do not think calling it "pop science" is fair or justified, though many 'official' linguists who have their own positions to defend would certainly agree with such a judgment. In this case, the scholarly consensus itself may reflect substantial bias. That is what Magnus' popularization stems from.
With time, we ought to be able to improve the treatment of this subject in Wikipedia. In my own view, the concept of the phonestheme deserves separate treatment from the many issues of sound symbolism generally. With hopes of further improvement to come in the future, I have at least taken the step of linking to the other article, to which I added a minor item, a couple of (initial) references, and some internal links.
I may not have enough experience as a wikipedian to fully grasp the point you were making, Mark, about merging some articles but I do concur that reorganizing material now in various different Wikipedia articles would help with the reorganization of this one. Bibliography for this subject is extremely problematic but is something I could help with (although I'm not sure how much detail would be appropriate in a generalist reference like Wikipedia).--Ph7five 14:43, 16 April 2006 (UTC)
I wrote the original article in the early days of Wikipedia when standards were not so high, and I'm sorry that people have not found my efforts up to scratch. It was a case of having to start somewhere and someone having to start - I was the one committing time and energy back then so that you'll would have something to complain about! Please let everyone note that my main source was Margaret Magnus' PhD thesis, particularly her extensive literature review, not her more popularly oriented book. Please note also that at the time she, in a letter to me, approved of my use of her material: "cool" was her adjective. (Perhaps she was just being kind?) Magnus has showed that initial word sounds in monomorphemic words are very closely correlated with 'meaning' for that word - e.g. words beginning with /m/ do fall into a very small no of what Lakoff would call "basic level categories", and when compared with other phonemes there is very little overlap of categories represented - well outside what a random (ie arbitrary) distribution would suggest. I've repeated some of her experiments for the whole of the Oxford Concise ed, 10,000's of words, and can confirm this. I have yet to see any serious effort to show how or why her reasoning is flawed, or an attempt at an alternative explaination for the considerable data that she amassed during her research. Meanwhile the University of Trondhiem apparently did not consider Margaret's research to be "junk science" since they granted her a PhD on the basis of it. Please do fix up my poor writing by all means. mahaabaala 20:04, 4 May 2007 (UTC)
There is nothing here on the commercial uses of sound symbollism which is a shame. While the academics fiddle the marketing departments of large multinationals have realised that this is a real phenomena and now use the research to design product names and pay big bucks for it. mahaabaala 20:04, 4 May 2007 (UTC)

I wrote the part on Japanese phonosemantics myself (section "Phenomimes and psychomimes"), and it certainly isn't a copy-paste from anywhere. I agree that I had a hard time finding a place for it in the article. Please feel free to move it where you think it belongs, and edit it if you think it's wrong. Other than that, I agree, the article is not the best around. — AdiJapan  16:29, 18 February 2006 (UTC)

I have in my last few edits pruned a lot of nonsense and I have tried to do away with the uncritical endorsements of Manske as much as possible. I have also put the Cratylus thing right, noting that Socrates conceded that "My first notions of original names are truly wild and ridiculous" after being faced by an overwhelming number of counterexamples. There is still very much to be done. Phonosemantics is an intriguing phenomenon, well worth a good article, as our anonymous editor above points out. I'm not sure if I have the time to do something about it in the near future; I surely hope others can help out and to that end I'll put up a notice at Wikiproject Languages. — mark 19:33, 18 February 2006 (UTC)
Yes my own understanding of the text has come along since I first used that quote. However there is a lot of humour and wisecracking in the dialogue. The more honest accounts admit that they are not sure when Socrates is being serious, or when he is being self-deprecating, or when he is denouncing his own statements. Overwhelming is surely an overstatement. A number of counter examples are offered and show up a possible weakness in the method, but Socrates does not completely abandon it - he steers a middle path between naturalism and convention. You seem to suggest that he takes up an extreme position, when the dialogue in fact concludes nothing and Socrates adopts no position. It is quite possible to elucidate a word based on similar sounding words as Socrates, and his Indian counterpart Yaska do. It does actually work. What you can't do is assume that every word with a similar sound is related. The problem lies in an assumption that everyone seems to make which has two aspects: One that "things" are unitary and not complex, or that something complex has a simple 'essence' to which a word can refer; and two that any given word/sound only has one meaning. Personally I see protestant preocupations with canons and singular truths and ideal languages in this, and find it frustrating that contempoary linguists still seem to see things as having an 'essence' that a word corresponds to even when they think of the word as arbitarily chosen. I think Lakoff has showed the way forward: things are complex and their membership of categories is complex; and names tend to correspond to how we interact with things; and are frequently metonymic, with the name relating to that part of a complex thing which most stands out for us, or with which we interact most strongly. Take this into account and Socrates' method is not only useful, but fun. mahaabaala 20:04, 4 May 2007 (UTC)

Sound symbolism applied in literature[edit]

There is not much in the article at present about sound symbolism (in many cases, before the modern formulation of what sound symbolism is) in literature. Poets especially have made much use of sound symbolism, e.g. Arthur Rimbaud, Stephane Mallarme, etc. I don't plan on discussing this in the article until I have references however. I remember for example a literary critic pointing out that Rimbaud's heavy use of the voiceless velar plosive in the first 5 or so sentences of his prose poem Villes was meant to give the impression of the clashing and dissonant sounds of the modern city. Alexander 007 08:48, 20 March 2006 (UTC)

See John Michell. Euphonics : a poets dictionary of sounds. Frontier Publishing, 1999.

Sound symbolism and synesthesia[edit]

As Ramachandran's co-author, I may not be the most impartial commentator, but I am certainly knowledgable about our attempts to link synesthesia with sound symbolism. I am quite happy to help work on the Relationship with Neuroscience section. I have been working mostly on the synaesthesia page recently, but added some mention of the links with other phenomena, including the kiki/bouba example. Note that this phenomenon was originally noted by Wolfgang Kohler (using the words takete and maluma) and expanded by Lawrence Marks. Subsequent to our 2001 article, follow-up experiments have shown that children as young as 2 1/2 show the "kiki/bouba" effect (Maurer et al., 2006), and that the effect can be replicated in more precise reaction time studies (Westbury, 2005). Edhubbard 23:45, 5 August 2006 (UTC)

What about people who do no speak Indo-european languages? mahaabaala 20:04, 4 May 2007 (UTC)


The bulk of sound symbolism is largely discredited by the vast majority of linguists. The article should begin with the fact that outside of onomatopoetics, trained linguists don't believe in it. This is junk science for the most part and it should be reported as such.

(The author of the above lines didn't sign them.)

On the contrary, there is a slowly growing body of evidence in favor of it from respected scientific sources. I'll agree that the vast majority of linguists "discredit" it-- but that doesn't mean much to me. The flat earth theory held wide sway at one point too. It would be worthwhile to give readers a sense of how controversial these findings are, and how they contradict established linguistic principles (Saussurian arbitrariness of the sign, etc.). I've done [of course unusable] original research in which participants agreed on the meanings of pseudowords with p<.001. Doesn't seem like junk science to me.

Some possible sources: Brent Berlin's Ethnobiological Classification, Bolinger's "The Sign is Not Arbitrary," Janice Nuckolls' "The Case for Sound Symbolism."

What I think is important is that etymology need not be entirely linear. If language speakers are aware of the sounds of their words along with their meanings, then some sort of Bolinger-style "associative lexical influence" can play into the evolution of the lexicon to develop clusters and phoneme-sememe correlations.

Hope this is helpful/convincing. Starvingpoet 07:48, 24 January 2007 (UTC)starvingpoet

I think the article does need a (brief) discussion of the fact that most mainstream linguists don't believe in this, though mainly to refute mainstream linguistics rather than to disprove sound symbolism. Linguistics for quite some time (and particularly in the past 50 years or so) has been obsessed with various theories of meaning that just ignore many philosophical issues of how meaning comes into being. Can a gesture or even basic body language have meaning? Can abstract visual art have meaning? Can music have meaning? Much philosophical debate in the past century has tried to classify these things in one way or another, but frequently the issue is a lack of imagination on the part of the theorists about how broad meaning can be. The fact is that things can often mean something (in a vague and nonspecific way) even if you couldn't write a specific lexical meaning down in a dictionary definition. And they can have a vague meaning that is context dependent to varying degrees (as most words/utterances are). If we're willing to accept a broader and more abstract kind of meaning, clearly the use of sounds and phonemes have enough patterns that we can say some of them have this kind of vague meaning.
Anyhow, the fact that there is little "scientific" evidence for most of this is because cognitive linguistics has been operating since its inception on the principle that such kind of meaning doesn't exist. As Starvingpoet points out, though, slowly but surely some cognitive scientists are catching on (even faster than most linguists).
But there does need to be a better discussion of how this all relates to mainstream linguistics, as well as a disclaimer pointing out that one could take either view too far (i.e., sounds can have some form of meaning, contrary to popular linguistic theories, but those forms of meaning also aren't going to be some sort of universal and specific meanings valid for all times and places). Just my thoughts.... 13:03, 2 July 2007 (UTC)


has a redirect to this article, but it isn't mentioned here. someone who is willing to explain that term? -- 21:36, 12 December 2006 (UTC)

sounds phonemes arent arbitrary[edit]

There is a great unknown thinker from Ipolyság Hungary currently Sahy Slovakia Balla Zoltán who attended Prague University in the thirties so he would be close to a hundred if still alive. He was an MD. Unfortunately my uncle stole my copy of his ouevre bequethed by my father who pretty much ridiculed his weird theories of carcinogenesis....something to do with muscle contraction causing the generation of carcinogens ........think of colon spasm and ca of the same....and even more lunatic theories of sound letter ie phoneme and grapheme concordance with meaning. Both knew Hungarian German Czech Latin Greek with a smattering of Russian English French Italian. The book of Balla whom my father considered one of the great intellects and original thinkers he ever met despite his odd theories has impressed me so much that unconsciously over the years I have totally internalised his theories so now I dont know what is his what is mine what I have discovered independently- I do not wish to subvert falsify his theories nor steal from his thunder but what follows is an amalgam of what I heard as verbal suggestion from my father who while he knew more languages than I perfectly he was a professor of silviculture but obviously widely read and cognisand of literatures in half a dozen languages or the book of Balla Zoltán his fellow student in Prague. A shameless and unapologetic proviso. No one in these linguistic wars mentioned the underlying Gestalt or Weltanschauung. Balla and my father myself and the thieving uncle all shared the same world view. But I will not pretend to explicate their views, this is mine. We or rather I do not know if there is a God, but we err with Pascal. I think that there is order in the Universe though it might be hard to discern. Newton Mendeleyev and Darwin agreed though for superficial what we call contemptously scientism they destroyed old shibboleths but their universe is more ordered than previously. So we look for connexion context structure meaning form rather than meaningless random chaos whether in physics medicine forestry musicology or finally in linguistics. This of course gets heavy as chaos itself is self replicating non random and self organising whether by fractals or apoptosis. I cannot obviously get into cosmology evolution or even the periodic table. Sorry for all those on the scientism side even the elements are not random but fall into a beautiful periodic table. People who deny the meaning of vocables are akin to petrologists who might deny the existence of elements in rocks.

  We will start with simple examples. i or the letter or grapheme i or

the sound i actually MEANS small pointed point thin needle. Hungarian itt means here ott means there itt ott here there the sound i means small pointed local here whereas ott with o means round far distant circular. No history of languages suggests any connexion between Hungarian and Latin yet the LATIN OF COURSE IS HIC and huc ubi or in other languages tam la da there etc. Here and ici both show the i. The simple single phoneme explication gets extraordinarily complex, take the Hungarian levél which means both leaf and letter and the French lettre and feuille and the Latin littera and folio and the Greek phyllis transposing the L AND THE F. As my knowledge of other languages is shaky and limited I have to stop there. Levél leaf feuille folio phyllis. While it is remotely possible that a nearctic language existed before the genesis of the IndoEuropean and FinnoUgric language families from which arose the L.F or F.L cluster or morpheme it is much more plausible that these hominids imitated the idea with their tongue and sound produced the ideas involved. F means light flighty frothy fair smoke like thin evanescent slight flimsy weak frilly, L of course means long thin linear flat lip flip lilting lily... more of these later.... So l.f or f.l actually means a flat weak object, a leaf a folio a feuille. To poets this of course is a joke, they always knew this and the trouveres troubadours who found the mot juste found exactly this, the phonological correlate of image thought and idea even of emotion. Now to our Bard's a rose by any other naME WOULD SMELL AS SWEET. wHICH OF COURSE IS Nonsense and is an example of authorial fallacy. Shakespeare isnt enunciating a general law of linguistics but is demonstrating the distress of Juliet as to why Romeo has to be the enemys son, ie he could be called Joe Bloe or rather Giovanni Bruni say. Then of course we would have no play by those names. So let us follow the techniques of linguistics and compare doublets. Take ROSE actually MEANS a fragrant round circular beautiful red flower. R is red. O is round circular perfect beautiful. S is odoriferous smelly or Z of course. So a rose by any other name of course wouldnt be a rose. Now take lily. We already know that L is linear long thin but it is also lilac blue white flat thin. i is of course pointed pin peak. So lily is a delightful thin long stemmed linear flower usually white blue or lilac.

Rose is a round sweet smelling red beautiful flower. The actual name and the graphemes say so.

l is a long linear letter. o is circular perfect round rolling spherical

That should be all for now. There are a myriad other examples we havent even scratched the surface. The poets have already delved into the depths.

littera and Greek phyllis transposing the L and F

but back to Cratylus de Saussure and Balla.

i is by its sound but also graphemic feature ie single dot means point small here

l is obviously long thin flat platte plate lip line linear in sound and writing

o is round perfect circular both in shape and lip muscular contour

ball globe bubble boule lob bulge need we go on?

The obvious glaring fact that the vast majority of words in any language cannot be so simply analysed need not detain or discourage us for it is not language which is meaningless random noise and chance script but that our understanding is rudimentary. A monkey doesnt comprehend embryology a dog evolution nor a crow geomorphology, which of course doesnt mean that they are wrong or unreal.

There is no pseudoscience here, you can either observe the same phenomena or not. If they are unreal irrational or meaningless then they are drivel, if on the other hand they are repeatedly observable and systematic then there is a real phenomenon first to be observed analysed proven or disproven finally to be incorporated into genuine knowledge.

As a minor afterthought zoologists have recorded and analysed the distress calls of mammals and birds and found the same pattern of descending ascending sounds notes and rhythms to signal grief joy distress alarm and reassurance.

And the same or similar musical intervals in dirges or funeral masses and lugubrious music of Mozart or Beethowen or genuine folksongs. So grief transcends not only languages but species.

And phonetic meaning languages.

Alexander Sándor Jablánczy —Preceding unsigned comment added by (talk) 08:43, 3 August 2008 (UTC)


This article is very POV. Given that languages often have many, many words (English at around a million), a much lesser number of phonemes, and there are only so many concepts to express, it's not surprizing at all that couincidences will be found. House, home, and hut mean the same. So what? Home, how, and hair do not. Homophones are a fabulous counterexample. (talk) 22:08, 22 June 2009 (UTC)

I was going to go the opposite direction and point out contrary to house, home, hut and hovel are abode, dwelling, demesne and residence. Theo in Greek and Teo in meso-American languages have similar sounds and identical meanings, but outside of "Gods from Outer Space" types, I don't think anyone thinks that more than a coincidence. And much of what is used as evidence here seems to fall into the same category of either cherry picking examples or simple coincidence. Or a bit of both. (talk) 19:31, 30 May 2013 (UTC)

Sound symbolism used in advertising[edit]

I certainly don't have the expertise to write about this, but if someone else does that would be great. I heard about a company, I think called Gap, that uses sound symbolism to come up with names for products. Most significantly if I remember correctly they came up with Dasani, the water brand from Coca Cola company. They have study groups that they give nonsense words to ask how it makes them feel, and there does seem to be a definite consensus about what words mean, for example if words sound like they describe something big or something small or something smart. I don't remember if this is cross-linguistic, which would be very significant, or if it is based on knowledge of other words (like teeny is smaller than tiny, so words with the first vowel sound smaller than with the second). —Preceding unsigned comment added by (talk) 04:46, 9 September 2009 (UTC)

  • True, we need to add sth about advertising and companies. I recently edited Altria to add this (addition in bold):
    • Philip Morris completed its rebranding to Altria Group... The name "Altria" is claimed to come from the Latin word for "high" and was part of a trend of companies rebranding to names that previously did not exist, Accenture and Verizon being notable examples,[1] though linguist Steven Pinker suggests that in fact the name is an "egregious example" of phonesthesia - with the company attempting to "switch its image from bad people who sell addictive carcinogens to a place or state marked by altruism and other lofty values".[2]
  • Maybe I'll add it here... but please, it'd be great if others could expand the section. Malick78 (talk) 11:24, 21 August 2013 (UTC)
Errrrr, what exactly does the Altria example have to do with sound symbolism?! The association of "Altria" with "altruism", even if perhaps not fully conscious in everybody who makes it, is based on the arbitrary meaning of "altruism", not any of the sounds/phonemes in "Altria". I think this example is plain wrong. --Florian Blaschke (talk) 13:23, 11 February 2014 (UTC)
Hmm, I think Florian has a point, although I fear I don't quite get it. Pinker's example seems perfect to me... the epitome of egregiousness, even...
however, referring to phonesthesia and not to sound symbolism, :-(((
Word-initial "altr-" is basically a hapax in English, at least for those of us who do not use "altricial" (completely different root) all that often. Thus, the coinage of "Altria" is especially felicitous (from the Philip Morris point of view) because it essentially creates an alliterative phonestheme with two or three members: "altruism" (with its derivatives), possibly "altricious," and the new brandname. Thus, I gratefully take Florian's point as evidence that we have created a problem in simply redirecting "phonesthesia" to "sound symbolism." I'm not happy about this because it would seem to open up a Wikipedian wormcan but -- at least to my mind -- it shows there is work cut out for us. I would like to hear others' opinions. - phi (talk) 15:05, 13 February 2014 (UTC)
I don't think altr- can be analysed as a phonestheme. It may not be a morpheme, but there is no basis of words starting in altr- with somehow similar or related seeming meanings to analyse it as a phonestheme, either. You say altr- is basically a hapax, and this demolishes your own point because a hapax cannot be a phonestheme virtually by definition if you consider the way phonesthemes are determined. There is simply no basis to associate anything in the way of even the vaguest meaning with an isolated singleton. You'd need at least two words with the supposed phonestheme. --Florian Blaschke (talk) 22:22, 13 February 2014 (UTC)
Moreover, I think altr- is too complex for a typical phonestheme. If you consider the examples given in the article, most of them are single phonemes or groups of two consonants, i. e., less than a syllable, even. I'd describe Altria simply as a more or less subtle allusion to the word altruism, evoking a positive, friendly, "humanistic" air. --Florian Blaschke (talk) 22:31, 13 February 2014 (UTC)

"Relationship with nature" Section[edit]

The "Relationship with nature" section looks like someone's pet theory and, frankly, metaphysical tosh. Thirteenangrymen (talk) 09:51, 3 May 2012 (UTC)
Cite error: There are <ref> tags on this page, but the references will not show without a {{reflist}} template (see the help page).