Monday, September 22, 2008

Fluttering Fleas, But Not Flabby Flannel

I’ve learned something new: There’s a name for something I have wondered about for a long time. Simply put, there’s a tendency for certain letter groups to appear in words that have similar meanings regardless of whether those words have a shared etymology.

The term is phonestheme.

The example first given in the Wikipedia page on the subject is the gl- phonestheme, which occurs in many English words referring to light and vision. (Glitter, glisten, glow, gleam, glare, etc.) An important aspect of these words is that neither the gl nor the letters that follow mean anything on their own. (No such word as itter or isten, for example.) They also don’t necessarily need to appear at the beginning of words. As Wikipedia puts it:
While phonesthemes have mostly been identified in the onsets of words and syllables, they can have other forms. There has been some argument that endings like -ash and -ack in English also serve as phonesthemes, due to their patterning in words that denote forceful, destructive contact (smash, crash, bash, etc.) and abrupt contact (smack, whack, crack, etc.), respectively.
Interesting, no? There’s a few aspects of this I haven’t wrapped my head around yet, so I feel like I should read more on the subject. (If anyone of you can point be toward good resources on the subject, I’d appreciate it.) For example, I don’t understand why this happens. I would also like to know what linguists make of outliers.

For example, I have the letters ack in my last name, but no one would lump that in with the -ack group. Is it because my ack occurs in the middle of the my last name? And what to make of back and rack and Zack? Why wouldn’t they fit the pattern?

As I said before, I had some idea that phonesthemes existed before I learned what to call them. A while back, the very cool and always informative word blog Bradshaw of the Future put up a post on the etymologies of various words that begin with the letters cl and that also refer to noise. Clack, clip, clap, cluck, and several others. I have always wondered about the fl words, which so often refer to various either quick motion or motion in the air, and I said so in a comment on the cl post.
I've always wondered but have never looked into the English words beginning with fl, which often describe some sort of quick movement, often through the air. Flick, float, flutter, flight, flap, fleet, flee, flit, flash, flea, fling, fly, flip, flurry, flag, flail, flare, flake, flay... There are even more than this. Not all share the association, of course. Flan, flabby, flannel, etc. All generally fun words to say, I've found. Have you any idea what the deal with these is?
Goofy, the Bradshaw blogger, responded last week with a post just on fl words, noting their various etymologies. Very helpful, particularly because it provided my first encounter with phonesthemes. Now I’m firing in a hundred different directions.
Some other phonesthemes for your face:
  • sn-, referring to mouth and nose goings-on: snarl, snout, snicker, snack, etc.
  • sl-, referring to frictionless movement: slide, sled, slick, etc.
  • -tter, referring to rapid, uneven motion: mutter, stutter, flutter, patter, titter, glitter, jitter, chatter, twitter
  • squ-, referring to eruptive sounds: squeal, squeak, squash, squirt, squawk, squabble, squall
As you may have noted, a lot of these words seem like they could be onomatopoetic. The -ack words, for example, could have arisen through imitation of noise created when the actions associated with them happen. This, however, doesn’t work for all phonesthemes, as the gl- words like glisten and glimmer don’t make noise. So where then must this mental association come from?

Are English phonesthemes shared by other languages?

If not, why shouldn’t they, if we’re apparently so hardwired to make these associations?

What is it about English, then, that prompts us to make such connections?

Could it be that people recognize the pattern on a few related words and then subconsciously only accept into their language new words that fit that pattern?

Do people even process language in this manner?

What would the advantage be of doing so?

I’m clearly full of questions, but that’s probably because this is the most interesting thing I’ve learned in a long time. I’m not clear on the linguistic definition of icon, which the Wikipedia page on phonesthemes uses in a section under the subject line “Important considerations.”
  1. If the words aggregated in these groups… were nothing more than arbitrary symbols, rather than being icons, their totally arbitrary nature would preclude any such systematic connections with meaning.
  2. Although groups of phonesthemes can be identified, this type of iconic sign, in any language, is rather rare.
  3. Despite [the above rule], there may well be far many more iconic words in, say, the English language, than we are aware of.
  4. It is also important that not each identified phoneme operates as a phonestheme every time it is used. (e.g., words square and squadron have no connection with discordant eruptive sounds common to the -squ group of squeal, squeak squall).
  5. Despite [the above rule], it is very clear that, in many cases, some level of iconic communication certainly does take place in a number of words within any language.
  6. Although people do seem to spend inordinate amounts of time creating a new word (such as deciding upon a “good name” for a new business), no one has ever been able to produce any evidence that speakers are consciously aware of such (iconic) signaling processes when they create new words.
Something about that sentence “There may well be far many more iconic words in, say, the English language, than we are aware of” strikes me as vaguely haunting.


  1. The "gl" phonestheme is shared by other Germanic languages, and the "cl" phonestheme seems to be present in Italic languages as well. So yes, we do find phonesthemes shared by more than one language. As for why all this happens, I don't know because I haven't done much reading on it either, but there are some references in the wikipedia article, and this LL post:

    afaik, an iconic sign is a sign that is perceived as resembling the thing it represents - eg, an onomatopeic word.

  2. another "frictionless" sl- would be slut.

    There's an Erica Jong joke in here somewhere, but I simply haven't the time.

  3. "There may well be far many more iconic words in, say, the English language, than we are aware of." That is creepy, kind of like "The calls are coming from inside the house" in the sense of something creepy already happening from within without your knowledge. I don't like the idea of my brain making decisions without my permission.

    At the same time, I wonder what the benefits might be to cracking this code. If you knew all of these and how they make people think, you could make words that have only positive connotations. You could manipulate people, completely invisibly.

  4. RW: there's no reason to suppose that iconic signs make people think in any specific way. Anyway, it was tried in 1984, and it didn't work.