Phonaesthetics don’t get no respect.

It is, or purports to be, a linguistic discipline investigating people’s reactions to particular speech sounds, and how those reactions influence spoken language.

Difficult things to study, it turns out. A few studies of phonaesthetics are well-known for the cleverness of their design and the nebulosity of their results. My Psych 101 book contained the famous Kohler maluma-takete experiment, in which speakers practically without exception called a rounded shape a “maluma” and an angular one a “takete.” A few studies have found that people can correctly assign meanings to polar antonyms (e.g. hot-cold) in languages they do not know with accuracy decidedly exceeding chance. How do they do it, though? No way to prove it’s the sound.

Introspection doesn’t help much. We tend to have some kind of idea that some words or names are pretty and others are ugly: witness the widespread disdain for the humble word “blog.” We tend not to have any idea on what we are basing our judgments. Worse, said judgments do appear to vary a bit depending on our native language.

Matters only worsen on investigation of phonaesthetics’s actual influence on language. If it has one at all, it’s extraordinarily slight; the operation of ordinary sound laws, never mind analogy, wipes it out easily. There is indeed the phenomenon of onomatopoeia, words representing sounds or soundmakers that resemble the sound in question. Even those, though… the obviously onomatopoeic Old English gans has become the much less obvious goose, just as OE crawe has become crow. The evolution away from the more onomatopoeic forms took place through perfectly ordinary sound laws, laws phonaesthetics appears to have been powerless to forestall.

So if it has no discernable effect on language change, why should a linguist bother with it? say a lot of linguists, washing their hands. Leave that stuff to those loonies over in lit-crit.

Just in this century, however, a phenomenon has popped up that ought to offer phonaestheticians (I assume that’s the word? or is it “phonaesthetes,” perchance?) new data and new hope: invented languages, in fantasy, science fiction, and elsewhere.

Take the grandmaster of language inventors, JRR Tolkien. Euphony (not to mention its opposite, whatever that is—dysphony, maybe?) was a major concern for him. He up and said so. So if we want to find out what an English speaker thinks is euphonious, Quenya and Sindarin ought to be major sources of data. Black Speech similarly illustrates dysphony.

I grant you that most fantasists aren’t as good at this as Tolkien. (Are any?) Even so, a sufficient sample ought to provide some curious and possibly valuable insights.

A couple-three years back I gave a con talk on invented fantasy languages. I seized on Lord Dunsany as an example, because he was not any kind of a linguist, but his nomenclature (once you strip out the odd bits of Greek and pseudo-Egyptian and whatnot) hangs together remarkably well. (My point being, one need not be a trained linguist to invent decent nomenclature, so why are so few fantasists doing it?)

So I made huge lists of Dunsanean place- and person-names and just stared at them for a while. And damn if patterns didn’t emerge. One such pattern turned out to be so striking both inside and outside Dunsany that I went and named it: Dorothea’s Law of Velar Villainy. The more villainous you are, the more velars and postvelars in your name or your language. Corollary: the more villainous you are, the more likely your name or language is to contain velars or postvelars in syllable-final and word-final positions.

Velar consonants are pronounced by pulling the back of the tongue up against the velum, the soft area at the back of the mouth. The “k” in “kite” is velar. So are the “g” in “go” and the “ng” in “sing.”

In Dunsany, with only a few exceptions, every character whose name ends in a velar consonant is a villain. (The most singular exception is Sarnidac of “The Relenting of Sarnidac,” and he may well have come by his name via the scorn of his fellow villagers.) The converse does not necessarily hold, incidentally, as Dunsany’s arguably evillest villain is the horrible Emperor Thuba Mleen, in whose name nary a velar is to be found.

Now consider Black Speech again, and the Ring Verse. Positively bristling with velars, particularly syllable- and word-finally.

Now consider Klingon, invented back when the Klingons were still Trek’s baddest villains. All over velars and postvelars, you betcha.

Coincidence? Honestly, I don’t think so. For some reason, we English speakers think velars are bad and ugly and nasty-sounding. Probably why some of us don’t like the word “blog” with its big fat velar at the end.

I happen to think there’s a similar but less strong effect with labial sounds, but I haven’t quite worked out whether this is through their frequent association with and transformation to velars in human language, or via some other mechanism. LeGuin’s Pravic (from The Dispossessed) is a possible data source here, as it actually includes labiovelar consonants (kw, gw) and seems to have been designed to be dysphonious, or at least prickly.

An interesting thing about Pravic (well, there are several interesting things about Pravic, actually, but I’ll limit myself to this one) is its use of clipping for familiar nicknames (Shev for Shevek, Dap for Bedap, and so on). Very Englishy, that. Most other languages I know of tack on a diminutive or transform the name entirely (e.g. Russian Yevgeny > Zhenya).

But we English-speakers clip names all the time, quite frequently down to one syllable. So LeGuin uses the same technique (consciously or not; she might well have known what she was doing as she did it, but I can’t say), pretty much entirely with sympathetic characters, to break down reader resistance to what are otherwise decidedly un-English names.

There’s plenty of fodder here for Legitimate Academic Research, but I frankly haven’t the patience. So here are the ideas for your perusal. If you want to do the legwork of data gathering, data analysis, citation tracking, and so on—be my guest.