Dr. Goodword's Office

A History of an Historical Quirk

Does English have Hs before Unaccented Syllables?

Dr. Goodword (Robert Beard, PhD, Linguistics)

One of the puzzles of the English language that has irritated us for longer than you might expect is the use of "an" before "historical" but not before "history," while "an" is used by some before both "hysterical" and "hysteria." To understand the issue here, we need to understand the nature of the sound [h] in the English language and its relation to the letter "h" - remember, letters and sounds do not always correlate in English.

Despite the torment endured by English speakers learning the spelling system of their native language (see 'The Chaos'), we often fail to distinguish the spelling system from the sound system (phonology) of English. In this essay, I will be speaking of the English sound system, not spelling. We already know some disparities: the [h] is not pronounced in "honor" nor before [w] in many dialects, so that "which" [hwich] is pronounced [wich], the same as "witch." Did you get the system? Double quotes marks letters while square brackets denote sounds.

[h] is a "glottal fricative." That means that it is pronounced by closing the glottis briefly, slowing the flow of air. The folds of the glottis are not completely closed or the results would be a glottal stop, the sound you hear between the two [o]s in "Oh-oh! Rather the folds approach each other until they cause friction with the air coming up from the lungs-hence, "fricative."

The sound [h] is pronounced in the very back of the vocal tract, that area from the glottis - the vocal cords - to the lips along which humans produce the sounds of language. In fact, it is produced by closing the glottis until a bit of audible friction is produced. Earlier it was pronounced by raising the tongue to a point on the soft palate (the velum) until it caused audible friction.

This is the same position where we pronounce [k] and [g] today by completely stopping the flow of air momentarily. [k] and [g] are thus called "velar stops." If you do not stop the flow of air but only cause friction by raising your tongue to that spot, you get the velar fricative of "ch" in Scottish "loch" and in German "Buch" and in all those Russian words spelled with "kh" in English, such as "Khrushchev."

This is the sound [h] developed from. The velar fricative was found everywhere in English words and was spelled "gh" when it occurred anywhere but the initial position of words. So, the "gh" [kh] in light" and "laugh" are of the same origin as [h], Proto-Indo-European [k]. (Proto-Indo-European or PIE is the language most of the languages of India and Europe developed from.)

That is why the Latin roots (the part of the word aside from the endings) for "light" (lux [luk-s]) and "night" (nox [nok-s]) end on [k] (the -s is a Latin suffix)while the English correlates contain "gh" [kh]. Proto-Indo-European [k] changed to [kh] in Germanic languages like English (see also German "Nacht" and "Licht") but not in Latin.

Now we can see that this sound [kh] has changed a lot in English. At the end of a word after [u] it is usually pronounced [f] today ("laugh," "cough," "rough"), while elsewhere it has disappeared from the pronunciation but not from the spelling: "thought," "light," "fraught." At the beginning of words, it became [h]: "howl," "high," and don't forget the order of pronunciation of "what" [hwaht], "which" [hwich], and "why" [hwI].

So why did the changes in this sound stop here? Well, who said they did? Since [kh] ("gh") has disappeared from most words unless preceded by a "u," we would expect [h] to do the same thing. We know that in many British dialects, it has already. As you recall from "My Fair Lady," Eliza Doolittle's Cockney dialect has no [h]s at all, to the great dismay of 'Enry 'Iggins, as Eliza called the eminent linguist Henry Higgins. Now, many of those Cockney speakers immigrated to North America, Australia, and New Zealand, bringing their concerted disinterest in the sound [h] with them.

In fact, you may be in the process of dropping consonant yourself! Listen to yourself very carefully and pronounce "history," then "historical." Do you hear as strong an [h] in "historical" as you do in "history?"

Now, try "the history" and "the historical." A little-known fact about English is that whether a noun begins with a consonant or vowel affects not only the selection of "a" and "an," but also the pronunciation of "the": before a consonant this article is pronounced [thuh] but before a vowel it is pronounced [thee]. If you want to say [thee] before "historical," or if this pronunciation sounds right to you, you are not pronouncing the initial [h] in that word.

Although the sound [h] is disappearing from many dialects of English, it is not dropping out in every word; it is dropping only from those syllables that are unaccented, hence pronounced more lightly. This means that "history" retains its [h] because the syllable it introduces is accented. The accent falls on the second syllable of "historical," so the word AS PRONOUNCED, not as written, begins with a vowel in many dialects. This means that "the" should be pronounced [thee] and "an" is the appropriate indefinite article in those dialects but only those dialects. "A historical" remains preferable in formal English.

Notice that the same principle applies to "hysteria" and "hysterical" but the accent is on the second syllable in both these words, so in the same dialects in which one would say "a history" but "an historical," we would expect "an hysteria" and "an hysterical."

Language is not a static phenomenon; it is constantly changing. That is one of the reasons we cannot count the number of words in any language. It is also why we have dialects and why we are not always sure which is the right way to say something.

Although languages are passed on from generation to generation, they are passed on imperfectly and with every new generation they change in very small ways. Over periods of hundreds of years can we actually map these changes and see them clearly, but they are happening right now in every language on earth. Our difficulty in deciding whether to say "an historical" or "a historical" is evidence of that fact.

Back to Dr. Goodword's Office