Wednesday, March 4, 2009

Linguistic egotism, or, “The I’s have it”

I — in English, the first-person singular personal pronoun, nominative case. One of the most common words you will encounter in the language. The second-person you sometimes appears before I in frequency charts — I before you, except after who? Or is it, I after me, except before thee? — but you have to remember that you is nominative, accusative, and dative, all rolled into a single form. But I + me > you, every time.

I is also one of our shortest words (only the indefinite article, a, and the vocative interjection, O, are as short). English, moreover, is the only language that capitalizes its first-person singular pronoun. Other languages (e.g., German, Italian, Spanish) bestow this nicety on the polite forms of the second-person, but not English. No, in English, it is I who is most important; it is my perspective that matters more than any other. The I of the beholder, you might say. The first Roman numeral, I, coming before all others, putting the I in veni, vidi, vici. Try to demote the letter into lower-case and you get an imaginary quantity. We simply cannot conceive of diminished personal primacy, can we? Id pro quo? No, rather, ego vincit omnia. No need to dot this I. But underline it? Aye.

What was it that started me eyeing this pronoun, you might ask? Some of you who know me might say I’ve been I-obsessed for years, but there was something else — a recent piece in the Times Online, which claimed:
A “time traveller’s phrasebook” that could allow basic communication between modern English speakers and Stone Age cavemen is being compiled by scientists studying the evolution of language. Research has identified a handful of modern words that have changed so little in tens of thousands of years that ancient hunter-gatherers would probably have been able to understand them. Anybody who was catapulted back in time to Ice Age Europe would stand a good chance of being intelligible to the locals by using words such as “I”, “who” and “thou” and the numbers “two”, “three” and “five”, the work suggests.
Three pronouns and three numbers — not much of a phrasebook, is it? I guess it might work for the Neolithic dating scene. More to the point, what of the suggestion that the pronoun, I (as opposed to the Jason, I), “would stand a good chance of being intelligible”? I don’t think so. Aren’t these scientists forgetting something rather important, namely a little thing we like to call the Great Vowel Shift? As pronounced today, I bears little resemblance to its earlier forms. Compare the sound of English I to PIE *(h)eĝ. Hmm, in the words of Struther Martin, “what we’ve got here is ... failure to communicate.” Ironically, (h)eĝ sounds more like the stereo-typical grunts of early hominids than I ever could.

Now let me be clear: to a philologist’s eye, there is indeed a likeness between I and *(h)eĝ — just as there is between I and Sanskrit अहम्, Hittite ūk, Greek ἐγώ, Latin ego, Umbrian eho, Gothic, Old Saxon, and Old Frisian ik, Old Norse ek, Old High German ih, Old and Middle Danish iak, æk, Old Swedish iak, iæk, and so on and so forth — especially when one compares them to the Old English ic (and Middle English ich). Likewise, there’s a philological resemblance between Modern English I and the same pronoun in other modern languages, e.g., Italian io, Spanish yo, Swedish jag, Czech , Russian я — just to give a more or less random assortment.

But to a speaker or a listener — that is to say, to the ear, not the eye — there is a great gulf between the sound of Modern English I, as it has been pronounced for the last several hundred years, and any of the earlier forms of the pronoun, let alone the earliest. It’s yet one more way in which English has differentiated — even canonized — its pronoun. Even Modern Frisian and Dutch, two languages arguably the closest to English, retain ik(ke) and ik, respectively. Dutch does have mijn “my”, pronounced just like English “mine” — that’s pretty close. If the nominative first-person pronoun had become *Ij in Modern Dutch, we might have had some company up here atop this I-ful tower, but alas, no.

It’s just English. And considering all the foregoing, I don’t think I should feel too bad about having rather a big ego. All English speakers are pretty full of themselves — aren’t they? — whether backed by the erstwhile imperialism of England, the rugged individualism of Australia, or the Manifest Destiny of America. Big egos — the Big I under a Big Sky — are built right into our language. Remember: you can spell solipsisme without je, but you can’t spell solipsism without I.


  1. Well... I think it's much more likely that I is capitalized because in miniscule it would be easy to overlook or mistake for a mimim (esp. since dotting i's, at least as a rule, is a modern convention), not because of egotism. Also, technically, as a diphthong the pronoun /ai/ consists of two morae (as do long vowels), and so is longer than the unstressed indefinite article (schwa).

  2. When I wrote that I was one of our shortest words, I meant only its length in letters, but you are quite right about the length of the sound.

    As to the other point, you may well be right, but I do not know. It’s true that the letter i was originally written without its characteristic tittle. (On the other hand, I don’t know if I’d call the addition of the tittle a “modern convention”, since I believe it goes back to the late Middle Ages.) It’s also true that the Old English pronoun was written lower-case, ic, as was the early Middle English ich. But, by the later Middle English period, we started seeing I, apparently always capitalized.

    So, this might corroborate your idea, that with the shortening of the word, upper-casing was needed to ensure that a short vertical stroke would not be missed; or it might speak to an emerging importance of the individual, and of self-identification. I’m not really arguing the latter (I know far too little to attempt it), but certainly today, there is no need to continue writing the pronoun upper-case (just as there is no need for Germans to capitalize all of their nouns). That we still do it may now speak more to self-importance than it once did, don’t you think? Of course, it might just be habit. :)

  3. The addition of the tittle per se is not a modern convention (in that it does occur in earlier times) but its use as a rule is (which is what I said).

    There are many conventions of English writing that aren't "necessary" (spaces between words, writing in a consistent direction, the use of miniscule, etc. etc. all were conventions adopted over time, not necessities), so I'd be reluctant to assign motive or value judgments to any convention on that basis.

  4. Fair enough. I missed your “as a rule” qualification by reading a little too hastily, and you make a good point about arbitrary conventions. My speculations are really nothing more than that. Well, that, and a little fun, I hope.

  5. Absolutely! Fun intended, fun had.

  6. Oh good! You had me worried there for a moment. :)

  7. Hahae! I see that you, like any non-Dutch, and like many modern Dutch labouring under the tyranny of Windows (one OS to rule them all ...), have fallen headlong over the 27th† letter of the Dutch alphabet. For ij (or ij in its proper form) is counted as a single letter.

    And that means that your hypothetical *Ij to educated Dutch eyes looks as uncouth as Uvashington to an American, or Vuestminster to a Brit. That is, it looks as weird as if someone tried to capitalise half a W. For in Dutch writing the development of ij as a doubling of i/j exactly parallels the development of w as a doubling of u/v.

    So, even if one writes it with two letters instead of a ligature, a stopgap measure that the Dutch have had to resort to since the domination of ASC-II*, ij can only be capitalised in its entirety, so the proper form is IJ, or IJ. Which, incidentally, does exist as a Dutch word: the name of a river - nowadays the harbour of Amsterdam.

    †Actually the 25th, Y being the 26th and Z the 27th.

    *Before that time, Dutch typewriters (I still learned to type on one) used to have a ij key - but only a lowercase one; the capital version still had to be written IJ, using two capital letters.

  8. Harm, that was a truly fascinating and informative comment — thank you! I am very happy to be corrected on the matter of the Dutch IJ (ij). Your examples (especially “Uvashington”) were well taken.

    You learn something every day, don’t you? At least, you do if you’re lucky,

  9. Pay no attention to that terminal comma in the last comment. It does not indicate some technical glitch where further comment was cut off. Nor is it a clumsy homage to e. e. cummings. It was just a typo. :)

  10. Harm J. Schelhaas9/09/2009 10:02 PM

    I've just found an article by another Dutchman on the internet about the problem of the Dutch 'IJ':

    There seems to be disagreement as to whether it constitutes an extra letter, or takes the place of Y ...

  11. Thanks for that, Harm. That was a fascinating article.

  12. Harm J. Schelhaas9/16/2009 4:05 PM

    You're welcome!