Talk:Three-letter rule

Rule or rule of thumb?

Contributors calling this a "rule" have no concept of the difference betwen a rule and a rule-of-thumb. --Wetman 06:17, 3 Mar 2005 (UTC)

As the sole contributor to date, I can assure you I did not coin either of the names for the...er...phenomenon. I'm not sure, but I think you might have to blame Otto Jespersen. Joestynes 07:01, 3 Mar 2005 (UTC)

Exceptions overwhelm the "rule"

This "rule" is not a particularly astute observation either:

a	article	F
I	pronoun	F?
o	interjecction?	F?
ab	noun (abdominal muscle)	C
ad	noun	C
ah	interjection	F?
Al	proper name	C
am	auxiliary verb	F?
an	article	F
as	preposition, conjunction	F
at	preposition	F
ax	noun	C
be	auxiliary verb	F?
bi	adjective (bisexual)	C
Bo	proper name	C
by	preposition	F
Cy	proper name	C
Di	proper name	C
do	verb, auxiliary verb, solfege	F??
Ed	proper name	C
eh	interjection	F?
em	noun (unit of width)	C
en	noun (unit of width)	C
er	interjection	F?
et	conjunction (loan from Latin)	F
fa	solfege	C?
go	verb, noun (loan from Japanese)	C
ha	interjection	F?
he	pronoun, interjection	F?
hi	greeting	F??
hm	interjection	F?
ho	interjection, noun	F?, C
id	noun (loan from Latin)	C
if	conjunction	F
in	preposition	F
Io	proper name	C
is	auxiliary verb	F?
it	pronoun	F?
Jo	proper name	C
la	solfege	C
Li	proper name	C
lo	interjection	F?
ma	noun	C
me	pronoun	F?
mi	solfege	C
Mo	proper name	C
mu	noun (Greek letter)	C
my	determiner	F?
no	interjection?, determiner	F?
nu	noun (Greek letter)	C
of	preposition	F
oh	interjection	F?
om	interjection	F?
on	preposition	F
or	conjunction	F
ox	noun	C
Oz	proper name	C
pi	noun (Greek letter)	C
Ra	proper name	C
re	noun/preposition (loan from Latin), solfege	F??, C
so	conjunction, adverb	F, C?
ti	solfege	C
to	preposition, particle	F
uh	interjection	F?
um	interjection	F?
up	preposition, adverb, noun	F, C
us	pronoun	F?
we	pronoun	F?
xi	noun (Greek letter)	C
ya	pronoun	F?
ye	pronoun, determiner	F?
yo	interjection, greeting	F??

That's not looking good, no matter where you draw the distinction between function words and content words. --MarkSweep 01:17, 4 Mar 2005 (UTC)

Well, it depends on whether you're measuring number of words or frequency of words. The average text contains more me's and no's than mi's and nu's. Also, many of the 2-letter words are recent imports or coinages. In Jespersen's time, abbreviations tended to have a full stop, and loanwords and jargon were written in italics or quotes. Nowadays, we're more relaxed about this, and also about hypocorisms. Perhaps the rule will eventually be regarded as part of philology. Nevertheless, for the moment, it's in the literature, whatever one may think of it. BTW "rule" is a standard term in descriptive linguistics; cf Meeussen's rule. Or for even geater shock-value, check out Verner's law, Grimm's law, Heaps' law, Grassmann's law. Joestynes 03:27, 4 Mar 2005 (UTC)

Sure, if you're listing words by frequencey, of course you'll get the function words near the top. But as I read it, the rule seems to be about type frequency (i.e. how many two-letter words occur in a dictionary), not token frequency (how often they occur in running text). Regarding Jespersen, should there be a remark in the article about him? Is Jespersen the ultimate source of this rule? Re "rule", I have nothing against this piece of jargon, it's just that the earlier version talked about this rule being "enforced", which made it seem like a positive law that someone was enforcing, rather than an empirical observation that is to some extent supported (did you mean "reinforced"?) by data. --MarkSweep 06:05, 4 Mar 2005 (UTC)

It's not simply a random coincidence producing a marginal mnemonic. There were processes at work in the 15th-18th centuries to encourage the distinction, which explain why "egg" isn't written "eg". Most exceptions postdate that time. I'll try to add some more detail when I can check it up (I'm not certain it was Jespersen, for one thing; if it was, it was far from his most significant observation). Joestynes 08:32, 4 Mar 2005 (UTC)

The current version is much improved, especially regarding the historical aspects. The rule makes sense as a(n) historical principle used at a time when English orthography was still in flux. --MarkSweep 19:14, 8 Mar 2005 (UTC)

Don't forget the many function words with 3+ letters. In addition, there are so little references (and the rule has such little grounding) that one wonders if the rule isn't just something someone made up once. Suppose this rule does exist, how different would we expect English to look without it? I think the differences would be absolutely minimal and insignificant if you take into account that we probably couldn't help imagining the existence of the opposite rule instead. — Preceding unsigned comment added by 82.139.90.173 (talk) 04:22, 24 February 2012 (UTC)

Paradise Lost example

Huh? It looks to me as if both occurrences of "he(e)" appear on the ictus. —Blotwell 14:55, 18 March 2006 (UTC)

examples of "augmentation"

since English spelling is very conservative, such examples would need to concern "content words" that did not use to be polysyllabic. These will be rather hard to find. Methinks the 'rule' would be better described by saying that some very frequent 'function words' had their spelling reduced to two letters, be < beon, if < gif, do < don, and some had less than three letters already in OE, such as ic, to, he, me. It is pointless to argue that "content words" could be spelled with fewer letters if they never were (queue, night, sleigh). Most homophones aren't homographs not because of some orthographical decision, but because they haven't been homophones for more than 300 years or so. dab (𒁳) 19:50, 16 March 2007 (UTC)

I take your point. However, when constructing the list of examples I specifically checked in the OED to pick words where a 2-letter spelling is attested. If a Middle English or Early modern English spelling without the "redundant" third letter is attested, then it gives credence to the three-letter rule, regardless of how the word was pronounced or spelled further back in the history of English. I don't personally discount your theory, but in keeping with Wikipedia ban on Original Research, you need independent attestation of your critique. I have reverted your edits acordingly. jnestorius^(talk) 14:58, 17 March 2007 (UTC)

Aren't verbs "content words"?

Particularly, in this case, "do", "go" and "be"? Matt 23:31, 29 March 2007 (UTC).

Auxiliary verbs are function words. "do" , "be", and "have" can serve as either auxiliary verb or main verb. "Go" is not really an auxiliary and is listed as an exception to the Three letter rule. jnestorius^(talk) 00:04, 30 March 2007 (UTC)

OK, so when "do" and "be" are used as main verbs, rather than auxiliary verbs, they are exceptions and should be listed as such? And what about words like "am" and "I"? In the sentence "I am alive" these words seem pretty contentful to me. Matt 00:28, 30 March 2007 (UTC).

The spellings reflect pronunciation rather than grammatical role. A weak form of pronunciation encourages a short spelling. Function words tend to have weak forms, but when "do" or "be" are used as main verbs, they are usually still unstressed ("I did the crossword"; "This is fun") There aren't separate spellings for function and content senses of the same word. Though there are for Contractions like isn't.

Regarding what counts as a function word: if the explanation at function word is inadequate, the question belongs at Talk:function word; failing that Wikipedia:Reference desk/Language. jnestorius^(talk) 01:18, 30 March 2007 (UTC).

That article seems reasonably clear, saying that content words "include nouns, verbs, adjectives, and most adverbs" and "function words may be prepositions, pronouns, auxiliary verbs, conjunctions, grammatical articles or particles". So if "be" and "do" can act as main verbs (as you seem to agree) then they are, in that role, content words, and should be listed here as exceptions just as "go" is. By the same token, "am", being an inflection of "be", is also content word and should be listed as an exception. Matt 01:38, 30 March 2007 (UTC).

Okay, I've noted it. I'm wary of making too much of it; I think it's pretty pedantic and not central to the point of the concept. jnestorius^(talk) 23:05, 30 March 2007 (UTC)