I teach an ESL course and last week a student asked me whether there were any "rules" in English for determining which syllable is stressed in English words. I went home and analyzed many 2 & 3 syllable words but did not find anything I could call a pattern. Can anyone provide me with some sort of answer?
Thanks.
1 2 3 4
I teach an ESL course and last week a student asked me whether there were any "rules" in English for ... syllable words but did not find anything I could call a pattern. Can anyone provideme with some sort of answer?

Yes, I have the answer. But I don't think you're going to like it...

Mark Barratt
Budapest
http://www.geocities.com/nyelvmark
I teach an ESL course and last week a student asked me whether there were any "rules" in English for ... words but did not find anything I could call a pattern. Can anyone provide me with some sort of answer?

By and large, words of Germanic origin stress the first syllable, and words of Latin origin stress a syllable that is reckoned from the tail of the word by some moderately complicated rules of classical Latin (often modified by a loss of the final syllable between Latin and English).
That distinction usually only shows up in words of three or more syllables; for shorter latinate words the stress is far enough from the tail to put it on the first syllable anyway, so you can't tell the difference.
And that's just the broad-brush treatment. Some Germanic prefixes don't get stressed, e.g. be- and un-, and some latinate words have had their stress shifted to the first syllable through a regularization process.
And then there are the Latin words with Germanic suffixes, such as "moderately complicated", or with Germanic prefixes and suffixes, such as "unfortunately".
Some illustrations from words in this message:
o Latin gerMANicus > English gerMANic. Latin's accent was never more than three syllables from the end, and here the stress is preserved on the MAN syllable even though a syllable disappeared off the end.

o Latin modeRAtus + Germanic -ly > English MODerately, using the Germanic first-syllable stress.
o Latin nominative disTINCtio, genitive distinctiOnis, > English disTINCtion, possibly preserving the nominative's accent on a form generally thought to be derived from the oblique forms, or perhaps a faux application of Latin rules to a truncated oblique form. (This is common in Lation "-tion" words: regulariZAtion, illusTRAtions.)
And of course we don't tolerate too many consecutive syllables without adding a secondary accent, as in "unfortunately" and "regularization".

And some function words don't ordinarily take any accent at all, such as "of", "the", "a", though they can be stressed for semantic contrasts.

Etc... It's no wonder that your students are confused.
Hopefully I have the Latin accents right; I'm not a native speaker.

Bobby Bryant
Austin, Texas
Teachers: We supply a list of EFL job vacancies
I teach an ESL course and last week a student asked me whether there were any "rules" in English for ... words but did not find anything I could call a pattern. Can anyone provide me with some sort of answer?

Stress is phonemic in English (meaning that differing stress alone can distinguish two words); thus it is not fully predictable.

Chomsky & Halle, in The Sound Pattern of English (1968), tried to show that stress is fully predictable from the morphophonemic structure of words at the cost of an extremely complicated scheme of differential morpheme boundaries.
But did they even consider syntactically determined stress assignment, as in Cornéll University vs. Córnell Avenue, where the stress retracts in the second example to avoid being adjacent to the stress on ávenue?
Peter T. Daniels (Email Removed)
I teach an ESL course and last week a student ... pattern. Can anyone provide me with some sort of answer?

I believe they did. IIRC their stock example was Tennessée (by itself) vs. Ténnessee Válley. But please don't make me go back and look it up (shudder).
Ross Clark
In sci.lang.translation benlizross (Email Removed) sanoi, hitaasti kuin hämähäkki:
Stress is phonemic in English (meaning that differing stress alone ... example to avoid being adjacent to the stress on ávenue?

I believe they did. IIRC their stock example was Tennessée (by itself) vs. Ténnessee Válley. But please don't make me go back and look it up (shudder).

Not that I would be surprised that Chomsky might be wrong about something, but at least I can say that this does not apply to proper names, which often have regional variations (Albany Georgia vs NY, for instance, or Lafayettes scattered across the US).
And did they take into account the more systematic variations between, say, London English and American English? Although I think perhaps regional dialects within both countries might swamp the national differences, in this as in much else.

rich

-to reply, it's hot not warm
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ \ Rich Hammett http://home.hiwaay.net/~rhammett / "Better the pride that resides in a citizen of the world; \ than the pride that divides
/ when a colorful rag is unfurled."
Site Hint: Check out our list of pronunciation videos.
I believe they did. IIRC their stock example was Tennessée ... don't make me go back and look it up (shudder).

You're not wanting me to look it up, are you?
Not that I would be surprised that Chomsky might be wrong about something, but at least I can say that this does not apply to proper names, which often have regional variations (Albany Georgia vs NY, for instance, or Lafayettes scattered across the US).

You're claiming that "Albany" has final stress in Georgia?

Do "Cornell" and "Tennessee" not look like proper names to you?
And did they take into account the more systematic variations between, say, London English and American English? Although I think perhaps regional dialects within both countries might swamp the national differences, in this as in much else.

They describe General American English.
Henry Lee Smith, in "The Concept of the Morphophone" in the Bloch Memorial Number of Language, suggested a way to capture facts about multiple dialects with a single formulation, but it was swamped out by the SPE frenzy a few months later.

Peter T. Daniels (Email Removed)
I teach an ESL course and last week a student asked me whether there were any "rules" in English for ... but did not find anything I could call a pattern. Can anyone provide me with some sort of answer? Thanks.

The following guidelines will get you a lot of the way, but there's a substantial minority of words they can't account for:
1. There are some comparatively rare endings which are always stressed:'-aire', '-ette'. In NAmE also '-et' (EI), but not in Britain
2. There are several common endings which always attract the stress tothe syllable preceding: '-tion', '-tious', '-tial' and '-ic' come to mind.
3. The suffix '-ly' does not move the stress when added.
4. Otherwise, stress is normally antepenultimate.
5. However, there are many (2 or 3 syllable) words in which the aboverules would indicate initial stress, but in fact the second syllable is stressed. There are only partial rules for this:
a. Germanic prefixes 'be-', 'for-', 'over-', 'under-', 'with-', 'un-' are almost never stressed.
b. In most words of two or three syllables where the first is not stressed, the first syllable is either a recognisable prefix (Germanic or Latinate) or homophonous with one. However, this does not get you far, since there are plenty of words with Latinate prefixes that are stressed, and I no of no rules to predict which.
c. In those words (all I believe with a Latinate prefix) which have different stress as a noun and a verb, the noun has initial stress and the verb non-initial.
6. Proper names are not generally excluded from these guidelines, but ina name of several words, the last is almost always stressed (even in cases where the last word is atypically a modifier: 'Llantwit Major', 'Weston-super-Mare', 'Frederic March III'. Interestingly, street names follow this rule unless the last component is 'Street', which is apparently felt not to be part of a name.
These are from my own observation. I'm sure somebody must have written them up more elegantly (and probably more correctly).

Colin
("Followup-To:" header set to alt.usage.english.)
6. Proper names are not generally excluded from these guidelines, but in a name of several words, the last is ... follow this rule unless the last component is 'Street', which is apparently felt not to be part of a name.

In AmE "Street" isn't the only street/road name final-generic-component that does not get primary stress. Another one is "Place". I can't think of any others, however.
"Broadway", by the way, is pronounced with "way" stressed by sufficiently elderly New Yorkers, at least with respect to the famous one in Manhattan (there's also one in Astoria, and there's an area of Flushing (and an LIRR station) called "Broadway"). Younger generations, as a rule, stress "Broad". We might be able to determine sci.lang's Dr. Daniels' age by finding out how he says it.
Students: Are you brave enough to let our tutors analyse your pronunciation?
Show more