Syllables written versus spoken

My question can best be illustrated by an example. In the dictionaries I’ve consulted, the word “decreasing” is given a pronunciation dee kree sing, however, the syllable breakdown is de creas ing. The only reason I can think of needing to know that latter is so i can hyphenate the word, if necessary.

I realize that decrease is the root word and so “obviously” the s belongs with the crea in some sense. However, syllables would seem to be a phenomenon of spoken language rather than written language and I’d think the spoken structure should be the determining factor. Is there any reason for breaking the syllables as is done other than “Well dammit decrease is the root word.”

What you describe is an example of a well-known linguistic phenomenon called “ambisyllabicity,” in which a consonant appears to belong to two syllables. Ask a native speaker which syllable the consonant belongs to, and they will be as confused as you are. For example, what’s the first syllable of “lemon”? Most people will say, “lem.” What’s the last syllable of “lemon”? Most people will answer “mon.” But, there’s only one m. Which syllable does it belong to? Uh…

In short, use your best judgement.

You’re correct that dictionaries show the orthographic syllable breaks for hyphenation. Words are generally hyphenated at positions such that one or both of the parts don’t end up confusing the reader. For example, “material” would usually be hyphenated as “mater-ial” rather than “mate-rial”, even though in pronunciation some people might consider the /r/ to belong to the third syllable. Less confusing syllable breaks tend to be those located at morpheme boundaries (such as between a stem and a suffix, as in your “decreasing” example).

Of course, there are many cases where multiple non-confusing breakpoints are possible. Where to put them is really up to whoever writes the dictionary. Regional varieties of English tend to have their own conventions; good typesetting software will come with different hyphenation patterns for British and American English.

We learned it that way in junior high, and I always thought it was totally arbitrary. Or I am not most people. I’d say “le - mon”.

English hyphenation is complicated, but one invariable rule is that the beginning part should not mislead the reader as to pronunciation. Let me give a couple examples. I was astonished to find equiv-alent in a book (of which I was a coauthor). The word is obviously made up from equi+valent, but you do not expect equiv from equi. (Note that a chemist would want it hyphenated equi-valent, because it is a different word with a different pronunciation.) My software (tex) will not hyphenate record, because the noun should be rec-ord while the verb should be re-cord. Tex will allow you to insert a conditional hyphen in such cases.

In French the hyphenation rules are based strictly on the orthography. For example: si-te although it is spoken with only one syllable. And in German, the work backen is broken as bak-ken. Presumably, bac would be pronounced as bats, so German must follow a practice similar to English. Tex creates hyphenation rules peculiar to each language, based on an analysis of a dictionary.

German hasn’t allowed hyphenation for <ck> since the Rechtschreibreform of 1996. The current rules for hyphenation can be found in Section F of the Deutsche Rechtschreibung Regeln; I’m sure TeX distributions were quick to adopt them.

Actually, I think the root word is the ‘crease’ part (from Latin* crēscere*: to grow), with the prefixes de- or in- added to it.

Not in English it isn’t; there is no English word “crease” which means “grow”. You could analyze “de-” as a prefix of “decreasing” only if you mean the word in the sense of “removing creases (i.e., folds)”.

I’m working from memory here, but thinking back to my undergraduate linguistics days…

A syllable in English at least is broken down into components: the nucleus (which must be a vowel [okay, it can be a nasal when spoken] or a diphthong and is ALWAYS present) and, possibly, an onset and/or a coda.

It’s hard to do this in just a type-box, but “decreasing”

[de] [creas] [ing]

[de] = nucleus of “e,” onset of “d”

[creas] = nucleus of “ea,” onset of “cr,” coda of “s”

[ing] = nucleus of “i,” coda of “ng”

Depending on the style of linguistics, it can be diagrammed into a syllable as consisting of Onset + Rhyme (where “rhyme” just means the rest of the syllable after the onset; sometimes the rhyme will be nucleus + coda, sometimes just nucleus.) I learned it in a slightly different, “bar” schema, where you sort of label onset twice–once at the same level as the nucleus and then once again up a level with the rhyme.

Anyway, if you go to this webpage and go about 3/4 of the way down, you can see the diagram better than my typing it out:

Hell, that page probably explains all of this better than I have, anyway.