Types of "stress" in language

Question

Wondering what are all the different kinds of "stress" (so to speak) in any language. I just found out about Prosodic Stress which is pretty cool.

I didn't take the test yesterday. (Somebody else did.)

That's cool because it totally changes how the sentence works, just by the way you stress the different words. That's really different from the acute accent like used in Spanish película. But that type of stress makes sense too. In this case it changes the meaning of the word, but we use this all the time in English it seems to just help with the understanding of words (disambiguating and other things I guess).

Then there is Secondary Stress which I don't know much about. Glancing here too, I don't get it yet. But it is said:

Most languages have at most one degree of stress on the phonemic level. That is, each syllable has stress or it does not.

However, I can think of it where you have (EXTRA FORCE) stress (acute accent type stress), and (sh) stress, or quieting stress / de-stress, relative to some normal level, sort of like low vs. high vs. normal tones. Wondering if anything like that exists in any language.

Then there is the exclamation mark ! for kind of adding some sort of stress-ish thing.

The question mark ? seems like it could be called a type of stress as well.

Then there is print stress, like bold and italic. The italic kind of accomplishes prosodic stress.

Then there is quote " stress, in which (in addition to meaning the grouping of quoted content) you might change your tone of voice to reflect a character of some sort.

So that means there are potentially these kinds of stress at least:

1. Lexical stress (acute accent).
2. Prosodic stress.
3. Secondary stress.
4. De-stressing perhaps.
5. Exclamation stress.
6. Question stress.
7. Quote stress.
8. Italic stress.
9. Bold stress.

The questions are the following:

1. If all of these things could be considered "stress" (or what a better name would be for it).
2. If there are any others across any other language that are important.
3. If there are any orthographies for any other types of stress (perhaps like if there is a "mood" stress in some language or something).

Answer 1

There are two main types of "stress". The first is what we still call stress, which refers to word stress, sometimes called lexical stress. The position of the stress may be completely predictable by surface-oriented rule, or may involve a mix of rules and lexical specifications such as in English "callow" vs. "allow". Within the set of such stresses, language may distinguish between primary and secondary stress (as well as lack of stress). In Odawa and Finnish for example one syllable in the word (the first) has a higher degree of phonetic prominence and other syllables are either unstressed or secondarily stressed. The pattern in Finnish, for instance, is ˈcvcvˌcvcvˌcvcvˌcvcv... There are some languages such as Waorani said to have multiple (alternating) stresses where all stresses are phonetically equivalent. In English, we have differences like the verb proˈgress, the noun ˈproˌgress, and the noun ˈtigrĕss illustrating the stress patterns "none+main", "main+secondary" and "main+none".

At the phonological level, it appears that languages can be described in terms of saying (a) which syllables are stressed and (b) which if any of the stressed syllable has the highest prominence. It is unclear to what degree (b) is actually part of the phonology of a language. Of course, people have historically posited much richer kinds of distinctions of stress, but these distinctions have not proven to be phonologically necessary or even experimentally validatable.

The other half of the "stress" picture arises when you combine multiple words into larger syntactic groups, and this gives rise to infamous differences such as between "blackboard eraser" (used to erase blackboards), "black board eraser" meaning a board-eraser which is black. While this was historically treated as being the same substance as word-stress, giving rise to an infinity of stress integers, the "received opinion" is that in fact this is the result of some kind of phonetic overlay on top of basic lexical stressed. The typical account of this overlay of intonation is that there are quasi-tonal elements such as "H", "L", "M" which are somehow lined up with syllables. The beginning of this kind of analysis was Mark Liberman's 1977 dissertation The Intonational System of English. Since then, myriad sub-theories and notations have emerged, so that you may find thing like "+H*". Such systems have been worked out that handle a fair amount of the intonational realization of English and Dutch, and the notional is constantly being applied to new languages. There has recently been some interest in enriching the system to account for the fact that pitch-modulations are not the only components of intonation, so that lengthening or shortening, breathiness or creakiness can also play a role in intonational systems.

There are two kinds of word-stress: primary and secondary. We have little idea how many kinds of intonational combinations there are. The problem with intonation has been identifying a non-phonetic correlate of a supposed intonational property. There are infinitely many ways to modify a given phonetic form by dynamically modulating pitch, messing with phonatory properties, changing duration patterns, fiddling with amplitude... what we need is a system for distinguishing various "kinds" of implementations of "then ˈI saw ˌher", and saying that "Tokens A and B are used the same way", "Tokens A and C are used different ways", since the premise is that we are looking for categories, not continuous phonetic measurements (and what kind of thing does a "category" represent?). Very roughly, intonations categorize some kind of semantic-pragmatic information, such as "contrast" or "focus". Creaky voice is used dialectally to convey disdain, etc. Progress is slowed by the inherent difficulty of mastering both the phonetics of disdain and the pragmatics of disdain.