Compounds: Comparing Hyphenated With Other Forms?

Question

Compounds can vary by:

• spanning multiple words without using a hyphen
• spanning multiple words and using a hyphen
• being only comprised of 1 word

Some compounds seem common in all forms. E.g.:

• pick up, pick-up, pickup

Some compounds seem common in only a couple forms. E.g.:

• freeze dry, freeze-dry
• rain fall, rainfall
• life-like, lifelike

Some compounds seem common in only 1 form. E.g.:

• swimming pool
• semi-automatic
• blackmail

I'd like to find out:

• What conditions incline compounds to vary in this regard?
• Under what conditions are they likely to transition from one dominant form to another over time?
• What implications or effect does one form have over another in writing?

Are there any research studies on this topic?

Answer 1

The only question I can answer is what conditions incline compounds to vary in this regard. It reflects the morphological type of the language, namely, isolating. Since affixes serve (despite many other functions) to delimit words, the fact that these languages lack affixes disturbs the system and calls for new means of marking it. It also causes context dependance to determine if the combination is of the form adjective+noun or noun+noun or even noun=noun+noun. In English the disctinction between the three is often invisible as in the pair "rain fall"-"rainfall": "rain" in the former may be intended to be either a noun or an adjective (like "rainy") - this process is called conversion, the latter is definitely a compound noun, whose formation came about in a deeper level of derivation and which is considered as a whole by default. Owing to the impoverished morphology, there is no clear formal means to show how a word or a compound is formed. The same holds true for the derivation of adjectives. Phrasal verbs are written separately for the principle to write separately whatever can be separated by insersion of something else, but when undergone convesion into nouns, they are written as one word. Here you see dependance on the context.

Answer 2

First, you have to be careful to not confuse compounds themselves with the orthography for compounds. What you're asking about is the orthographic variation in how compounds are represented.

As far as orthographic variation goes, there has been very little research done in this area, most likely because there's been very little reason to think to look into it up until the internet age as, before now, written communication was limited to very few domains and so was likely to be far more influenced by standardization attempts than domains like texting or tweeting are.

So, there have of course been many attempts in many societies to standardize spelling, and these almost certainly play a role in why compounds are represented in one way or another, but actual use most likely varies along social dimensions. For example, those living in one region might tend to spell compounds in one way while those living in another region might spell them a different way. You can see regional variation even in standardization attempts, whereas there are British standards (e.g. behaviour) that differ from American standards (e.g. behavior). Other examples might be people of different age groups spelling them different ways or people of different ethnic backgrounds.

A good correlate to look at to better understand why a compound would be spelled one way or another is studies on phonological variation in sociolinguistics, of which there is a massive literature starting from Labov (1966) and continuing to the present day. The same approach can be taken to orthographic variation, including how compounds are spelled, and there are some signs of that happening (see Tatman (2016) or Eisenstein (2015), for example), but as of yet, the best anyone would be able to tell you are the invented rules of some prescriptivist grammarians, which are really just suggestions, not explanations of why things are the way they are.

Eisenstein, J. (2015). Systematic patterning in phonologically-motivated orthographic variation. Journal of Sociolinguistics, 19(2), 161–188. https://doi.org/10.1111/josl.12119

Labov, W. (2006). The Social Stratification of English in New York City (2nd ed.). Cambridge, England: Cambridge University Press. (Original work published 1966)

Tatman, R. (2016). “I’m a spawts guay”: Comparing the Use of Sociophonetic Variables in Speech and Twitter. University of Pennsylvania Working Papers in Linguistics, 22(2). Retrieved from http://repository.upenn.edu/pwpl/vol22/iss2/18