Morphology of words

[clara]

Hi all,

I wondered if we could discuss the preposition 'during' and if this  is made up of two morphemes.

I understand that the inflectional affix 'ing' marks the present participle form, and that 'dur' is used to mark some sort of time frame. My reasoning is that 'dur' can occur in other combinations of morphemes without changing its basic meaning, such as 'duration', 'endure', etc. Equally, 'ing' is a bound morpheme that can't stand alone and is also used in other combinations to make the present participle.

What do you think of my reasoning here?

[vox]

I agree with your reasoning but ‘dure’ is not a free morpheme in english, isn’t it ? It’s supposed to be a bound morpheme too. Yet dictionaries say that ‘during’ comes from an obsolete verb ‘dure’ which actually was a loan word from latin via old french ‘durer’ (to last). By the way in contemporary french ‘durant’ is the present participle of ‘durer’ as well as a preposition. In some ways it calls into question the synchronic/diachronic distinction. It’s well-known that lexicon is the area where that distinction is the most problematical.

I remember two old papers written by R.Jackendoff and M.Aronoff. Aronoff in particular wondered if forms like -sist (assist, insist, persist, consist...), -fer (prefer, confer, infer, transfer...), -mit (permit, commit, admit...), per- (persist, permit, perceive...), con- (confer, consist, consume, conceive...) are morphemes in english. All are latin verbs. He concluded that they are meaningless morphemes (which he ended up calling ‘morphomes’ few years later if I’m not mistaken, I don’t remember well). I think it’s indirectly related to your issue since I think we can find one arguing that these forms have or had a constant meaning.

[Daniel]

Vox answered this well.

The main distinction is between etymology and active morphology in (someone's current mental) grammar. Etymologically yes, that's -ing, but it can't be easily analyzed that way in modern English grammar.

There's also the question of productivity, with other verb+ing combinations like "considering", or other forms like "given", that act like prepositions. But because that morphological process is not productive for all/most/many verbs, these are also, in a sense, frozen forms, rather than an active morphological process today. So it is also a question of etymology there, for the most part. Once in a while the same etymological process might happen for a new word though, so it might be just a little productive.

[vox]

Yes, if we analyze ‘-ing’ as the present participle marker the analysis must necessarily correspond with a productive word-formation process. It’s not the case here. So synchronically it’s better to consider it as a frozen form.
Diachronically, it’s a present participle converted into (or employed as) a preposition. Then the formation process is not derivationnal or inflectional but grammatical. Conversion is a common word-formation process but this one in particular is not productive since participles converted into prepositions are very rare. It’s a case of grammaticalization, which is mainly a diachronic process.

By the way, I’ve never understood what do morphologists exactly mean by ‘productivity’. Is it availability in creating new words or a high frequency rate in attested words ? I’m not an english native speaker but it seems that suffix -ity is quite frequent in nouns derived from adjectives, but -ness is much more available to create new nouns from adjectives.

[Daniel]

Then the formation process is not derivationnal or inflectional but grammatical. Conversion is a common word-formation process but this one in particular is not productive since participles converted into prepositions are very rare. It’s a case of grammaticalization, which is mainly a diachronic process.

That's a good point. It probably is grammaticalization rather than morphological for the suffix. The etymological explanation conflates that.

By the way, I’ve never understood what do morphologists exactly mean by ‘productivity’. Is it availability in creating new words or a high frequency rate in attested words ? I’m not an english native speaker but it seems that suffix -ity is quite frequent in nouns derived from adjectives, but -ness is much more available to create new nouns from adjectives.

I don't think it's very well defined either. It's sort of a combination of defaultness, frequency, and being grammatically active. Most important, a productive affix is grammatically active-- one that can still apply to form new words. But it sort of brings along the other properties with it-- usually it's the default suffix for that property, and it's also usually relatively frequent. And often we see frequency being used as a substitute for productivity, but what we really need to see is it being (frequently?) used with new words, or even in an experiment rather than in observational data. But that's obviously not easy data to acquire.

Edit:
This isn't too bad, but also not too clear: https://en.wikipedia.org/wiki/Productivity_(linguistics)

The opposite of productivity is certainly obvious enough: it's when a morpheme/process is no longer used to form new words. So, productivity is when it's still used, but how do we (easily) measure that?

Lieber's Introducing Morphology textbook has some good information in it. I used that last year to teach about this (and other morphology topics). I could get some of my notes for that and write more if you want to know more about it. It's an interesting topic. Let me know if I've answered your question enough, or if you want more notes about that.

[vox]

Most important, a productive affix is grammatically active-- one that can still apply to form new words. But it sort of brings along the other properties with it-- usually it's the default suffix for that property, and it's also usually relatively frequent.

That makes sense but what do you mean by ‘default’ ?
If there’s only one existing affix to form the word one needs to express (like -ism for nouns of ideologies) then it’s an affix by default. But what if there are several rival affixes to form the same meaning ? How do we know that -ness is the default affix to form nouns of property derived from adjectives, by contrast with -ity and -hood ? (I thought -hood was only for nouns derived from nouns but I found ‘kindhood’ in online dictionaries.)

I think we know that defaultness only a posteriori, after having counted observational data. That means people obey to a paradigmatic pressure which constrains their lexical creations to conform morphologically with the paradigmatic range. So I think it’s likely the ADJ-ness paradigm applies a stronger pressure on speakers than the other semantically equivalent paradigms. Maybe the ADJ-ness paradigm contains some sort of leader words which have given to it a high pressure capacity on people’s mind when they coin new words. Thus -ness would be the default affix originally because of (or thanks to) certain words.

[Daniel]

On the one hand, that is a good question and I agree with your objections.

On the other, there are some ways to reasonably approach this:
1. Yes, frequency of new productive usage.
2. The less restricted affix. Plurals in English can be formed via -s or -i, but -i is very restricted in forming new words, even if it is once in a while productive. It doesn't seem controversial to me to say that -s is the default, although perhaps 'most general' (rather than special case usage only) would be a better term. Technically -i is default in exactly those cases where it fits best, leaving -s as the 'elsewhere' type. So rather than 'default' we might say 'general' or 'elsewhere'.
3. Blocking appears to be a real thing in language (*stealer, because of thief). So if two possible suffixes could be used, probably only one becomes the norm. So which is it?
4. Speakers have an amazing ability to track frequencies of usage in addition to just grammatical rules. Indisputably this is part of our 'competence' as well even though it may be irrelevant to (or not considered in) some theories. Zeldes (2012) Productivity in argument selection from morphology to syntax (De Gruyter Mouton) is a good book about some of those effects, specifically about how speakers appear to know and use collocational frequencies.