3

-al can be added to a word to form a an adjective or a noun. If the adjective-forming suffix is added after a base word that ends in [l] then it can change to an -ar suffix (e.g. polar).

But are there any examples of this, or another sound change, with the noun-forming suffix?

Danger Fourpence
  • 547

3 Answers3

2

I think your underlying premise, that -ar is a variant of -al as far as English is concerned, is not completely accurate or how these suffixes are perceived in presentday English.

In previous stages of Romance languages, speakers may have perceived the two as variants of one another, but I'm not sure this is the case today. The suffix -al is used either to form an adjective from a noun or to form a noun from a verb, whereas -ar primarily is an ending on adjectives, without the corresponding noun necessarily existing in the first place.

Interesting cases to consider include "peculiar", "lunar", "nuclear", "vulgar"...

Neil Coffey
  • 19,622
  • Lunar is 'pertaining to Luna, the moon goddess', and is used, I always understood, to avoid confusion with moony having a different meaning. Nuclear is 'pertaining to the atomic nucleus'; vulgus meaning the populace at large (as the root of vulgar) is in Chambers, never mind the OED; and peculiar has been a noun as long as it has an adjective. Your argument is an interesting one, but I think you might need better examples. – Tim Lymington May 03 '12 at 18:24
  • 1
    No. The words "lunar", "peculiar", and "vulgar" were borrowed as adjectives. The pattern N+al doesn't apply here. – Alex B. May 03 '12 at 20:11
  • These were borrowed as adjectives (sort of, probably). But that's kind of my point: rather than -ar being a productive suffix per se in English, probably most of the adjectives ending in -ar were borrowed from French, or at least created under influence from French. It's hard to be sure, but the motivation for saying that -ar is a variant of -al, at least natively in English, is far from strong. – Neil Coffey May 03 '12 at 20:41
  • Though I should also say that to a large extent, what matters is how speakers perceive of the form. Do speakers intuitively think of "-ar" as being a variant of "-al" (as they might, for example, think of the plural "-es" ending as being a variant of "-s"). I suspect not. And I suspect that for the vast majority of speakers, "vulgar" is not derived from another word (even though historically that may be how it came to be) but simply now an adjective in its own right. – Neil Coffey May 03 '12 at 20:55
  • The premise is correct: many -ar adjectives have matching nouns with a root ending in /-l/: 'circle'~'circular', 'triangle'~'triangular', 'pole'~'polar', 'cochlea'~'cochlear', 'curriculum'~'curricular', 'spectacle'~'spectacular', 'module'~'modular', 'muscle'~'muscular', 'testicle'~'testicular', 'title'~'titular', 'globule'~'globular', 'granule'~'granular', 'molecule'~'molecular'. – nohat May 03 '12 at 22:14
  • @nohat - yes, but do speakers perceive this -ar as being a variant of -al? – Neil Coffey May 03 '12 at 23:02
  • Perceptions are very subjective and can't be used as a basis for linguistic analysis. The other day someone here "perceived" of the prefix "unac" in unacceptable. – Alex B. May 03 '12 at 23:05
  • @Neil I think that people who know many Latin roots might perceive this. It is a pretty subtle thing, but my hypothesis is that speakers would reject made up adjectives with -ar if the root doesn't end in -l, and vice-versa. – nohat May 03 '12 at 23:09
  • Alex - it can if used appropriately. Obviously I'm indicating a hunch here which to be scientifically valid would have to be subject to a broader survey. @nohat - I suspect that speakers are aware of the distribution effect as you suggest, with a few "corner" cases: because of e.g. "lunar", I would hypothesise that speakers' judgements of other formas ending in -nar may vary more. But... – Neil Coffey May 04 '12 at 15:32
  • ...you then come back to the issue of whether that distribution effect actually makes -ar a variant of -al. It's a similar situation to e.g. the phonemes /h/ and /ŋ/ in English which if it was purely down to distribution would probably be considered variants of one another, but it's very questionable whether speakers consider instances of these phonemes as being manifestations of 'the same sound'. – Neil Coffey May 04 '12 at 15:38
  • @NeilCoffey, but "-ar" *is" an allomorph of the suffix "-al". At least, that's what most linguists think. – Alex B. May 08 '12 at 23:56
  • Alex -- and do you buy the arguments for that analysis? – Neil Coffey May 09 '12 at 03:54
  • Sorry, that was a bit glib, so just to expand: I wonder if it's actually true that "most linguists" make this categorisation without qualification (e.g. if you see them as "allomorphs" purely as a statistical, distributional phenomenon, then this seems to have more grounds; saying that "-ar" is perceived as a 'version' of "-al" is a less convincing argument, I think, weighing up the evidence). However, in any case, it doesn't per se matter what "most linguists" think: what's more important is to weigh up the available evidence and reach a reasoned conclusion based on this evidence. – Neil Coffey May 09 '12 at 04:00
1

There seem to be a number of nouns ending in the suffix -ar. Apparently even in Latin, there were nouns ending in -⁠āre or -⁠ar which were derived from the neuter declension of Latin adjectives in -⁠āris. Some of the following words come from Latin nouns like this:

  • collar, from Latin collare, from collum "neck"
  • scapular, from medieval Latin scapulare, from scapula "shoulder"
  • altar, from Latin altare, which the Oxford English Dictionary says is from adolere (I don't know how the "t" got in there)
  • exemplar
  • cellar

Not sure if the following words count, since they're obviously derived from earlier adjectives, but:

However, I was not able to find any examples of a native English word that can be suffixed with -ar (there are examples of this with -al, such as betroth + -al = betrothal).

herisson
  • 81,803
0

In its entry for the noun-forming suffix -al, the OED gives no examples nor makes any mention of any sound changes. The only slight variation is in spelling as with battle. The answer, then, would appear to be no.

Brett Reynolds
  • 9,274