The existence of the morphemes (AL), (IC), (LY) and (Y) is demonstrated

An assumption underlying this investigation is that our four suffixes are morphemic. This assumption is justified by the following demonstration:

Compare form formal

(FORM) (FORM) + (AL)

base basic

(BASE) (BASE) + (IC)

man manly

(MAN) (MAN) + (LY)

horse horsy

(HORSE) (HORSE) + (Y)

In these four examples the first word, being a free form, is thereby established as morphemic. It is then recognized in the bound form as the same morpheme and leaves the suffix as – at the least – a residual morpheme.

This method of demonstrating the morphemic status of our suffixes is not as simple as it looks and it will receive further comment later. But as we shall not have cause to quarrel with these particular conclusions we can accept them for the time being and go on to gather our stock material.

The difficulty of identifying -al, -ic, -ly, and -y as instances of suffix morphemes (AL), (IC), (LY) and (Y)

Reliable stock material implies a list of genuine occurrences of our suffix morphemes. This requires that we have means of telling whether or not an instance of final -al, -ic, -ly or -y in a word of adjective function is identical with the morphemes (AL), (IC), (LY) or (Y) respectively. Our means for doing this are unfortunately rather inadequate.

The difficulty is that we are trying to identify units which are both short (thus having small phonemic and graphemic distinctiveness) and, most often, semantically vague. It is their deficiencies in these respects which make them difficult to identify with any confidence.

Two simple examples may illustrate this:

antic (adjective)

Can we really identify final -ic here as our suffix morpheme (IC)? They are phonemically, graphemically and syntactically identical, but it would be presumptuous to deduce from this and identify it as an instance of (IC). Our intuition tells us that this may not be a case of a suffix at all (what, then, is the stem?). Intuition is “evidence” which should be called on as seldom as possible of course, yet the weak distinctiveness of these suffixes seems likely to involve us in many such cases.

Or take basic. To identify -ic here with (IC) would of course be right, but by good luck rather than by good logic. Logically basic may consist of a stem /beys/ + /ik/ or /bey/ + /sick/ or even /beysi/ + /k/ and accordingly to recognize (IC) here is really an example of wishful thinking. If one takes base into account the picture is quite different, but this only confirms the inadequacy of attempts to identify -ic as (IC) without recourse to external evidence, i.e. to the morphemic status of the rest of the word. Identification of such distinctively weak units falls down if it is not supported by other means.

We could of course propose that every occurrence of final -al, -ic, -ly and -y in adjectives be taken as an instance of one of our suffix morphemes. There is, however, one good reason for rejecting this suggestion outright, and that is that many such occurrences are certainly not instances of our suffix morphemes; we have only to think of the common adjective suffixes (ARY), (ORY) (in e.g. military, accessory) to realize how false a picture we should obtain of our suffix (Y) if these were included in our statistics.

The first task remains: we must devise some reliable means for identifying these indistinctive units with the morphemes (AL), (IC), (LY) and (Y).

Date: 2016-04-22; view: 737