Learn English – Why is it “geometric” but “theoretical”

etymologyic-icalmeaningmorphologysuffixes

I just came across a course name: Geometric and Theoretical Optics. The mismatched endings bug me. Why do we have both -ical and -ic endings?

Is there any difference in meaning between, say, theoretic and theoretical? I know that acoustic and acoustical mean slightly different things, but is this difference inherent in the words, or is it just usage — we happen to use the former for sound, and the latter for ceilings?

Best Answer

Origin

The suffix -ic comes from Greek -ikos, while -ical is a combination of -ic and the French suffix -al. Originally, -al was suffixed to scientific nouns ending in -ics, e.g. mathematics - mathematical. Eventually, the -ical portion of those words was reanalyzed as being a single unit. This is what Marchand (1969) had to say about -ic vs. -ical at this time in their history:

There was, at the beginning, indiscriminate coexistence of two synonymous adjectives. But language does not like to have two words for one and the same notion, and competition was bound to come.

What Marchand is talking about is that these two redundant suffixes eventually settled into certain niches where they exist today (more on this later).

-ic and -ical doublets

There exist many -ic/-ical "doublets" in English, where both forms of the word exist and are used regularly.

For example:

  • historic and historical
  • electric and electrical

The words in common doublets have generally developed distinct meanings (or domains of use). For example, historical refers exclusively to things that happened in the past, while historic can refer to an event occurring at that very moment that will have a profound impact on history.

However, the differences in meanings between -ic and -ical words (in cases where both forms are used) cannot be generalized to a specific unique meaning that -ic or -ical itself contributes to the word.

That is to say, there is no general and consistent meaning that -ical uniquely contributes to historical, electrical, etc. Instead, each word as a whole just settles into its specific meaning.

Distribution Today

According to a systematic productivity measure by Lindsay and Aronoff (2010), the more productive of the two suffixes is clearly -ic, which is favored by approximately an 8 to 1 ratio using this measure. So, you are much more likely to have an -ic form of a word, or to have that form be the more commonly used form.

Now, if -ic is clearly used more than -ical, then why do we have both of these suffixes in the language?

Well, -ical has managed to differentiate itself from -ic, but not by contributing a specific meaning to the words it attaches to. Instead, -ical has found a specific morphological domain in which it attaches very productively: stems that end with the morpheme -(o)log-. Examples:

  • biological preferred over biologic
  • technological preferred over technologic
  • typological preferred over typologic

There are nearly 500 stems in Webster's 2nd dictionary that end in -olog. Using the same measure of productivity, Lindsay & Aronoff found that within the domain of these stems, the -ical suffix was itself preferred by an 8 to 1 ratio over -ic. That is to say, when there is an -(o)log morpheme at the end of the word, then people heavily favor -ical.

The notion of one suffix being productive when attaching to a certain suffix is known as "potentiation", coined by Williams (1981). (Another example of potentiation is what we see with words ending in -able. In general, -ness is considered a more productive suffix than -ity, but in words ending in -able, -ity is strongly preferred.)

Bear in mind that these are tendencies; while there is a strong preference for one form over another in certain cases, that does not mean that it is a perfect split.

Summary

-ic and -ical do not differentiate themselves semantically; however, both forms of a word can exist if each one settles into a different meaning, or otherwise has a different domain of usage (for example, one formal and the other informal). The actual differences in meaning between two forms in a doublet are essentially arbitrary.

While -ic is much more common overall (and could be argued to be the "default"), -ical is strongly preferred in stems ending in -(o)log-. Again, this is a tendency that began somewhat arbitrarily during the development of the language, but because words ending in -olog form such a large and relatively uniform group, a subpattern was able to emerge and stabilize, even as -ical remains in the minority.