Considering the -able and -ab(u)le Suffixes

I am currently thinking about the -able and -ab(u)le suffixes. I have previously studied both but want to refresh my knowledge. Specifically, I want to answer the question: “Why are -able and -ab(u)le two different suffixes?”

First, -able suffix is not etymologically related to the word able. The base of the word able is <Able>, which denotes “having power or means, capable, can, permitted to, not prohibited from, qualified to.” The word able comes from Middle English able from Old Northern French able from Old French able, hable from abile, habile from Latin habilis from habēre + -bilis. Notice that the Latin started with the letter <h>, which spelled [h]. Due to regular sound changes as the word passed through French, the initial [h] became zeroed and the spelling changed, eliminating the <h>, to reflect the pronunciation.

The word ability similarly comes from Middle English abilite from Old French ableté from Latin habilitās from habilis +‎ –tās from habēre + –bilis. The same sound and spelling changes occurred in French as with able, resulting in the loss of the <h> from Latin habilitās to English ability. The Latin –tās developed into the English -ty. The preceding is the connecting vowel -i-. Because the changes from habilis to habilitās occurred in Latin and English borrowed ability through French from Latin, the base of the word ability is <Ability>.

(As a side note, <Able> is not further analyzable as *<Abe + ile> because the suffix of habilis is -bilis, not -ilis, which did develop into the –(i)l(e) suffix in English as in <Gent + (i)l(e) → gentle>, <Gent + (i)l(e) → gentile>, and <Cive + (i)l(e) → civil>. English also retains the full Latin -ilis but largely in scientific contexts. English also has the free base <Habile>, which forms the word habile meaning “generally able or adroit, handy.” The <Habile> base comes directly from Latin habilis rather than passing through French. Thus, <Able> and <Habile> are doublets.)

Moving back to my original question, I investigate the etymology of the -able suffix, which is an adjective suffix denoting capability or possibility. Although semantically related to the word able, the suffix has a completely different etymology. English borrowed -able from Old French -able and directly from Latin -ābilis. The Latin -ābilis developed from a rebracketing of first conjugation Latin verbs, which refers to Latin verbs with an <a> in the stem. The infinitives of first conjugation verbs end in -āre. Latin -ābilis thus comes from -a- + -bilis. (The same etymological route occurred with -ible but with an from the stem of third conjugation verbs.)

Etymologically, the <Able> base that forms the word able and the -able suffix clearly differ. Although related, the meanings of the two forms also differ slightly. Different etymologies and different semantics mean different morphemes.

The -ability suffix also differs etymologically from the word ability, which comes from <Ability>. The suffix -ability comes from Latin –abilitas, which comes from –ābilis + -i- + -ty. By surface analysis, -ability is -able + -i- + –ty. However, because the word sum <able + i + ty> produces *ablity, which is not the spelling of the suffix, I know that -ability is a suffixal construction.

Now, as I write that -ability is a suffixal construction, I begin to wonder why I cannot represent the suffixes -able and -ability as the parenthetical <ab(i)le>. Both suffixes developed from Latin –ābilis with the between the and <l>. If the source is the same but the surfaces only in -ability, then why can I not identify the morpheme as the parenthetical <ab(i)le>?

I know that <ab(i)le> surfaces with the because of the words amabile and cantabile. The –abile suffix comes to English from Italian –abile, and the Italian -abile comes from the Latin -ābilis. Both -able and -abile mean “adjective suffix denoting capability or possibility,” then why can I not represent both with <ab(i)le>? Both have the same meaning and come from the same Latin source.

If the morpheme is <ab(i)le> with the parenthetical , then the <ab(i)le> can surface as either the common -able or the rare -abile. If the morpheme is <ab(i)le>, then the suffixes -ability and -abilize would consist of the -ab(i)le suffix plus additional suffixes.

The English -ability comes from Latin –abilitās, which comes from –ābilis + –tās. The Latin –tās is a source of English -ty. I also know that the suffix is never -ity, always -ty. Thus, I know that the before the -ty in -ability is a connecting vowel. Thus, if the morpheme is <ab(i)le>, then words that end in the -ability suffix consist of <ab(i)le + i + ty>.

<ac + Cept + ab(i)le + i + ty → acceptability> (<ac + Cept + ability>)
<Break + ab(i)le + i + ty -> breakability> (<Break + ability>)
<Note + ab(i)le + i + ty → notabilty> (<Note + ability>)
<Memor + ab(i)le + i + ty → memorability> (<Memor + ability>)

The -abilize suffix in stabilize would also consist of <ab(i)le + ize>.

<St + ab(i)le + ize → stabilize> (<St + abilize>)
<St + ab(i)le + ize + ate + ion → stabilization> (<St + abilize + ate + ion>)
<St + ab(i)le + ize + (e)r → stabilizer> (<St + abilize + (e)r>)

Two other suffixes related to -ab(i)le would not be further analyzable as <ab(i)le> plus additional affixes: -abilia and -ably.

The word memorabilia comes from Latin memorābilia, which is the plural of memorābile. Both plural -ābilia and singular -ābile are forms of -ābilis. Thus, -abilia and -abile are separate, but related, suffixes in English.

<Memor + abilia → memorabilia>
<Memor + abile → memorabile> (rare singular form)

The adverb suffix -ably is also a separate but related suffix to the -ab(i)le suffix. First attested in Middle English (and therefore not a Latin suffix), -ably is a suffixal construction formed from -able + -ly. This particular -ly is an adverb suffix of English-origin, ultimately from Old English -līċ. The Modern English -able combines the -able that surfaces with the -ly suffix. In addition to combining morphemes of different language origins, the -ably cannot come from <ab(i)le + ly> because the word sum would produce *-ablely or *abilely, neither of which are -ably. The -ably suffix is just <-ably>.

If the base of the -able suffix that surfaces is <ab(i)le>, then the base of the -ible and -ibility suffixes would likewise be <ib(i)le>. English does not have an -ibile suffix like -abile. (Italian does, and the suffix comes from Latin -ibilis.) But English has an -ible from Latin –ibilis and an -ibility from Latin –ibilitas from -ibilis + –tās. Both suffixes would thus consist of the morpheme <ib(i)le>. The -ibility would then also have the connecting vowel -i- and the -ty suffix.

<Flex + ib(i)le → flexible> (<Flex + ible>)
<Flex + ib(i)le + i + ty → flexibility> (<Flex + ibility>)
<Hor(r) + ib(i)le → horrible> (<Hor(r) + ible>)
<Ter(r) + ib(i)le → terrible> (<Ter(r) + ible>)
<Crede + ib(i)le → credible> (<Crede + ible>)
<Crede + ib(i)le + i + ty → credibility> (<Crede + ibility>)

The -ibilize suffix in compatibilize would also consist of <ib(i)le + ize>, the -ibilism in impossibilism of <ib(i)le + ism>, and the -ibilist in impossibilist of <ib(i)le + ist>.

<com + Pate + ib(i)le + ize → compatibilize> (<com + Pate + ibilize>)
<im + Poss + ib(i)le + ism → impossibilism> (<im + Poss + ibilism>)
<im + Poss + ib(i)le + ist → impossibilist> (<im + Poss + ibilist>)

Like -ably, the –ibly suffix would remain an unanalyzable modern suffixal construction from –ible + –ly because <ib(i)le + ly> would produce *-iblely or *ibilely, neither of which are –ibly.

Finally returning to my initial question about -able and -ab(u)le, I already know the etymology of the -able that surfaces in words like movable and notable. But what about the -able and -abule that surface from <ab(u)le> in pairings like stable~constable~constabulary and fable~fabulous?

In all these words, the <ab(u)le> comes from Latin –ābulum. The development of –ābulum is similar to <ab(i)le> in that the initial <a> results from a rebracketing the stem of a first conjugation Latin verb. The Latin -bulum is an instrumental suffix denoting means or agency which gives English the singular -bulum and plural -bula suffixes, the -b(u)le suffix, and the singular -bula and plural -bulae suffixes.

To finally answer my initial question, -able and -ab(u)le are identified as different suffixes because the two morphemes are different suffixes. The etymologies are different. The -able comes from Latin –ābilis. The -ab(u)le comes from Latin –ābulum. The meanings are different. The -able is an adjective suffix denoting capability or possibility. The -ab(u)le is an instrumental suffix.

While on my roundabout journey to establishing that -able and -ab(u)le are different suffixes, I also uncovered that -able is one form that surfaces from the <ab(i)le> morpheme. I already knew that <ab(u)le> surfaces as both -able as in fable and -abule as in fabulous. Once I realized that -able and -abile surface from the <ab(i)le> morpheme, I could clearly see the difference between the -able denoting capability or possibility and the -able denoting means or agency. The <ab(i)le> and <ab(u)le> are the morphemes. The -able, -abile, -able, and -abule are the allomorphs that surface. The -able from <ab(i)le> and the -able from <ab(u)le> are homographs but are allomorphs of different morphemes.

Please correct me if I am wrong.

Addendum

I need to correct myself because I was wrong, or, rather, partially wrong. I have been thinking about the -able and -ab(u)le suffixes obsessively since finishing the initial publication of this post.

One thought was whether the suffix is <ab(u)l> or <ab(u)le>. If the morpheme did not require the replaceable <e>, then no words would end in -able or -abule. While no words end in the -abule form because another suffix always follows (constabulary, fabulous), the -able form does surface word-finally (stable, constable, fable). Thus, the forms that surface are -able and -abul. The parenthetical does not surface in -able, and the replaceable <e> need not surface in -abul. I could there represent the morpheme as <ab(u)l(e)>. However, because only vowel-initial suffixes that replace the the replaceable <e> ever follow the -abule, I can leave the morpheme as <ab(u)le>. If a word in which a consonant-initial suffix did follow the <ab(u)le> and the <e> were not present, then I would need to amend my analysis to <ab(u)l(e)>.

Second, I have previously and repeatedly learned that -able, -ability, -abilize, and any other yet-discovered or -coined suffixes are distinct and unanalyzable suffixes. The discovery of the rare yet existent -abile in cantabile and amabile sparked my reconsideration.

While different forms of the same Latin ending or suffix result in different suffixes in English (memorabile with singular -abile versus memorabilia with plural -abilia). One Latin ending or suffix can surface as more than one allomorph in English (gentile with -ile, gentle with -le, and civil with -il, but all from the morpheme <(i)l(e)> ultimately from Latin -ilis.)

The -able in words like readable and washable and the -abile in the words cantabile and amabile both developed from Latin -ābilis. The difference in spelling is because -able arrived to English through French and -abile through Italian. The ultimate source, however, is the same: Latin -ābilis. I am thus confident that -able and -abile are allomorphs of the <ab(i)le> morpheme.

English does not have an -ibile suffix that surfaces word-terminally, only an -ible, so the etymological and synchronic evidence supports identifying the morpheme as <ible>. Then again, the modern -ibilize, -ibilist, and -ibilism make me reconsider whether -ible could be an allomorph of an <ib(i)le> suffix.

I must thus again continue my investigation to make sense of the semantically and etymologically related -ability, -abilize, -ibility, -ibilize, -ibilist, and -ibilism. Returning to the etymology of -ability and -ibility, I immediately realize my mistake. Both suffixes developed in Latin, -ability from -abilitās from -ābilis + -tās and -ibility from -ibilitās, from -ibilis + -tās. Neither developed in English from <ab(i)le> or <ib(i)le. When a morpheme enters English matters! Thus, -ability and -ibility, although etymologically developing with the -tās suffix that became -ty in Modern English, are not an English <ab(i)le + i + ty> or <ib(i)le + i +ty> but unanalyzable suffixes from Latin -abilitās and -ibilitās.

I then look at the other suffix related to the -able: –abilize. I can see the -ize suffix, which forms a verb, on the end. I can also see the potential <ab(i)le> surfacing as -abile. But is -abilize <Ab(i)le + ize>? The earliest attested use of -abilize that I could find dates from only 1861 in the word stabilize from stable + -ize. (Other words ending in -abilize appear around the same time or later.) The word stable is <St + ab(i)le> with the <ab(i)le> surfacing as -able. Suffixing -ize to <St + ab(i)le + ize> results in stabilize with the <ab(i)le> realized as -abile.

I then finally reconsider the -ibilize, -ibilist, and -ibilism suffixes. Like -abilize, are these suffixes built on the -ibile allomorph of the <ib(i)le> morpheme? The -ibilize is much less common, so finding the earliest use is harder, but the OED lists sensibilize as dating from 1860. Both British compatibilise and American compatibilize gain usasge in terms of computers and are thus even more modern, dating from the late twentieth century. I found flexibilization in Google Scholar, with the oldest usage dating from only 2024. The -ibilize is clearly a modern creation.

Despite the limited use of -ibilize (compared to -abilize), I can see that these few examples have been built from an -ible word: sensible~sensibilize, compatible~compatibilize, flexible~flexibilization. I can see the same allomorphic process arising as with <ab(i)le> and <ab(u)le>. The medial surfaces when followed by another suffix but not when word-final.

The same is true for the even less common -ibilism and -ibilist. Both endings developed from -ible words: fallible~fallibilism~fallibilist, impossible~impossibilism~impossibilist, perfectible~perfectibilism~perfectibilist. Both endings are also modern creations.

Of course, identifying -abilize, -ibilize, -ibilism, and -ibilist hinges on the morphemes <ab(i)le> and <ib(i)le> surfacing as the allomorphs -able~-abile and -ible~-ibile. The etymology from Latin -ābilis and Latin -ibilis supports my allomorph hypothesis. That the able~-abile and -ible~-ibile pairs both denote “adjective suffix denoting capability or possibility” also supports my hypothesis.

TLDR: The suffixes -ability, -ably, -ibility, and -ibly are distinct morphemes. Both -ability and -ibility developed from Latin suffixes. Both -ably and -ibly are modern suffixal constructions. The morphemes <ab(i)le> and <ib(i)le> surface as the allomorphs able~-abile and -ible~-ibile. The suffixes -ize, -ism, and -ist can affix to <ab(i)le> and/or <ib(i)le> to form the complex -abilize, -ibilize, -ibilism, and -ibilist.