The English spelling system is a rule-based system. A system is a group of parts that operate together as a complex whole. The rules of the system govern the ways in which the parts work together. In my spare time over the past few years, I have been investigating words and word families using Structured Word Inquiry as a means of understanding the English spelling system and its governing rules are completely and accurately as possible.
One way that I have been keeping track of my knowledge gained is through my Morphodex. As I uncover morphemes, I save those morphemes along with their denotations, etymologies, and evidence so that I do not have to perform every word study from scratch. I can refer to my collection of morphemes when I think that I have found another example of a previously-identified morpheme.
As I studied the word trachea, I asked myself, “When do I need to clip a new base from another word, and when can I identify a new suffixal construction?”
I first brought up the question with myself while studying the words gravity and gravitate. The -ty on gravity is a noun suffix from Latin -tās. The word gravity thus consists of the free base <Grave>, connecting vowel -i-, and suffix -ty. The English word gravity comes from Latin gravitās, which contains the Latin form of the -ty suffix. English also directly borrowed the word gravitas.
<Grave + i + ty → gravity>
<Grave + i + tas → gravitas>
The word gravitate, however, did not develop until the early 1600s, which means, in terms of the history of the English language, the word is fairly new. New Latin has the verb gravitāre, but New Latin in not Classical Latin. New Latin gravitāre was created from Classical Latin gravitās from gravis + –tās, which means the ultimate base of gravitate is <Grave>.
But <Gravity + ate> does not produce gravitate. <Gravity + ate> produces *gravitiate, which is not an English word. Instead, the -itate in gravitate is a modern suffixal construction from -i- + -ty + -ate (not a word sum). A suffixal construction is a new unanalyzable suffix that originates from a combination of other suffixes and typically found in Classical compounds or modern coinages. English created the new -itate suffix by combining -ity with -ate.
<Grave + itate → gravitate>
Identifying the base of gravitate as <Grave> preserves the morphological continuity with gravity, gravitas, and other <Grave> words whereas positing a new clipped base like <Gravite> would obscure the shared etymology, semantics, and morphology. Evidence for suffixal constructions also comes from Latin as in the -ability suffix from Latin –abilitās from –ābilis + –tās, which is a suffixal construction from –ābilis + -i- + -ty, with the -ābilis rebracketed from the <a> of stem of first conjugation Latin verbs + –bilis.
English has many suffixal constructions, some from classical suffixes and some more modern creation. For example, -aneity as in spontaneity is a noun suffix from Latin –āneitās from –āneus + –tās. In the word audacious, the –acious is a modern suffixal construction from –acity + –ous. The –acity as in audacity is also a suffixal construction from Medieval Latin –ācitās, which comes from Latin –āx + –tās.
Studying the word trachea, I found the relatives tracheal and the erroneously identified suffix tracheo-. (Dictionaries often identify trachea- as a prefix or combining form, but the word part is a bound base and connecting vowel.) The word trachea comes from Latin trāchēa from Ancient Greek trākheîa. The Latin -ēa and Ancient Greek -eîa endings become the singular noun suffix -ea in English. The base is the bound base <Trach> denoting “windpipe.”
<Trach + ea → trachea>
Looking at tracheal, I immediately see that the <al> is the adjective suffix. However, I also wonder about the preceding <e>, which could be a connecting vowel but might also have been retained from the -ea. According to the OED, the earliest attested use of tracheal is from 1710. Wiktionary tells me that the word comes from trachea + -al. Because <Trach + ea → trachea> and tracheal comes from trachea + -al, then the -eal is a suffixal construction from -ea + -al. The same process of suffixalization occurs with trachean, tracheole, tracheid, and tracheary from the -ea and then -an, -ol(e), -id, and -ary.
<Trach + ean → trachean>
<Trach + eole → tracheole>
<Trach + eid → tracheid>
<Trach + eary → tracheary>
Although uncommon, these suffixal constructions can occur in other words such as corneal from <Corn + eal> and obsolete cochleary from <Cochl + eary> (and the modern cochlear from <Cochl + ear>).
But what about words like tracheoscope and tracheotomy that begin with trachea- (misidentified as a prefix)? The etymologically and semantic connection with trachea and the <Trach> base is clear. Words built with an initial tracheo- cannot simply consist of the base <Trach> plus a connecting vowel and then additional morphemes because both the <e> and the <o> would have to be connecting vowels, and a word cannot have two connecting vowels in a row. If the base were <Trach> then the part of the word that follows would be some sort of suffix. However, words like tracheoscope and tracheotomy have additional bases: <Scope> in tracheoscope and <Tome> in tracheotomy. The part that follows the <Trach> cannot therefore not be a suffixal construction.
Instead, the <Trache> in words like tracheoscope and tracheotomy is a new, albeit related, base clipped from the word trachea. The final <e> is pronounced, not a replaceable <e> (like in the bound bases <Ge> and <De> and the free bases <Me> and <He>).
<Trache + o + Scope → tracheocope>
<Trache + o + Tome + y → tracheotomy>
The same process of clipping a word to create a new but related base also occurs for other words such as the <Kinese> from kinesia in kinesiology, <Chore> with a pronounced final <e> from chorea in choreographer, and <Card> from Latin cardi in carditis.
Finally, to answer my initial question, I must clip a new base from an existing word or base when the morphemes that follow the new base include at least one additional base. However, when two or more suffixes combine into a new suffix in a way invalidated by a word sum, then the result is a new suffixal construction.