After I posted on presentee (vs. absentee), commenters on Facebook wondered about the derivational suffix –ee, and I noted that there was a fair amount of literature about it. Here I cite some of this, including a famous proposal (well, famous in linguistics) that will take us to a strange and unexpected place: ergative-absolutive languages (vs. nominative-accusative languages).

The great resource on –ee is a 1998 paper by Chris Barker in Language (74.695-727), “Episodic -ee in English: A thematic role constraint on new word formation” (stable URL here), which uses a database of “fifteen hundred naturally occurring tokens of some five hundred word types” to analyze the semantics of the suffix; it also has a full bibliography of relevant literature on the subject. I’ll come back to Barker after a detour into ergativity.

Wikipedia has an excellent article on ergativity, beginning:

An ergative [more fully, ergative-absolutive] language maintains a syntactic or morphological equivalence (such as the same word order or grammatical case) for the object of a transitive verb and the single core argument of an intransitive verb, while treating the agent of a transitive verb differently.

This contrasts with nominative–accusative languages such as English, where the single argument of an intransitive verb and the agent of a transitive verb (both called the subject) are treated alike and kept distinct from the object of a transitive verb.

There are three arguments distributed in two different ways here. Using now-standard symbols for the three arguments —

O = object of transitive verb (also symbolized as P for “patient”)
S = core argument (subject) of intransitive verb
A = agent of transitive verb

these are the two ways they are aligned:

ergative-absolutive alignment: A (ERG) vs. S and O together (ABS)
nominative-accusative alignment: A and S together (NOM) vs. O (ACC)

(The term ergative derives from the Greek root erg– ‘work’.)

English is a nominative-accusative language, with the arguments distinguished by word order and (for personal pronouns) morphological case. Similarly, Japanese, where the arguments are distinguished by word order and by marker words (the postpositions NOM ga and ACC o). In contrast, Basque distinguishes its arguments by suffixes (with ERG –ak for A and ABS –a for O and S).

Note (1) on intransitivity: semantically, intransitive verbs come in two varieties: those with agentive subjects (run, talk, resign) and those with non-agentive subjects (die, fall, arrive). For complex historical reasons, the two types are known as unergative and unaccusative verbs, respectively.

Note (2) on intransitivity: English dictionaries customarily take object to refer to direct objects only, so that verbs with oblique objects, marked by prepositions — like adhere, with an object marked by to — are labeled in these dictionaries as intransitive. In discussions of the two systems of marking, as above, objects are objects, no matter how they are marked.

Note (3) on intransitivity: English allows for a considerable range of object omissions — types of object ellipsis — giving rise to instances of verbs that lack an object but are nevertheless “understood transitively”. (Recipe Object Omission, for instance: Take a cup of noodles and boil for 20 minutes, with boil understood as a transitive, understood as having the noodles as object.) In discussions of the two systems of marking, object-less verbs that are understood transitively do not count as intransitives.

Now, more from the Wikipedia article, which (finally) takes up English –ee:

English has derivational morphology that parallels ergativity in that it operates on intransitive verbs and objects of transitive verbs [that is, on absolutives]. With an intransitive verb, adding the suffix -ee to the verb produces a label for the person performing the action:

“John has retired” → “John is a retiree”
“John has escaped” → “John is an escapee”

However, with a transitive verb, adding -ee does not produce a label for the person doing the action. Instead, it gives us a label for the person to whom the action is done:

“Mike employs Susie” → “Susie is an employee”
“Mike has appointed Susie” → “Susie is an appointee”

Etymologically, the sense in which “-ee” denotes the object of a transitive verb is the original one, arising from French past participles in “-é”. This is still the prevalent sense in British English: the intransitive uses are all 19th-century American coinages and all except “escapee” are still marked as “chiefly U.S.” by the Oxford English Dictionary.

Chris Barker’s 1998 article, using a large database, concludes that the formation of nouns in –ee is moderately but genuinely productive, but that analyses based on the syntactic argument structure of the stem verb are unsatisfactory (the hypothesis is suggestive, but doesn’t work in detail; in particular, there’s a big class of cases like amputee, in which the referent doesn’t correspond to a syntactic argument of the stem verb at all). Barker proposes three essentially semantic constraints, the first of which — the referent of the derived noun must be sentient — is pretty much implicit in earlier discussions. The third — positing a relative lack of volitional control on the part of the derived noun’s referent — captures part of the spirit of the absolutive proposal, by uniting the semantics of the subjects of unaccusative intransitives and the objects of transitives.

The third condition — the denotation of the noun must be episodically linked to denotation of its stem — requires more development of the notion episodically linked than I can provide here.

Leave a Reply

%d bloggers like this: