Ed,

My impression is that there are two ways to define “indirect object” and (as is frequently the case with definitions) no universally accepted way to arrive at a pronouncement about the “right” definition. From this and previous discussions, I think we are dealing with which one(s) of the following three claims a given analyst wants to accept:

(1)   [Semantic] Indirect Objects are Beneficiaries (to cadge a term from Fillmore).

(2)   [Semantic] Only Beneficiaries that are *required* by the semantic structure of the verb count as indirect objects.  “Requirement,” in this sense, is adduced from the observation that making a sentence with the verb without an overt Beneficiary sounds strange unless the Beneficiary is clear from context.

(3)   [Structural] Objects have to be bare complements to the verb. PPs do not, therefore, count as objects.

Playing these off against three example sentences yields different judgments:

(a)               Bjarki gave Bjorn some lutfisk.

(b)               Bjarki gave some lutfisk to Bjorn.

(c)               Bjarki made some lutfisk for Bjorn.

Those who accept (1) but not (2) or (3) will see an indirect object in all three sentences (and I would consider “hit the ball to me” analogous to (c)). Those who accept (2) but not (1) or (3) will see an indirect object in (a) and (b) only. Those who accept (2) and (3) will only see an indirect object in (a). Those who accept only (3), eschewing semantic definitions entirely, will simply talk about complements – some verbs have one, some verbs have two, and you can call them whatever makes you happy.

The problem, of course, is that “Indirect Object,” as a grammatical concept, was originally formulated to deal with the distinction between accusative and dative objects in Latin, but English doesn’t have a dative case, really. Early English grammarians frequently decided to consider prepositions as being exactly equivalent to case marking, even to the extent of making case tables with prepositions in them (accusative: him; dative: to him; ablative: from him). This led quite naturally to considering both the “bare” complement in (a) and the “prepositional complement” in (b) as being in the dative case, thus exactly analogous to their Latin equivalents. I’m not bringing this up to try to argue that calling the third constituent in (b) is a bad thing, but rather to argue that the argument itself is irresolvable. It is quite literally a matter of definition.

 

Bill Spruiell