In traditional English grammar, we're taught that phrases like those boldfaced below are "indirect objects":

  1. I gave the book to Ted.
  2. I gave Ted the book.

But this appears to be based on semantics (namely, Recipient role) rather than syntax. In (1), to Ted is a prepositional phrase, which by most definitions means it can't be an object of any kind. In (2), Ted doesn't seem to be syntactically any different from other objects: such arguments use the objective form where this is distinct (I gave him the book), and they can serve as the subject in a corresponding passive construction (Ted was given the book).

In other languages, "indirect object" is sometimes used to refer to an obligatory argument that appears in an oblique case, as in German:

  1. Ich helfe dem Mann. "I help the man-DAT"

Here there is syntactic justification for regarding this argument as something different from a regular object (different morphology, also can't be a passive subject), yet still regarding it as an object of some kind (given that it's an NP subcategorized for by the verb). But this is a different situation from that of English, and I can think of yet other ways that languages might treat NPs as somewhat-but-not-completely object-like (e.g., an NP that's cross-referenced by an agreement marker on the verb, but doesn't pass some syntactic test for objecthood).

So, is there a coherent syntactic definition for "indirect object" in English, in other languages, or cross-linguistically? Do syntacticians and typologists use this term at all anymore, and if so what do they mean by it?

  • I've never thought it was a particularly linguistic term. More folk-linguistic in that way.
    – curiousdannii
    Nov 24, 2018 at 10:52
  • 1
    @curiousdannii Well, it's certainly part of traditional grammar, and it seems linguists still use it to some extent, as in the WALS chapter linked in the answer.
    – TKR
    Nov 24, 2018 at 18:54
  • 1
    There is a small error in your German: 'I help the man' should be 'Ich helfe dem Mann'.
    – dumetrulo
    Nov 26, 2018 at 9:34
  • @dumetrulo Thanks for the correction, fixed!
    – TKR
    Nov 26, 2018 at 18:49
  • The way I see is that we can come up with syntactic tests for IO that will work for a specific language (e.g. English or Russian) or those that might work - albeit less accurately - for a number of languages (not necessarily genetically related) but I'm very skeptical of universal syntactic tests. see anything written by Martin Haspelmath. e.g. his 2010 paper Comparative concepts and descriptive categories in crosslinguistic studies jstor.org/stable/40961695
    – Alex B.
    Aug 29, 2020 at 20:35

2 Answers 2


I am by no means a lingustics expert, just fluent in German (native), English (native-level) and Spanish (native-level). As you observed, German retains enough case markers to distinguish four cases much like they existed in (say) Latin. To take up your example '1. I gave the book to Ted' / '2. I gave Ted the book', in German it would look like this:

Ich gab Ted das Buch.

This is equivalent to your number 2. Here, Ted is in the dative case but you wouldn't know that from looking at the word, since names don't get inflected in German (except in genitive case). It helps to add an article (this usage is quite colloquial, and you wouldn't usually see it in written/formal language):

Ich gab dem Ted das Buch.

Here the dative case is clearly visible due to the article 'dem.' Grammatically, 'dem Ted' is the indirect object in the dative case, 'das Buch' is the direct object in the accusative case (here identical to the nominative case since 'das Buch' is neuter). For emphasis or other purposes, you can freely switch the word order around without affecting the inherent meaning of the sentence, e.g.:

Ich gab das Buch dem Ted.

In English you no longer have a case marker to distinguish between dative and accusative case, therefore you need either a convention about where in the sentence you find the direct object, and where the indirect object, or you need to add a preposition (or another morphological marker) to distinguish the two, and make it possible to change word order.

EDIT: Without the article, the sentence is 'Ich gab das Buch Ted.' While correct, it sounds awkward, and most Germans would insert a preposition like in your sentence number 1, making it 'Ich gab das Buch an Ted.'

Why are the objects called direct and indirect? That is because the direct object ('das Buch') will turn into the subject of the corresponding sentence in passive voice, while the indirect object ('Ted') stays in the dative case:

Das Buch wurde dem Ted (von mir) gegeben.

EDIT: Not pertinent to the question but as an aside, there is also a passive voice using 'bekommen' as the auxiliary verb which will turn the indirect object into the subject while leaving the direct object in the accusative case:

Der Ted bekam das Buch (von mir) gegeben.

I am not aware of an equivalent construction in English.

EDIT: As far as the 'ordinary' passive goes, your sentences transform as follows:

  1. I gave the book to Ted. => The book was given to Ted.
  2. I gave Ted the book. => Ted was given the book.

In both cases, 'the book' is transformed from an object of the active voice to the subject of the passive voice, therefore 'the book' is the direct object in the active voice sentence. 'Ted'/'to Ted' remain in object position, and are therefore indirect objects.

Since English is in essence a Germanic language, I suppose linguists felt inclined to take the terminology that existed for German (or for Latin), and apply it to English even though it lacks syntactical markers to make sense of the distinction between direct and indirect object.

  • Absolutely true, except that if you passivise "Ich gab das Buch dem Ted", you get "Dem Ted wurde [von mir] das Buch gegeben" (Ted is still dative, the book is still the subject). In your sentence, "Der Ted bekam das Buch gegeben", isn't das Buch an Akkusativobjekt? Nov 26, 2018 at 10:49
  • @Wilson Yes, in the sentence, 'Der Ted bekam das Buch gegeben,' 'das Buch' is the accusative object. This construction is known as bekommen-Passiv. In the 'ordinary' passive (werden-Passiv), 'das Buch' becomes the subject in nominative case, and 'Ted' remains in dative case, same as in the active voice.
    – dumetrulo
    Nov 26, 2018 at 11:10
  • @Wilson Reading your comment again, I think you got confused as to what is what in the active voice sentence. In 'Ich gab das Buch dem Ted,' 'ich' is the subject in nominative case, 'das Buch' is the direct object in accusative case, and 'dem Ted' is the indirect object in dative case. In 'Dem Ted wurde das Buch gegeben,' 'dem Ted' is still the indirect object in dative case, while 'das Buch' turned into the subject in nominative case (since it is of the neuter gender, nominative and accusative case look identical). This is what happens when you transform a sentence from active to passive.
    – dumetrulo
    Nov 26, 2018 at 11:23
  • That is exactly what I wanted to say. bekommen-Passiv is not pertinent to the question because the question is about English, so I wanted to suggest that you improve your answer by using a passivisation process that befits OP's examples #1 and #2.. Nov 26, 2018 at 11:32
  • In "Ted was given the book", Ted does not move to the book. It uses an archaic syntax, like "To Ted was given the book". Both mean "The book was given to Ted." Note that implicit dative needs a direct object: *"The book was given Ted." Note that agent case marker 'by' has different meaning than preposition 'by'. The same is true for dative case marker 'to', and 'for' is always dative (unless a conjunction). Note archaic direct object marker 'of'.
    – amI
    Nov 26, 2018 at 12:04

You have different manners to define syntactic functions. There are the syntactic position and the morphological marking. These markedness' can be combined or opposite. Moreover, nothing restrains a syntactic function to one markedness. Finally, the syntactic function is determined in relation to other arguments too. So that being said, English I.O. has two markings that are used in complementarity.

First, when IO is placed in the third position and is not marked by a morpheme: S V IO O.

Second, when IO is placed in the fourth position and is marked with to-: S V O to-IO.

These conditions are sufficient not to confuse the arguments. Everything is well discriminated, because the Object, here, has other marks opposite to the Indirect Object.

As for other languages, there is WALS website where you can find more informations: https://wals.info/chapter/105

  • 1
    But when it's inside the prepositional phrase, it's no longer an argument of the verb phrase - I can't see what justifies it being called an object at all. And when in a prepositional phrase it can be freely moved anywhere in the sentence.
    – curiousdannii
    Nov 24, 2018 at 10:53
  • 2
    You are saying that English IO is determined in relation to this semantic features, because, according to you, there is no morphological marker. But, it is not true, because there is a mark, it is the syntactic position. These two considerations, accounting for other arguments, are sufficient, so useful, to characterize the grammatical functions. Your third example seems to be a differential object marking with a marker borrowed to the dative as in Spanish "a".
    – amegnunsen
    Nov 24, 2018 at 21:01
  • 2
    I don't know why you think prepositions can only express spatial relations -- that's obviously not the case. And to is clearly not a prefix because of its distribution; if anything, it would be a proclitic, but that too seems ruled out by e.g. the girl I gave the book to.
    – TKR
    Nov 24, 2018 at 23:37
  • 2
    So a PP can't be an argument of the verb? I find this whole line of reasoning too absurd to pursue further, to be honest.
    – TKR
    Nov 25, 2018 at 18:55
  • 2
    Put takes two arguments, of which the second can be a PP or a locational adverb like down (though some would actually call down a preposition too -- I believe CGEL takes this approach). What is your alternative analysis? More importantly, I don't understand what you're trying to establish -- even if these weren't PPs, how would that answer my original question?
    – TKR
    Nov 25, 2018 at 23:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.