3

I am assuming that the number of transitive verbs that take both a direct and indirect object in English is a subset of those that just take just a direct object. Does anyone know how much smaller? How does this compare with other languages (e.g., German, Spanish, Russian)? Are there databases that can be used to explore this further?

5
  • 2
    Some would argue that verbs can be indirectly monotransitive – that is, have an indirect object without a direct object (e.g., “I told you already!”). Others would argue that the indirect object gets bumped up to direct object status in such cases. I don’t personally buy that, but some do. Even so, I don’t think there are any verbs that can only take indirect objects: every IO verb can also, sometimes optionally, take a DO in the same construction. Feb 15 at 22:59
  • @JanusBahsJacquet You read my mind because this was also something I had been wondering about, but I didn't want to complicate the initial post. Thank you for your contribution.
    – LISA
    Feb 15 at 23:28
  • 1
    There are also verbs that can't take only a direct object on its own: *I put the book, *I gave the letter.
    – Draconis
    Feb 16 at 1:15
  • Despite doing a full linguistic degree, I never learnt the difference between "direct" and "indirect" objects. They're all just complements to me!
    – curiousdannii
    Feb 16 at 1:30
  • @JanusBahsJacquet otoh, such verbs never explicitly mark their complement as an indirect object, e.g. it's always "I told the man already" not "I told to the man already". This isn't necessarily a problem though as there are doubtless explanations of the "I gave him it" vs "I gave it to him" (and regional "I give it him") reordering that would cover this too
    – Tristan
    Feb 16 at 11:49

2 Answers 2

3

As John Lawler notes in his comment, the big problem is defining the class of verbs you mean. Typically, we use "give" as an example of a verb that can have both an indirect and direct object: you can say "I gave Bob a cake". You can also say "I gave a cake to Bob": does that version count as an example of IO+DO? Let's say "no" (if you say "yes" then the answer probably has to be "pretty much all of them"). An Indirect object is an "object-like NP" that doesn't have a preposition. So we also have "I sent Bob a letter" which has two bare O-like nominals.

We also have "I baked Bob a cake", which we could derive from "I baked a cake for Bob". Now we can start looking for verbs that we can do "for" people, and see to what extend we can shift "V NP1 for NP2" to "V NP2 NP1", i.e. "dug Bob a hole", "skinned Bob a vole", "bought Bob a pole", and also try more examples of the analogous shift with "to", like "threw Bob a rope", "showed Bob the dope".

So a paraphrase of the question could be: which verbs engage in the V NP prep-NP ~ V NP NP alternation? For example, *"introduced Bob the pope", *"lobbied Bob his vote". As far as I can tell, "underneath" can never enter into that relation (*"I found the table the book" meaning "I found the book underneath the table"). It's not always or obviously always prepositional phrases, though, for example "They elected Bob mayor" matches "They elected Bob to be(come) mayor" or "They elected Bob as mayor", without movement or a second PP (unless you think of "as mayor" as a PP). Of course nobody would say that "I hammered the metal flat" has an IO and a DO, because "flat" is an adjective, not an NP.

In Bantu languages, the general answer is "everything", because there are valence-increasing suffixes that allow all forts of thematic roles to be integrated as bare NPs.

5
  • Benefactive constructions like buy a book for Bob can do Dative to buy Bob a book if Bob winds up with the book; otherwise not. Cf I fixed John a hamburger and *I fixed John the roof.
    – jlawler
    Feb 17 at 23:45
  • @jlawler Good points all. Another thing I got to pondering is the concept of transitivity, independent of any language in which it is conveyed. For example, if the concept of "give" is something that can involve both a direct object and an indirect object, then wouldn't this really be true in any language? What would cause statistical variance, then, are the number of words any given language has for verbs like "give" and its synonyms that share this same quality.
    – LISA
    Feb 23 at 2:11
  • According to some sources, German has eight times as many words as English. And, if verbs that can take both a direct object and an indirect object are a small subset of the total number of verbs, it would be logical to conclude that disparities between languages will be multiplied accordingly.
    – LISA
    Feb 23 at 2:12
  • So, for example, if one out of four verbs in English can take both a direct and indirect object and only one in six in some other language that has more verbs in total, it is possible that the number of concepts we have that involve both types of object is virtually the same and that the difference, statistically, is merely a reflection of a larger total number of verbs.
    – LISA
    Feb 23 at 2:12
  • It's not a question of number of verbs - it's a question of what an object is, and that's a big rabbit hole. The semantics of subjects is determined by the verb -- agent, experiencer, instrument, etc, depending. Indirect objects are even more determined by the verb -- mostly they're receivers, with dative verbs, but occasionally they're resultatives, like elect. But the semantic relations between the verb and its direct object are part of the meaning of the verb, and vary wildly from verb to verb. And, of course, from context to context, and from speaker to speaker.
    – jlawler
    Feb 23 at 16:15
2

I am not a linguist by trade, so I don't know if I should even attempt to answer this, but here it goes because curiosity is killing this cat. In an attempt to answer my own question, I simply took the first most frequent 25 verbs in English and then compared their transitivity to a randomized selection of the most frequent 1,000. I did this simply because I was interested in how they compared.

There were some flaws in this methodology because I did not use the same source for the most frequent 25 as I did for the most frequent 1,000, but if you want to take a look at a rough idea of the distribution, look below:

enter image description here

So, after doing that, I attempted to do the same thing with German verbs. As with the English, the first most frequent 25 German verbs were from a list of infinitives. That may be the most consistent aspect of the methodology.

But the methodology for the collection of German words is contaminated with flaws as well because I also did not use the same lists for the first most frequent 25 and the most frequent 1,000. In addition to that, the English list did not appear to show any past tense forms, but it did have conjugated present tense forms. This is not true for the German which appeared to contain not only all conjugations, but all tenses as well. Nevertheless, perhaps some of you can draw some conclusions out of these charts just the same. (Or, perhaps this poor answer will inspire someone to contribute a better one!) The German distributions are below:

enter image description here

Due to the lack of rigor in the methodology and my lack of experience with linguistics, I won't draw too many conclusions from these charts, but what I didn't know before I began this attempt to answer my question is just how many verbs in English aren't just either exclusively transitive or exclusively intransitive. Many verbs can play different roles along the transitivity spectrum depending on context. Even so, it does appear that this is less of a tendency with German verbs. Of particular note is the larger percentage of exclusively transitive and intransitive German verbs compared to those for English. (44% vs. 32% for transitive and 20% vs. 12% for intransitive.) Perhaps this is common among languages with case systems. Perhaps in languages without case systems, verbs become more flexible in usage which affects their transitivity as well.

For the record, I used WordReference.com to assign transitivity for both the English verbs and the German verbs.


*Note: I realize that this answer morphed into content regarding transitivity rather than what percentage of verbs take both a direct and an indirect object, so let me close by saying that when I looked at the randomly selected transitive verbs that were included in the tabulations for this answer and determined which ones took both a direct and indirect object, I came to this estimate:

enter image description here

Granted, the numbers are really, really small, and the methodology was a bit shoddy, so these estimates are a bit of a leap. In fact, if this were an academic paper, I would never settle for such limitations, but as it is, especially due to a lack of experience and linguistic toolsets, I'm content to leave it at this for now.

4
  • 1
    It's pretty good for back-of-the-envelope figuring. As you noticed, the real problem is defining "transitive", and much ink and many pixels have been spilled doing it. The syntactic theory I follow says that 3-argument predicates, like give or elect, distinguish the two NPs in their VP semantically, and usually also syntactically. "Indirect object" is a term syntacticians don't use much, because it almost always implies the existence of a "direct object" as well, which may not be present in a given sentence. For that matter, neither object may be present: I gave at the office.
    – jlawler
    Feb 17 at 18:43
  • 1
    @jlawler You're so kind and supportive. I love it! In fact, posting on a forum on a topic of which you are not much of an expert is a bit daunting, but I have found the members of this stack exchange to be polite and encouraging ... much more so than some others I've been a part of! BTW, I read your bio. Very interesting. I look forward to exploring your links further.
    – LISA
    Feb 17 at 21:53
  • 1
    You might find Relational Grammar useful. It starts off by assuming that the grammatical relations of subject, object, and indirect object (called simply "1,2,3" in the theory) are the basis of syntactic analysis. Passive in Relational Grammar is "promote 2 to 1", i.e, make the object into the subject. Dative is "promote 3 to 2", i.e, convert give a book to Bob into give Bob a book. If you do dative before passive, you get Bob was bought a book.
    – jlawler
    Feb 17 at 23:57
  • @jlawler Sounds very interesting. Thank you for the recommendation!
    – LISA
    Feb 23 at 2:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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