Is the theta role in one language (ex. English) - L1 the same as in another language - L2, when this two sentence are about the same ? Can anybody give me example, when they are different ?

  • 4
    Theta roles are theoretical constructions; they aren't phenomena that can be detected or measured in any language, let alone compared in two languages. So the answer is that they can be the same if the person writing the definition says they are the same, and otherwise not. – jlawler Sep 4 '14 at 20:03
  • Yes, I know that theta roles (or thematic roles) usually don't have influence on the form of the words. But if the meaning is the same in both languages, so theta roles also should be the same, isn't it ? Maybe they are diffrent when the same action is different perceived in another culture - using L2 language. – Adrian Sep 5 '14 at 18:14
  • 1
    The problem is deciding whether "the meaning is the same in both languages". In most cases, it's not, so it's hard to talk about the theta roles. You're right that things -- all kinds of things -- are different in different cultures, and in individual grammar and semantics as well, whatever languages are being spoken. But that doesn't say much about theta roles, I'm afraid. Maybe what you're talking about is part of FrameNet, but I doubt whether it's called "theta roles" there. – jlawler Sep 5 '14 at 18:18
  • I think that the meaning is usually similar if we want to translate the sentence from L1 to L2. For example, in L1 (Spanish) we could say: yo naci en colombia, and in L2 (English) I was born in Columbia. And in both of them "I" and "yo" is the same think (person), so we can assign it the same theta role - patient. FrameNet is a good example of similar roles in verbs, in different languages, I think. – Adrian Sep 6 '14 at 8:04

I believe that theta roles have to be considered universal or else they are a useless concept. By definition they have to do with the underlying meanings and relations in a statement. When it comes to general word definitions it is very difficult to find one-to-one correspondence between languages because ordinary words tend to be polysemous and can have nuanced meanings with various connotations and histories in a language. But when talking about theta roles, or semantic roles as I would call them, we are talking about a purported underlying system in language that is necessary for any language to function at all. Basic concepts like SOURCE, PATH, and GOAL which do seem to be present in all languages (to my knowledge.)

However, that is not to say that "when two sentence are about the same" that the semantic roles in each language will match up perfectly. I don't have any example off hand, because I don't know any second language well enough, but I feel like I can almost think of an example where an expression is translated into another language and in the process the semantic roles have to change. Like the patient in L1 may be expressed as a recipient or something in L2, even though you are translating as directly as possible. Does that make sense?

This isn't an example but it can give an idea of how languages can "think" differently about the same concepts. Some languages like Finnish have no verb for "have" so they instead speak about things as being on someone. Like "minulla on koira" -- literally "on me is a dog".

The answer to this question Thematic roles in some languages seems relevant.

| improve this answer | |
  • +1 for you, I think this caputeres it quite well: Semantic roles attempt to be universal, after all that's what they're made for (and like everything which is supposed to be univeral, there is a lot of dispute about what those primary roles are) , but languages might make use of different thematic roles to convey the same meaning. – lemontree Jul 6 '16 at 8:58
  • An example I could think of is a Spanish sentence like Me gusta la paella, where me has the role of probably an EXPERIENCER, while the most likely English sentence for the same meaning would be I like (the taste of) paella, where the subject is, at least judging from the nominative case, more AGENS-like. You could try to translate the Spanish sentence literally to something like (The taste of) paella pleases me, which would be closer to the distribution of the thematic roles, but just not the way the language would naturally express this menaing. – lemontree Jul 6 '16 at 8:58
  • I also think Finnish with its rich case system is very nice to show of differently languages can express meanings. In Finnish, as you say, you don't have something, but something is on you (adessive); you don't tell about your thoughts, you tell out of your thoughts (elative); you don't love someone, you love of someone (partitive for an irresultative aspect, probably resembling that your love has no effect on your loved one - or that you can never fully love someone ;) ), ... – lemontree Jul 6 '16 at 9:01
  • You shouldn't be too fast judging from that whether Finnish speakers really think differently, but at least the language does. – lemontree Jul 6 '16 at 9:01
  • @lemontree I think you're confusing semantics with syntax. Me gusta la paella and I like paella are different syntactically, but if their meanings are the same, the semantic roles involved are also the same (in this case e.g. Experiencer and Stimulus, depending on the inventory of roles in your favored theory). I agree with the first part of Moss's answer -- semantic roles are intended as being universal -- and therefore not with the second part, if it implies that semantically equivalent sentences in different languages will involve different semantic roles. – TKR Jul 7 '16 at 1:50

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.