bubble   This is an OPINION PAGE. It may contain different points of view about different parts of Klingon. You may add useful thoughts, but please remember this is not a forum.

The Transitivity Issue

A brief note

My original motivation for creating the questionable verbs page was that some Klingon verbs have ambiguous English glosses. However, some Klingonist disagree that there is even an issue here. I believe there is an important issue, and in the interest of academic reconciliation, I will explain the reasons on this page.

The transitivity issue is probably only of incidental importance with these verbs. The true question is what the role of the subject is, rather than whether there is (or can be) an object. For example, tetDI' vay', is the subject now in liquid form, or did it cause something else to become liquid?

Please keep in mind that just because something has been found to be transitive or intransitive in a single canon sentence, that does not mean it can't also be the other way. Some Klingon words do this, and Marc Okrand has never stated that Klingons take notice of transitivity.

The issue

If anything can be said about the KLI "family", we are precise almost to a fault in our use of tlhIngan Hol. To that end I would like very much to know that I am using words correctly. We rely on Marc Okrand's glosses of Klingon words to know how to use them. Though he glosses most words adequately, there are a few words that have stood to be clarified in the past, and there are a few such words remaining now.

One set of Klingon word which stands to be clarified consists of the verbs which are glossed by an English verb with variable transitivity. For example, let's consider the verb SIH. It is glossed with the English verb bend. Merriam Webster Online lists six transitive senses and four intransitive senses in which bend could be used. The first transitive sense is to constrain or strain to tension by curving: bend a bow. The first intransitive sense is to curve out of a straight line or position; specifically: to incline the body in token of submission. In the transitive sense, it is the object of the verb that is actually changing form, viz. becoming curved, while the subject is merely causing this change to occur. In the intransitive sense, there is only one argument (the subject), and it is that which curves. So it really comes down to a question of what the role of each argument of the verb is. For the English verb bend the thing-that-is-newly-curved can be bumped back to object position while a causal agent is placed in the subject position.

Now, some people would like to claim that Klingon doesn't concern itself with transitivity. But transitivity is a general linguistic concept. It is not limited to English, nor to Indo-European languages. If Klingon really did not concern itself with transitivity at all, then the concepts of subject and object would have no permanent meaning for the language. For simplicity let's temporarily forget about the word "transitivity" and focus on what we know for sure about Klingon. In Klingon, there are some verbs that take only a subject, e.g. Qong, Hegh, vum, Hagh, SaQ, nargh, vang, ratlh, qet, yIt, naj, etc. We could call these subject-only verbs, since they simply do not ever take an object. Indeed, for most of these verbs an object would not even make sense. Now, there are of course the rest of the verbs, which do take objects, e.g. yaj, ghaj, tu', ghoj, lo', legh, laD, tIv, rur, Sop etc. Let's call these object-taking verbs. So far, so good?

Now here is where it gets tricky. It is not always clear what the object is supposed to be. For example, what is the object of ghun? Is it the computer, the program? If you said, 'oH vIghun, would Klingons even be able to guess what 'oH was referring to? Or does ghun simply take no object at all? Is it one of the subject-only verbs? Without further evidence, we do not necessarily know. One might say, "Well, the object of the English verb program is the computer, so the object of ghun must be the same." But is it really that easy? Can we always assume Klingon is just like English? Let's look at another example: Har. Is the object of Har a trustworthy person or a trustworthy fact? For the English verb believe it can be either one. The English verb is flexible. But is that flexibility inherently connected to the meaning of believe? Or is it merely an accidental (though useful) feature of this language? If the latter is the case, then we cannot simply assume that Klingon verbs have the same relations to their subjects and objects (or lack thereof) as English verbs do. Here is another point that serves to demonstrate the accidental quality of language: In English we can say we believe in something or someone. That is rather different from just believing something or someone. But what is it about the meaning of in that actualizes that difference? There is nothing about in that would have allowed a student of English to know the nuance of believe in even after learning the individual meanings of believe and in. It is just an arbitrary feature of that verb. Natural languages are replete with this kind of arbitrariness. That's why they are sometimes called idioms. In general we cannot predict the accidental features of any given language based on the meanings of each isolated word. This idiomaticity extends to verbs and their semantic relation to the grammatical objects and/or subjects. All these "accidental" expressive features of a given language are so vital to proper understanding of it. The idiomatic expressions must simply be learned rotely.

Klingon is not immune from this idiomaticity. Just to demonstrate, we will examine the verb pegh. Obviously pegh has to something do with secrecy. Its gloss in TKD is keep something secret. That seems to imply that the subject is the agent preventing certain information from being known. (If there is an object, it would probably have to be the secret itself or the "hidden" thing.) But from CK we have the example pegh De'vetlh, translated That is classified information or literally That information is kept secret. What's going on here? By the TKD gloss we should have expected De'vetlh peghlu'. The point is that pegh is being used in a way quite different from how it is glossed. Might this be an instance of idiomaticity in Klingon? I would have to at least entertain the possibility, because if pegh really had be secret as its only interpretation (as in the CK example) why would Okrand have glossed it as he did in TKD? Maybe it has alternative relations to its object and/or subject.

Anyway, let's turn out attention to those verbs represented under the Transitivity section of the questionable verbs page. The issue with them is that just because the English verbs are of variable transitivity, it does not necessarily follow that those (and only those?) Klingon verbs should also be of variable transitivity? As I have shown with ghun and Har above, it is too late to hope that Okrand's glosses will be completely trustworthy here. Now is it out of the question that Klingon has verbs of variable transitivity? No, of course not. Many languages have such verbs. But which verbs these are in a given language can be quite surprising. In many languages the verb for want, desire has an intransitive sense meaning be absent, be needed. On the other hand, that same language may have totally different words for the transitive and intransitive senses of bend or drop.

The woman dropped the handkerchief to the ground.
The handkerchief dropped to the ground
The same verb has these two grammatical distinct senses in English, but in another language two totally different words may be required.

The woman wanted a handkerchief.
A handkerchief wanted
A language may allow this literal difference in senses of the verb for want, though English certainly does not.

So even if I were to concede that Klingon might have verbs of variable transitivity, I would never assume that I could always rely on the English glosses to determine that. Moreover, there is something else that supports the idea that Klingon's verbs have fixed transitivity (or if you like, a fixed sense in which they are semantically related to their arguments). Most of the instances of variable transitivity in English are to introduce a causal agent. But Klingon has the highly productive -moH suffix for just this purpose. Why would Klingon ever need to use variable transitivity like English does? Klingon is so different from English in so many other respects. Let us err on the side of caution.

ravDaq ghIchnav chaghmoH be'
The woman dropped the handkerchief to the ground (lit. The woman caused the handkerchief to drop [intransitive sense] to the ground )

ravDaq chagh ghIchnav
The handkerchief dropped to the ground

Of course if it turns out that the object of chagh is the thing that descends, then we would have to use chagh differently: The second sentence would have to add -lu'. That is why we need clarification for many verbs. For some we have already received clarification. For instance, the respective usages of Dub and ghur were ascertained by consistent examples after TKD, especially in SBC.

If anyone still doubts that transitivity is an issue, let me submit to them the prospect that a student of Klingon with a certain native language may very well offer neH ghIchnav as a suitable translation for a handkerchief is needed. Then most Anglophonic Klingonists would be quick to flag it incorrect, and they would have to raise the transitivity issue in doing so. Yet the same Klingonists may keep insisting that a verb like nup is correct with variable transitivity (The water level decreased, The water level decreased the pollen count). If Klingon nup can have both senses, then why not neH as well? Or is Klingon really so closely related to English? The reason that I say we need clarification on these certain verbs is that our only alternative is to rely on our English bias, and that's an alternative that I personally find completely unsatisfactory. It could be argued that because Okrand speaks English natively, all his glosses will be accurate enough for English speakers. But I would assert that Okrand's English bias is the very obstacle to our accurate understanding of Klingon words. An English bias may prevent a field linguist from thinking to test certain verbs for one transitivity or another. (If at any point in this article you object to my use of "transitivity", substitute "place structure" or "respective semantic roles of predicated arguments" or whatever suits you, because that is of course what I am talking about.)

Remember: The subset of verbs that exhibit variable transitivity is completely arbitrary with respect to the choice of language. This becomes clear when you compare any two languages, especially if they are not historically related.

There is yet another reason to remain conservative about transitivity in Klingon. If students of Klingon are made aware of the issue of verb transitivity, then they will be more likely to avoid certain usages which are surely incorrect. For example, some students may learn 'oy' as hurt, as it has that gloss on the English-Klingon side of TKD. The verb hurt has variable transitivity: She is hurting all over, She is hurting him all over. This might lead a student to say something like mu'oy'. But we actually have a clarification for 'oy': ache, hurt, be sore on the Klingon-English side of TKD. mu'oy' is wrong, and the correct version is 'oHmo' jI'oy' or mu'oy'moH. The issue of variable transitivity of English glosses has actually resulted in so many errors that it seems pertinent to raise the issue in general, so that the definite-violations are avoided, and the apparent-violations at least inspire a healthy suspicion.

In the end I would like to be using Klingon in the most plausibly accurate way that I can infer from canon, but I must trust my instincts. Given that Klingon is wholly unrelated to English, I have to at least consider the perspective I have gained from experience in other languages. That experience shows that languages differ so drastically with respect to verb place structure that it is folly to assume correspondences between arbitrary grammatical features in respective languages. Essentially it boils down to this: Just because a Klingon verb X corresponds semantically to an English verb Y, that does not necessarily mean that X should also correspond grammatically to Y.

-- Andrew Strader aka Guido#1

Comment 1

I welcome criticisms and comments on this article, but I do request the following:
  • Use examples wherever possible, even if it seems obvious.
  • Explain uses of linguistic terminology; do not let it disintegrate into a definitional dispute (of a term like "transitivity" or whatever).
  • Respect that this is my opinion, and neither I nor anyone else would consider my word on Klingon to be a final word -- so flaming is totally unnecessary.

For the most part, I agree with this analysis. As I see it, you're suggesting that Okrand was defining the sense of the words, and not necessarily the grammar.

Suppose this isn't the case? There are a number of examples, e.g. pegh, loS, DoH, 'ab, from various sources, that include the English prepositions necessary to indicate the grammatical structure. From this, one may suppose that most of the entries in the word lists indicate both grammar and sense by design. So, just as DoH back away from, back off, get away from shows that the the object of the verb must be the thing that the subject is moving away from, shouldn't a lack of a preposition in the definition (as in Har believe also indicate that the correct object by demonstrating it in its English definition? If Har could mean either "believe" or "believe in," why didn't Okrand write this?

Naturally, those verbs which are flexible in English and which are not further detailed in the definition must be labeled as "unknowns." We don't know whether nup can be used in the sense of The water level decreased, The water level decreased the pollen count, or both. We must seek clarification for these, noting that a single example one way does not necessarily preclude use of it the other way.

There are exceptions to this idea, but not enough, I feel, to invalidate it. qIm, for instance, is defined as pay attention, but it has been used in canon (I can't recall where at this time, but I'm certain it's there) to mean pay attention to. Some words have been used in multiple ways, like pegh as noted above, or mev (compare bIjatlh 'e' yImev with not mev peghmey).

I don't think transitivity is a non-issue, but neither do I think it is completely open to interpretation. Okrand's word lists may be indicating grammatical use as well as the sense of the word. We must look to canon to guide us in understanding and learning how Klingons interpret their words, rather than trying to come up with self-imposed rules on how to look at these words.

-- David Trimboli aka SuStel

Comment 2

teH. I think the problem is just that there are a few words (admittedly, not many), unused in canon, whose glosses are inadequate for actually putting the word into use. I think DIng is one of those.

However, I'd say that since we have jIr rotate and jIrmoH (rotate (something), cause to rotate) , and in the spirit of what Krankor has already written about this idea, DIng is likely to be intransitive: DIng rutlh the wheel spins. He spins the wheel is probably rutlh DIngmoH. Not for certain, but it's likely. (But the appearance of examples like mev clouds the issue, and seems to indicate that variable transitivity might have some relevance. Based upon the idea of mev as in not mev peghmey, one would expect bIjatlh 'e' yImevmoH rather than the bIjatlh 'e' yImev that we have.)

-- RhonaFenwick aka QeS lagh

See also

Category: Grammar    Latest edit: 11 Apr 2017, by MarcZankl    Created: 02 Mar 2014 by LieVen
History: r14 < r13 < r12 < r11 - View wiki text
The Klingon Language Wiki is a private fan project to promote the Klingon language. See Copyright notice for details.