(Originally appeared in Lognet 99/1)
I know some of you have been expecting me—as the author of mia—to say something useful about the possibility of a “dual subjunctive” for some time. For I’ve often hinted that the topic needs discussing, and that it was an open issue for me. So I am genuinely sorry to have deferred my remarks on this topic for so long.
Leaping to the conclusion first, let me say that, now that mia has been accepted by the К (Keugru), I am more than willing to say aye to adopting a second subjunctive form as an alternative to mia, one that would be grammatically, and on most usage questions, identical to mia, but semantically distinct from it. I am hopeful that this move on our part, when and if we make it, will satisfy our logicians’ sense that we need a stronger translation of the NL (natural language) subjunctive than mia provides, an alternative conditional that will presumably be used by people who really are claiming something about the external world, even if conditionally, when they use E (English) were/would and corresponding forms in other NLs.
The critical moment for me, on this long-lurking issue, came when I was riding with Alex in his car in the Auvergne. [As described in LN 97/1, p. 16, c.2] Alex and I were approaching a “Y”-junction along one of its “ascending arms”—the left one, from the standpoint of our approach to it—and all 3 branches of the “Y” were utterly empty of traffic except for our own vehicle. ‘Is it still true’, I think I said as we approached the “Y”-junction from the left, remembering my own experiences in Paris some decades earlier, ‘that a car on the right...’ (gesturing to the empty road behind and to the right of us) ‘...still has the right of way?’ ‘Yes,’ said Alex, ‘if there were acar on that road, it would have the right-of-way.’ I noticed A’s use of the subjunctive—as I always did in peoples’ speech, those days. ‘A clear case,’ I murmurred to myself; ‘of the non-subjective use of the subjunctive.’ Then to Alex: ‘For surely you are reporting the laws of France here, and not just something going on inside your head?’ He agreed he was; and I went on to say that, in cases like this one, I was myself strongly inclined to use the indicative, to say not only in L (Loglan) with nacevi but also in E with when-and-where: ‘When and where two French vehicles are approaching each other at any angle less than 180 degrees, the one on their respective rights has the right-of-way.’ Nothing subjunctive about it, in my opinion. It is simply bad usage—in both L and E, I’d argue—to use the subjunctive to make such hard, factual, even though conditional claims, or to report any truths, for that matter, about the external world with the subjunctive. On the other hand, I do agree with Alex that the use of the subjunctive to make even testable claims conditionally is strongly habitual in many dialects of E and other NLs.
I have for many years tried to cure myself of this bad habit, especially in expository writing. For example, if I catch myself saying in the JM (Job Market) book ‘Under <certain circumstances> the job market would have <certain effects>’ I promptly change ‘would’ to ‘will’, the weak subjunctive to the stronger predictive form: ‘Under <certain circumstances> the job market will have <certain effects>’. If that prediction needs moderating, I can always do so with ‘probably’ or even ‘perhaps’, where either is justified. But the important thing to realize is that, as an inventor, I’m precisely in the business of making predictions; so to use the subjunctive in such expository writing is to duck my responsibilities and to mislead the reader into thinking that I’m writing about some imaginary “never-never land”. I’m not; I’m talking about the real physical world. So it is bad form to use the mia-type subjunctive wherever one is confidently talking about events that one believes will take place—though conditionally—in some possible future state of the real world...though it is one that may never be realized.
How different this is from O’Leary’s saying that “if Casey had lived 'til next Thursday, he’d have been dead a month"! That amazing little verbal invention is, in my opinion, the best illustration I have yet come across of the imaginative use of what I think of now as the true subjunctive, the one that warns the listener/reader that he is being told something about some imaginative goings-on in the speaker/writer’s head.
Now if I were on a desert island thinking about such Loglandical problems, I would simplу say to myself: ‘Ah! That’s still another case where NL blurs an important logical distinction. In L, we must do better than that. We must use mia and kin only when we are making subjective claims—claims about the imaginary worlds we create in our heads—and use the readily available indicative when we have something to say about the external world. And that would be the end, I’m afraid, of my insular thinking, my thinking as an empiricist on an island called Empiricon on which all inhabitants are also empiricists.
But I’m not living on an island with fellow empiricists as my sole companions. I am living in a Loglandian community that is, fortunately, diverse enough philosophically that I am intimately aware, now, that at least three of my nearest logli neighbors are platonists! And platonists, god bless them, really do think they are making hard claims—though they are claims that remain difficult if not impossible to test—when they use the subjunctive to report features of their “ideal worlds”. These platonic worlds of theirs are not just creatures of their imaginations, they argue. They are to be treated as real places—logically possible places, as they are also inclined to tell us—about which they have “real knowledge” to impart. And, understandably, having discovered that an inventive logician named ‘Kripke’ shares their platonist convictions, they now want to use the forms of Kripke logic to do so.
Good. What we are designing, here—actually, what we are constantly improving by expanding—is a language, and not a device for conveying catechisms among cosy little groups of true believers. So we must provide, for these platonists of ours—and for others like them with similarly special epistemic needs—a third alternative...third, that is, to mia and necevi. But what is abundantly clear is that this alternative, if adopted, must be clearly distinguished from the other NL senses of the subjunctive. In particular, it must never be confused with the mia sense, which is, I’ve observed, the sense in which most people use the subjunctive on most of the NL occasions for which they use it.
Still, when I consider this problem in my ecumenical role as a member of the К, I find myself saying, ‘But we are building a logical language, one to inspire, canalize, make easy and efficient, and so promote, “good thinking”. And isn’t it in fact good thinking to use the indicative mood and not the junctive in these few “right of way” cases?’ ‘Yes,’ I am obliged to say, ‘it is. And I will personally continue to try to rid myself of this misleadingly objective use of the subjunctive—in both E and L—and to encourage others to do so, as well.’
But the more generous and more worldly part of me rebuts the strongly moralistic position I’ve just expressed. The language must also accommodate every kind of belief system, including platonism and theism and polytheism and animism and everything else people have thought and may still be thinking on this planet, about which they may still want to make claims. We must be able to say Four of the three men came, as well as There are exactly three gods and all of them are green. So I’ve moved to the К that we adopt a “2nd subjunctive” even though I believe it leads to intellectual error, and so will privately discourage its use in L by using the indicative myself, by using nacevi (when-and-where), in short, on all these “quasi-subjunctive” occasions when some objective truth—like who’s got the right-of-way in France—is conditional on there being vehicles about to bring it into play.
The K adopted my motion and we then discussed what the non-mia subjunctive should be phonemically...apart from being, like mia, cvv in shape. The foi derivative of fikco, which is in any case not really appropriate, was ruled out as busy, and this left the three words forma, lodji and vidre as possible sources of a cvv-form derivative. Each member of this triplet does suggest one semantic dimension of our platonic operator; for the kind of truth going on here may legitimately be thought of as “formal”, “logical”, or “idea(l)”...although not in the usual sense “factual”.
Of the cvv forms still available for these words, vidre turns out to be the most productive, /vie/ and /vio/ are still available, though the other three /viV/-forms are all assigned. I rather like vie, don’t you? It virtually commands the recall of vidre; it can be nothing else; and vidre (idea) evokes the long platonic tradition that only ideas are “real” (whatever that means, I am forced, as an empiricist, to ask).
By the way, we don’t need a “yank-back” companion for vie, any more than we need one for mia or pa. The question of the discursive scope of pa words has still not been exhaustively studied, and whatever we find works for pa and mia will, I expect, work for vie as well.
—Hue Braon Djim
Copyright © 1999 by The Loglan Institute. All rights reserved.