(A page from the Loglan web site.)

(From Lognet 93/1. Used with the permission of The Loglan Institute, Inc.)

Hu Logla Sanpa Toi?
(How Do You Say This?)

by Bill Gober

Lopo melii, Ama malbi, ica ei ba norkalpebmao kicmu vi le hasfa
(The A-disease: Or, Is there an indefinite doctor in the house?)

Loi, logla stude. In this column, we explore how to express the indefinite article ('a', 'an', 'some') in Loglan. I've been promising you this column for some time, but with the title, 'Lo, it's an indefinite article'. I had to change the title because that one contained, well, certain serious inaccuracies—in particular, lo isn't an indefinite article. In fact, as I discovered in preparing for this column, Loglan has no indefinite articles—none.

The best place to start is with page xviii of the 1975 Dictionary. Since few of you have this obsolete dictionary, here is the entire relevant paragraph. (I've silently changed the Loglan specimens into their current equivalents.)

3.7 Translations of the Indefinite Article: As Loglan has no indefinite article, the problem of translating English 'a' or 'an' before Subjects or Objects arises. (Articles in English Verb Phrases require no special translation; thus 'is a man' = mrenu.) Four translations of 'a' or 'an' are usually possible:(1) by a non-designating variable (ba, be, etc.), (2) by ne ('one') followed by a description (ne le mrenu = 'one of the men'), (3) by the mass descriptor lo (lo sinma = 'the movies') or (4) by the definite descriptor le ('the'). Thus, in English sentences in which 'a' or 'an' indicates genuine ignorance on the part of the speaker about the identity of someone or something—as, for example, in 'A man was here', spoken on smelling cigar-smoke in an empty room—translation with ba is best: Ba pa mrenu vi = 'Something x was (being) a man here.' If however, the speaker knows something about the identity of the designated person or object—e.g., that i[t] was one of a definite group, as in 'I ate a cake'—translation with ne is best: Mi pa titci ne le grato = 'I ate one of the cakes.' On the other hand, if 'a' or 'an' indicates indiscriminateness—as in 'Let's go to a movie', spoken by someone who doesn't care which—translation with lo is best: Ea mu godzi lo sinma = 'Let us go to the mass individual composed of all the movies there are', i.e., to the movies. Finally, 'a' or 'an' is sometimes used in English to mark a designation of a once-definite individual who may now be hard to locate, as in 'A dog bit me.' Here translation with le is the best we can do: Le kangu pa ditka mi = 'The dog bit me.' (Ie le kangu = 'Which (the) dog?'; Le kangu ji pa darprano = 'The dog that ran away.') All these circumlocutions are, of course, more precise in Loglan than the English expressions they translate. They are, therefore, inexact translations. As always when translating English into Loglan one must first decide what the English expression "really means."

Pages 216-218 'Indefinite Reference' in Loglan 1 introduce what I'll call the fifth option: (5) by a numeric descriptor. Ne mrenu pa hijra used to be illegal; now it's legal, with the meaning 'Some one member of the class of men was here'. Similarly, Ro mrenu pa hijra means 'Many members of the class of men were here'. Such usages are more compact and feel more "natural" than the Option (1) equivalent that uses a non-designating variable; e.g., Neba mrenu, e pa hijra. Ne grato is more compact (and more to the point) than its Option (2) equivalent, Ne lo grato.

Some people—not least, Prof. Brown—consider that Option (5) has added genuine indefinite articles to Loglan. Unfortunately, the "indefinite articles" that it makes possible are all quantifiers... their primary job is to tell 'how many'. Ne kangu pa ditka mi is vaguer about which dog than is Le kangu pa ditka mi, but at the cost of becoming specific about how many dogs.

I've noticed a lot of ill-conceived use of Option (5) to translate indefinite articles. The uses are not so much bad, as unnecessarily anglophile, i.e., they tend to mark you as an English speaker. I'll call these usages 'The A-disease'.

In the fall of 1991, James Jennings sent me an example of the A-disease. He was translating, 'Time flies like an arrow. Fruit flies like a banana', and he had come up with this: Lo ckemo ga fleti lia su bortcidi. I lo fruta flaki ga fundi su banhane. This shows two classic features of the A-disease. (1) Mass or plural indefinite references (i.e., those with 'some', or unmarked) are unerringly translated with lo. (2) The speaker wants to translate singular references (those with 'a' or 'an') with ne 'one'; sometimes, 'one' doesn't seem big enough, so the speaker uses su 'at least one, some'. As we'll see, probably the most Loglandical translation uses lo for every one of the articles: Lo ckemo ga fleti lia lo bortcidi. I lo fruta flaki ga fundi lo banhane 'Time flies like arrows (bow-food). Fruit flies like (are fond of) bananas'.

I've encountered other logli showing symptoms of the A-disease. (I escaped the A-disease, although I'm still recovering from a related syndrome: I have trouble choosing correctly between le and lo. More on this later.) Enough logli, in fact, that I decided I had material for a column.

I've been slowly raising the curtain on this column. In LN92/2, I had the sentence Mi pazda lo taksi 'I wait for a taxi'; I hoped it would be thought-provoking for you. In LN92/3, the narrator in my story suffered the A-disease; n had the following exchange with Ioséf the barkeep: (simplifying a bit) "Ao mi plizo ne telfo lepo frekra ne taksi." "Mi fa frekra lo taksi dii tu."

Then, last August, I convened a conference on Internet and CompuServe to discuss indefinite reference. We spent several weeks on an electronic street corner, discussing how to wait for taxis; occasionally, someone would duck into a stationer's shop, so that we could talk about buying pens and pencils. (Prof. Brown was the most daring of us all: He enjoyed ice cream, movies, and the company of women!) All flippancy aside, the exchanges were invaluable (and not just to me). I'm grateful to everyone who contributed: Gary Rector, Richard Kennaway (in the UK), Colin Fine (UK), Chris Handley (New Zealand), Armando Ramos (Spain), Bob McIvor (Ontario), Randall Holmes, James Cooke Brown, James Jennings, and Wes Parsons.

Instead of summarizing the discussion, I'll excerpt key points from it, in chronological order. Not only does this save me work, but it also hints at the fascinating evolution of the discussion. (But it no more than hints at the discussion. The whole discussion, all 38 contributions, would fill this entire issue of Lognet!) (I suggest you re-read L1 pp. 186-188 and the associated note before you start reading the discussion....)


Gary Rector: Since the old dictionary gives 'any one of...' as the meaning of ne, I think we ought to say, Mi pazda ne taksi.

Richard Kennaway: Maybe, but I feel that the oneness derives from the English expression 'a taxi', rather than the other way round. In Loglan, Mi pazda lo taksi seems to me a very natural way to say that I am waiting for a taxi.

GR: I would like to say that my choice of ne in Mi pazda ne taksi was not influenced by the English usage of the indefinite article here but by the fact that generally when I'm waiting for a taxi, I'm waiting for exactly one taxi.

Bob McIvor (RAM): Gary expressed exactly my objections to lo taksi. As I told Bill once, I have a friend that is a bus fan, and when I was in Britain I saw many boys that were train fans, and went to the tracks with notebooks etc. to pazda lo tren[a]. This could be translated as 'a train' in English, but more likely as 'trains'.

RK: Supposing that lo tren[a] should only mean concepts that would be expressed in English as 'trains', or supposing that 'trains' and 'a train' must be expressed differently in Loglan, is just gleco penso. The singularity of 'a train' need not be expressed, and the more obvious it is from context, the less need there is to express it.

Randall Holmes (MRH): I'll put my opinion on record; I think that one can construe waiting for a taxi as waiting for lo taksi to manifest itself.

JCB: Now we come to a matter of personal judgement. Having played this eerie game for some years, I am personally convinced that that is exactly what I am doing when I am waiting for 'a taxi'...or going to 'the movies'...or liking 'ice cream'...or enjoying the company of 'women'. I am waiting for, going to, liking, and enjoying the company of, respectively, some manifestation in my experience of all the taxis there are, all the movies there are, all the ice cream there is, all the women there have ever been. In fact, that is precisely what I am doing when I am waiting for a taxi! I am waiting for an appearance out of the mist of this mass individual. And interestingly enough neither Mi na pazda ne taksi nor Mi na pazda su taksi—and certainly not Mi na pazda le taksi unless I called one!—gives anything like the right spin on my meaning. For at the moment, when I am actually waiting for one, I am totally uninterested in the fact that taxis can, under other circumstances, be lined up in ranks and be counted.

James Jennings: Gary is waiting for one taxi. However, Bill is 'waiting for taxis'. This seems to emphasize the waiting, as if you had said La Bil, durzo lopo da pazda ne(lo?) taksi.... Bill is focused on the waiting, not on the expected taxi.

JJ: I just noticed th[at] the list of non‑numerical quantifiers (L1, pg 214) has nothing for 'an undetermined number of'. I've only noticed the lack by noticing that we don't need it. An unmarked le works nicely. What really amazes me is that I never noticed this before. It seems so obvious in hindsight. (Did everybody know about it but me?) [James asked me to add the following note: "It is better to say that le describes an 'unspecified' number rather than an 'undetermined' number."]

GR: Now I see that we should use lo when we don't care whether we're talking about a mass of things or an individual manifestation, when it doesn't matter how many whatevers are involved or even whether they can be counted or not. If that is correct, then of course I would prefer to say, Mi pazda lo taksi. I chose ne before because I thought Loglan gave me no way to express what I would call the bare noun rather than the mass noun. (I think of it as bare because the unadorned noun in Korean—and in Japanese and Chinese, for that matter—works the way [JCB] described the Trobrianders' nouns: all nouns are like 'furniture' or 'equipment' or 'soap' in English, and if you want to talk about a specific number of them, you have to use an appropriate counter word with the number rather than just putting the number directly before the noun.)

Bill Gober (WGGJ): I'm having trouble getting used to using le for plural objects, too. It really seems to me that le is the general‑purpose definite article, while lo is the matching indefinite article—and neither of them contains any number designation.

MRH: Thus, I prefer lo taksi—he is actually waiting for a polyp of taxidom, not any individual taxi.

MRH: Both le and lo are definite articles. Loglan has no indefinite articles other than numbers (in the most general sense). The best translations of 'a/an' are ne or su.

GR: In English we use 'some' as if it were a sort of indefinite article for plurals and uncountable nouns, without being concerned about what quantity is actually involved. In answer to the question "What did you buy?" we might answer "I bought a pencil," "I bought some pencils," or simply "I bought pencils"; or if we bought something "uncountable," we might more likely use the noun alone, as in "I bought beef and pork," though we might also say "I bought some beef and pork" with no difference in meaning. I don't see why lo shouldn't be the Loglandic descriptor of choice in all these cases.

JCB: Thus, an L speaker who said to me Mi pa furvea lo pinsi would be telling me he had a brush with the Mr. Pencil and bought a piece of him. I might think that a slightly peculiar way of speaking—even suspect, possibly, that there was a Trobriander lurking behind his loglandian garb—but I would certainly understand his sentence. I might, however, understand this possible Trobriander's report with an even higher degree of semantic precision—would think of him as exercising the 'option of clarity' rather elegantly, in fact—if he had said 'Mi pa furvea piri (point‑few of) lo pinsi', that is, a small portion of this presumably massive individual is now what my interlocutor is telling me he had purchased. Right, the (admittedly implausible) ambiguity between his buying little, much, or all of it is now resolved.

MRH: This is a problem with English usage, not Loglan usage. One is not waiting for one taxi when one is waiting for a taxi. English cannot be translated literally into Loglan, that is all. Try saying 'I am waiting for one taxi (I don't necessarily know which one)' in English. This is the literal sense of Mi pazda ne taksi. The meaning of the Loglan is quite direct, but the naive literal English translation is deceptive. There are plenty of places where ne (su is usually better) translates English 'a' fairly well.

JCB: I believe it was someone else who remarked, in this long interchange about taxis, that lo was "the L indefinite article"; and presented in L1 as such. Of course it isn't and wasn't. Lo sometimes translates the E indefinite article satisfactorily—as it does in 'Let's go to a movie' or 'Let's go catch a thief!'—but that doesn't make lo an indefinite article, or indeed, an indefinite designator of any kind.

JCB: As Randall also suggested, le, ne and su are often perfectly sensible alternatives to lo. This is especially true when the speaker is talking about something that has already happened.

RAM: Nobody seems to be taking up the use of loe.

JJ:Loe seems like a special purpose word, used only when you really want to stress the typicalness of the object. I don't care if the taxi is typical or not. If it takes me where I want to go in reasonable comfort and for a reasonable fee, it can have a sod roof for all I care.

MRH: Don't get too involved with the apparent helpfulness of temporal concepts in the taxi problem. The meanings of quantifiers have nothing to do with time.


Unfortunately, I can't round out this column with a couple of good rules of thumb; I can't give you crib notes that cover the topic exhaustively. For the A-disease isn't a speech issue: it's a world-view issue. If for 'I'm waiting for a taxi' you say Mi pazda ne taksi, you've communicated your meaning clearly and correctly. You've also communicated an obsession with one taxi that a Trobriander—and, I like to think, a native logli—would find odd.

Or, as Prof. Brown put it in one of his contributions to the discussion: "It is, in short, a question of mood, of how one means to enjoy—or at least experience—the world. Trobrianders do it one way; counters do it another way; we logli ought to be able to do it both ways. Which way are you going to do it when you are waiting for a taxi? Or enjoying the company of women?"

Okay, so I do have a piece of advice to round out the column: Remember the passage from the 1975 dictionary. And think carefully before you exercise Option (5)!

I haven't determined the topic of my next column yet. I'm planning to convene an electronic conference to help me pick one. So, if you have a topic you'd like me to cover, get on the Net and make your preference known!


Copyright 1993 by The Loglan Institute, Inc. All rights reserved.

Send comments and corrections to:

djeimz AT megaseattle DOT com