These are the LE + PO and LE + ZO (see PO and ZO) compounds that make arguments out of sentences. They are recognized as LEPO words by the preparser. LEPO words are known as event operators, although they designate properties (for example, with lepu) and quantities (with lezo) as well. Their grammar is quite different from that of LE-words, which they resemble morphologically, in that LEPO-words but not LE-words take whole sentences as operands. Sometimes it looks as if the operand of an event operator is nothing more than a predicate expression—e.g., lepo sucmi = 'the swim'—but this is because the operand is here the simplest of all sentences, a one-word imperative. The event of swimming could be fully specified: lepo da sucmi de di = 'X's swim to Y from W'. So unlike the operands of ordinary descriptors, the operands of event operators may be embellished without limit: Lepo lemi brudi ji la Djan, pa kamla lepo fomtitci, e stolo mu ne nirne = 'The time my brother John came to dinner and stayed with us for a year.'
Notice that the events, properties or quantities described with event operators may be particular (lezo da pa sucmi = 'the amount X swam') or general (lopo sucmi = 'swimming'), or general but constrained in various ways: e.g., lovipo sucmi = 'The here-swimming (the swimming here)' and lenapo penso = 'The now-thinking (the thinking now)'.
Loglan event descriptions are philosophically of some interest. Their designata are the objects of thought, for example; and so event descriptions accomplish what in the Indo-European languages is accomplished by so-called "indirect discourse". Thus, 'He thought that it was going to rain' is translated into Loglan with an event description: Da pa jupni lepo ti fa crina = 'X opined the event of this (place) being later rained-on.'
Lexeme LI: The Left Quotation Operator
LI is monolexic and is the opening mark for ordinary, or "weak" quotation. Li is used with the close-quote lu to quote grammatical Loglan, usually in strings of two or more words; see LIE and LIU for other varieties of quotation.
Lexeme LIE: The Strong Quotation Operator
LIE is also monolexic; it is the leading mark of "strong" quotation. That is, it is the mark by which non-Loglan strings or even strings of malformed Loglan may be unequivocally quoted. Lie is used with two instances of any arbitrarily chosen word X which are placed at each end of the string to be quoted. The boundary marker X is usually a letter-word (see TAI) and each instance of it must be separated from the quoted string by a pause: lieX, , X. /.X/ must not of course appear adventitiously within the string being quoted.
Lie-quotation is most often used to quote foreign language strings. When it is, the upper-case Loglan letter-word for the Latin letter which is initial in the Loglan predicate for that particular foreign language is used conventionally as the boundary marker X: LieSai, No hablo Español, Sai. Sai is the sign of spana, the predicate meaning 'is Spanish/an element/expression of the Spanish language'.
Lexeme LIO: The Number Designator
LIO is the monolexic number designator. Lio is used before mathematical expressions to designate numbers. For example, Lio ne natra numcu = 'The number one is a natural number' and Ti langa ta lio topifemei = 'This is longer than that by 2.5 meters.'
Lexeme LIU: The Single-Word Quotation Operator
LIU is used to quote single words. It is monolexic and used only before single, well-formed Loglan words to designate them. For example, Liu liu logla purda = 'The word 'liu' is a Loglan word'. See LI and LIE for other varieties of quotation.
Lexeme LU: The Right Quotation Operator
LU is monolexic and is the closing mark for ordinary, or "weak" quotation. Lu is used with the open-quote li to quote grammatical Loglan, usually in strings of two or more words; see LIE and LIU for other varieties of quotation.
Lexemes M1* through M11*
Ml is a machine lexeme inserted by the preparser before any instance of A that is connecting linked arguments.
M2 is a machine lexeme inserted by the preparser before any instance of KA that is connecting linked arguments.
M3 is a machine lexeme inserted by the preparser before any instance of KA that is connecting predicates.
M4 is a machine lexeme inserted by the preparser before any NO that it is negating a modifier.
M5 is a machine lexeme inserted by the preparser before any A that is connecting sentence modifiers.
M6 is a machine lexeme inserted by the preparser before any A that is connecting argument modifiers.
M7 is a machine lexeme inserted by the preparser before any PA that is inflecting a predicate.
M8 is a machine lexeme inserted by the preparser before any NO that is negating a marked predicate, as shown by PO, ZO or M7.
M9 is a machine lexeme inserted by the preparser before any ACI that is connecting predicates.
M10 is a machine lexeme inserted by the preparser before any A that is connecting predicates.
M11 is a machine lexeme inserted by the preparser before any AGE that is connecting predicates.
Lexeme ME: The Predifying Operator
ME is the monolexic operator that turns any designation into a predicate ("predifies" it). Prefixed to the first word in that designation, me creates a predicate with a meaning, often vague, that can be associated with that designation. Some examples: Ta metu cadre = 'That's a you-ish dress'; Plizo le mela Kraislr = 'Use the Chrysler'; Da mela Nuiork = 'X is a New Yorker'; De mele sitci = 'Y is of the city (citified? cityish?)'.
Such usages create "false compounds", i.e., forms like metu, mela and mele which are recognized as not being members of single lexemes by the preparser, as true compounds always are, and so are decompounded. Thus, metu is lexed as. the two-lexeme string ME DA before it goes to the parser; see DA.
Lexeme NI: Quantifiers
The quantifier lexeme NI has in principle a limitless number of allolexes, since it contains not only the digits ni ne to te fo fe so se vo ve ('none' 'one' 'two' 'three' 'four' 'five' 'six' 'seven' 'eight' 'nine'), but all the compounds of whatever length that might ever be made from them. Thus, netotefofesosevove as well as neni ('ten') is an allolex of NI.
NI also includes ho, the interrogative quantifier ('How many?' or 'How many of…?'); thus Ho da pa kamla = 'How many of X came?' It includes three of the non-numerical quantifiers as well, namely re ro ru ('most' 'many' 'enough'); the other two r-words, ra ri ('all' 'few'), having certain compounding duties to perform—they make the cardinal and ordinal compounds that will be classified as PREDA—must occupy a separate lexeme, namely RA.
Other allolexes of NI are sa si su ('approximately (all)' 'at most (one)' 'at least (one)'), which may either be prefixed to other NI-words (savo = 'approximately eight' sivo = 'at most eight' suvo = 'at least eight') or used alone with their default values understood (sa = 'nearly all', si = 'at most one', su = 'at least one'). Finally, there is a series of mathematical expression-building operators, not yet complete, consisting of the decimal point pi [.], the double-zero ma , the triple-zero mo , the division sign kua [/] (read 'over'), the inverse division sign kue (not available on this font, but read 'under'), the addition sign poi [+], the subtraction sign nio [-], the negation sign niu (not available), the multiplication sign tia [*], the power sign sua [|] (read 'to the …-th'), the left and right parentheses kie and kiu [( )], which in this context, may often be read as 'the quantity', and for more complex nesting, the left and right brackets gie and giu [[ ]], also often readable as 'the quantity'. Mathematical expressions formed with these operators—for example, tokuafo [2/4] and tosuaniufo [2|-4] (read '2 to the power minus 4')—are also NI words.
NI has another group of allolexes which are formed of some numerical expression plus one or more letter-words of the TAI-Lexeme, e.g., nevepinifomeimei [19.04mm]. These are the dimensioned numbers; they, too, as well as the TAI words themselves, may be components of mathematical expressions. For example, consider tobeisuato [ 2b|2 ] = 'two-bee-squared' vs. tobeikiusuato [2b)|2] = 'two-bee-the-quantity-squared'. The "grammar" of such expressions, i.e., the set of compounding rules for NI compounds, has not yet been written.
A final type of NI compound may be formed by prefixing ie (see Lexeme IE) to any other NI word, simple or compound, e.g., ieto or ietobeigiusuato. Such quantifiers have the sense of 'which…of…?', as in Ieto le mrenu = 'Which two of the men?'…or, indeed, as in Ietobeigiusuato le protoni = 'Which two-bee-the-quantity-squared of the protons?' The internal structure of such compounds may be as complex as the speaker or writer likes; but grammatically, such expressions count as just another quantifier.
There is one usage convention involving NI and TAI words that resolves an ambiguity that cannot be solved be preparsing. This is the apparent ambiguity of the expression nei preda. Is (i) nei a quantifier, in which case this is an indefinite description meaning 'n predas'? Or is (ii) nei a letter-variable, in which case it means 'n is a preda'? Consider that the speaker has already established that the letter word nei [n] in some mathematical expression is to represent a certain large unknown number. May da then lift it out of that expression and use nei as a quantifier, saying, for example, *Nei mrenu pa kamla (interpretation (i))? Da may not; the parser would declare this expression to be unparsible (and so I have starred it). What is happening here is that Nei has been lexed as a TAI word (interpretation (ii), which happens to be the correct one); but so-lexed, the expression will not parse. Standing alone, as it does, Nei will not be parsed as a NI word, i.e., a quantifier. Whatever da's intent, da's auditor will hear Nei as a letter variable, presumably as a replacement of some longer argument in which 'n' figures; because that is grammatically what Nei is. So what da has said would translate into the equally ungrammatical English non-sentence '*Enn is a man came'. In other words, da would like to use nei as a quantifier, but may not because any well-behaved Loglan parser will treat it as an argument.
The way out of this dilemma is to adopt a certain usage convention. We may use the dimensioned number nenei [1n] whenever we would like to use the number represented by nei [n] as a quantifier. Nei is not a quantifier; but nenei [1n] is. Moreover, this usage is mathematically correct. The number one is the identity operator in this group; so 1X = X for any X. So it doesn't matter mathematically whether we say nei or nenei. But it does matter grammatically; for only by using nenei will the parser behave in the way that we intend. To English-speakers, this usage has a certain cost. It will be a trifle awkward, at first, to have to say Nenei mrenu pa kamla ('One-enn men came') when one feels, following English usage, that one ought to be able to say more economically *Nei mrenu pa kamla. But later on I trust we loglanists will discover the satisfaction of speaking precisely and so, understandably. Besides, by observing this convention we preserve the much more important use of TAI words as anaphoric variables; see TAI below.
Lexeme NO: The Negation Operator
There is just one negative in Loglan and it is used in all of the ways that the various negatives of the natural languages are used. No is very widely distributed in Loglan grammar, occurring in 13 rules. In particular, no occurs (i) in R28 where it is used before predicate words and other predicate units (Da bilti no nirli ckela = 'X is a beautiful non-girls' school'); this is the close-bound negative that we often express with 'non-' in English. It occurs (ii) in R34 where it is used to negate kekked predicate units (Da bilti no ke botci, e nirli ckela = 'X is a beautiful not both boys and girls' school', that is, a school for beautiful boys or beautiful girls or beautiful things that are neither boys nor girls, but not for both beautiful boys and beautiful girls; the scope of this no is confined to the kekked pair of predunits ke botci, e nirli). No is used (iii) in R62 before modifiers (Titci no na = 'Eat (but) not now') where the preparser will have marked it with a leading M4. It is used (iv) in R64 where it negates kekked modifiers (Titci no ke vi, e na = 'Eat (but) not both here and now'); the preparser will also have marked this no with M4. It is used (v) and (vi) in R134 and R143 to negate tensed or otherwise marked predicates (Da no pa mutce gudbi takna = 'X was not a very good talker'); this is long-scope no, the tense operator having spread the scope of no over the entire predicate expression. The preparser will have looked ahead and found the pa or other scope-extending mark on the other side of no, and then inserted M8 before the no to tell the parser that this is a long-scope negative of a predicate it is dealing with. No is also used (vii) in R149 to negate identity predicates (Da no bi de = 'X is not equal to Y'); and (viii) in R151 to negate kekked predicates (Da no ke gudbi takna, ki bilti ckela = 'X is not both a good talker and a beautiful school'; here the scope of no covers the entire kekked predicate). It is also used (ix) in R159 to negate statements (No da bilti ckela = 'It is not the case that X is a beautiful school'); as suggested, this no has the sense of the logician's 'It is not the case that…'. No is used (x) in R161 to negate kekked sentences (No ke da bilti ckela ki de gudbi takna = 'It is not the case that both X is a beautiful school and Y is a good talker'. It is used (xi) alone in R168 where No is a fragmentary utterance, perhaps an answer. No is used (xii) in R183, where it is used with a "gap" (a comma or a gu) before any utterance, including fragmentary ones (No, na la Nemen = 'No; in January'); in such uses no often has the sense of being one of several fragmentary answers being made in the same utterance. Finally, no occurs (xiii) in R184 where such leading negatives may be preceded by free modifiers (la no, na la Nemen = 'Certainly not; in January').
NO also occurs in /noV/ compounds (noa, noe, etc.) that are some but not all of the A words. In this use it is rather like NOI (see next lexeme) which also combines with vowels and other elemental lexemes to produce A and other kinds of connective words.
*Lexeme NOI: The Negation Suffix
This lexeme does not occur as a separate word. It is used only by the preparser to detect some sorts of A, CA, ICA, KA and KI words, namely those that end with the suffix -noi For example, the A words anoi and enoi, the CA words canoi and cenoi, the I words Icanoi and Icenoi, the KA words kanoi and kenoi, and the KI word kinoi all require that NOI be recognized by the preparser, in particular, by that portion of it that analyzes compounds, in order that they be assigned to their proper lexemes.
Lexeme NU: Conversion Operators
The simple allolexes of this lexeme are nu fu ju, the three conversion operators which in Loglan effect the passive voice and a great many other "place-shifting" constructions as well. Thus, nu exchanges the meanings of the 1st and 2nd places of a predicate (Da nu bloda de = 'X is hit by Y'); fu, of the 1st and 3rd places (Da fu ketpi de = 'X is the departure-point from which ticket Y is valid'); and ju, of the 1st and 4th places (Da ju ketpi de = 'X is the carrier on which ticket Y is valid'). NU also includes compounds like nufu and nufuju, and the series of numerically marked conversion operators nute (= fu), nufo (= ju), nufe, nuso, etc., an alternative series of conversion operators that accommodate extremely long place-structures. All these words are recognized by the CPD-lexer.
NU words may also be used in incomplete utterances: Da nu bloda = 'X is hit', Da fu ketpi = 'X is a departure-point of some journey for which tickets are required', and Da ju ketpi = 'X is a carrier on which tickets are required.'
Lexeme PA: Inflectors/Adverbs/Prepositions
PA is a "portmanteau" lexeme in that its allolexes belong to 5 semantically distinct series: tense operators, location operators, modal operators, causal operators, and the predicate marker ga. These very different kinds of words are members of the same lexeme not because their meanings are similar but because their grammar, though not their usages, happens to be identical.
As the title of this lexeme suggests, any of the PA words may be used in three grammatical contexts: (i) as an inflector of some predicate (Da pa titci ta = 'X ate that'); (ii) as an adverbial modifier of the main predicate expression of the sentence (Da titci ta pa = 'X eats that earlier') or, when linked to an argument by a JI word (q.v.), as a local modifier of that immediately preceding argument (Da titci ta ji pa = 'X eats that earlier one, i.e., that one that was before'); and (iii) as the prepositional head of a phrase or a clause which modifies the main predicate expression (Da titci ta pa la Yen = 'X eats that before nine' or Da titci ta lia lo horma = 'X eats that like a horse') or, if linked to an argument, then as a local modifier of that argument (Da titci ta ji lia lo horma = 'X eats that (thing) that is like a horse').
If the prepositional meaning of a PA word is known, then its adverbial meaning can generally be inferred by regarding the adverb as an ellipsis of a phrase or clause headed by that same word used as a preposition or conjunction. Thus, we may infer that Da titci ta vi = 'X eats that here' is short for Da titci ta vi ti = 'X eats that in/at this (place)'; hence that the sentence without ti means 'X eats that here'. Similarly, the inflectional meaning of a PA word may usually be inferred by constructing a phrase using that word as a preposition and some designation of the place or time of speech as the object of that preposition. Thus Da pa titci may be regarded as shorthand for Da pa ti titci - 'X, before this occasion on which we are speaking, eats'; hence 'X ate'.
So much for the three contexts in which PA words may appear. We may now describe the five semantic types of PA words:-
Tense Operators: These are pa na fa, the simple past, present and future tenses, respectively, and their numerous derivatives and compounds. For example, there are the continuous tenses pia nia fia ('was …-ing' 'is …-ing' 'will be …-ing', which yield 'since…' and 'ever since' for the prepositional and adverbial senses of pia, respectively; I shall let the reader work out the corresponding senses of nia and fia. There are the habitual tenses pua nua fua ('habitually …-ed' 'habitually …s' 'will habitually …'). There are the intervalized tenses pazi paza pazu (the immediate, intermediate and remote past, for example) as well as fazi faza fazu and even nazi naza nazu with their analogous meanings. There are the nine narrative or compound tenses, the first element of each compound relating the point of speech to the reference point in the narration…often called the "narrative present", the second relating the reference point to the event. The first three of the narrative tenses are the so-called "perfect" tenses of concluded action papa napa fapa ('had …-ed' 'has …-ed' 'will have …-ed'); the next three are the "progressive" tenses of predicted action pafa nafa fafa ('was going to…' 'is now going to…' 'will be going to…'); the last three are the less often used but equally interesting tenses of "coincident" action—coincident because the reference point and the event coincide in these tenses—pana nana fana ('was then …- ing' 'is just now …-ing' 'will be …-ing'). Beyond the narrative tenses are the quantified tenses, of which rana rona rena rina suna tona nena nina ('always' 'frequently' 'usually' 'rarely' 'sometimes' 'twice' 'once' 'never') are just a few of the many quantified possibilities which the loglanist is invited to explore. Various combinations of the above are also possible, e.g., papia ('had been …-ing') and pazufazi ('was long ago going to … immediately'). There are the connected tense words such as pacenoina ('no longer' or 'once', literally 'before-and-not-now'). Finally, any PA word may, of course, be incorporated into a descriptor via the LE + PA compounding formula; for example, one of the narrative tenses pafa might be so incorporated, as in Lepafa ditca = 'The-one-who-was-going-to-be a teacher'; see Lexeme LE. Some of these tensed descriptors are very long, but even the longest so far built are easy to decipher. Take Lepacenoinacefa bragai. The pacenoinacefa-part of this descriptor unravels easily as 'before-and-not-now-and-after'; whence the whole expression might be elegantly translated as 'The once-and-future king.')
Even this long list of PA words is not complete. Loglanists are invited to explore their many possibilities.
Location Operators: These are vi va vu, 'at' or 'in', 'nearby' and 'far away', and their many derivatives, e.g., via vii viu, and so on. This list is incomplete.
Modal Operators: Unlike the tense and location operators, which are inherently systematic, the modal operators are a disordered set. We can do no better than list the ones in current use; no doubt others will be added.
as much as/to the same degree as…
for/on behalf of…
in manner/mode…/by method…
re/concerning/as for/with regard to…
like/as/in the way that…
for/in order to please…
as well as/in addition to…
with…'s help/through agent…
instead of/in place of…
with…, a tool or means
Notice how the modals differ from the case tags of the DIO Lexeme. In a certain sense, modal phrases may be used to extend the place-structure of nearly any predicate; so they may be thought of as "itinerant cases". But because such phrases can turn up nearly anywhere, they may never be distinctive features of any predicate. Therefore, the modal preposition may never be omitted from its argument, as a case tag may be, relying on the auditor to infer the modality of that argument from its context. So the modality of an argument must always be explicitly announced by some attendant modal preposition.
Loglanists are just beginning to explore the uses of modals as inflectors and adjectives. We expect adventurous speakers to make many interesting discoveries in these uncharted waters. With a little thought, almost any of these strange new usages may be sensibly interpreted. For example, what does Da durzo de hea mean? In particular, what is the adverbial sense of hea? Well; if you think about it, hea used non-inflectionally and without an argument must at the very least mean that X did Y with the help of someone else, that is, "helpedly". So Da durzo de hea must mean 'X does Y with help'. What about hea as an inflector? Reasoning analogically from, let us say, the sense of pa itself in these three positions—as a preposition pa means 'before (this designated time)', as an adverb 'before some undesignated time (presumably inferable from context)', and as an inflector 'before this particular time, namely the point of speech'—Da hea durzo de must mean that X's doing Y was helped by someone or something current, perhaps a person present at the time of speech. Could we translate it as 'X does Y with your help', you the listener? Extending the currency principle to another case, what does Da sea durzo de mean? As a preposition sea means 'instead of'. So I would assume that the specimen means that X did Y instead of someone or something else, someone or something which was in some sense present at the time of speech. Could it mean that X does Y instead of your doing Y? I.e., in your place? But note that the speaker, too, is present at the point of speech…but perhaps less interestingly so, since da is always present. As I say, loglanists are just beginning to explore this vast new domain of meanings that has been opened up by the machine grammar work.
Causal Operators: These are the PA words which, when prefixed with /i/, form the causal connectives which are members of the I-Lexeme; q.v. Here the same sixteen causal relationships are invoked by causal prepositions; and these same sixteen words may, if interpreted by ellipsis, be used adverbially and even as inflectors. The most common use of these elemental causal operators is as prepositions, and that is the sense of the English translations given here. Some of these meanings do not exist in the natural languages, so their Loglan meanings are occasionally hard to think out.
When the prepositional sense is plain, the adverbial and inflecting senses are easily inferred.
Ga:Ga is the fifth kind of PA word, and there is just one of its kind. Ga is a boundary-marker; it is used as the left boundary marker of any predicate whose left boundary would otherwise be unclear. Ga is therefore one of the "spoken punctuation marks" of Loglan as is suggested by its g-initial form. (The other boundary markers are ge gi go goa goi gu gue, q.v.) The left boundary of a predicate happens to coincide with the inflecting position of the PA words; so ga has the grammar of PA words, and is therefore lexemically a PA word even though it is semantically quite unlike the rest of them. Moreover, unlike other PA words ga is used only in the inflecting position, and then only when the immediately preceding word—exclusive of the "free modifiers” which, for these and other purposes, constitute a kind of "grammatical noise"; see Lexeme UI—is a predicate. Thus ga is used only in contexts like Le sadji ga fumna ('The wise one is a woman') from which its omission or removal would cause an unintended description to be heard: Le sadji fumna = 'The wise woman'.
It is considered bad usage to use ga where its removal would not cause this ambiguity. Thus *La Selis, ga fumna and *Tu ga fumna, although grammatically understandable, do not occur in well-formed Loglan speech. Here, then, is another grammatical superset, only some of whose members are actually used.