Brefic

Overview

As of now, Brefic is a sketch of a language, primarily an experiment in grammar design. I wanted to see if it was possible to design a language in which nouns, verbs, adjectives, adverbs, and even prepositions are merged into a single part of speech, without requiring any sort of part-of-speech markers (as in Esperanto) or otherwise seeming too artificial. Perhaps one day it will grow into a full language.

The grammar of Brefic is designed to minimize "grammaticalization." As many concepts as possible are included in the category of content words, and the small set of particles serves only to make the connections between them clearer. This maximizes the freedom to combine concepts with other concepts, and the freedom to include or omit concepts as desired.

At present, Brefic has a skeleton vocabulary stolen from European languages, mostly used in examples to illustrate the grammar. Its grammar, however, is entirely constructed, resembling a bizarre hybrid between Chinese and Japanese grammars if anything at all.

=Basic Grammar=

Parts of Speech
Content Words - All words which carry any sort of semantic content whatsoever. This includes nouns, verbs, adjectives, adverbs, and even post-positions. Brefic content words can move between these functions without any sort of modification, though the meaning of using, say, a "nounish" word as a "verb" or vice-versa is well-defined. It is possible to express entire sentences using nothing but content words (and not just tiny sentences, either.)

Particles - Words which do not carry any semantic meaning at all, but mark the relationships between the content words to help reduce ambiguity in parsing the sentence. The aforementioned all-content-word sentences can become very ambiguous, especially as they grow longer. Therefore Brefic contains a small word-class of particles to make them more precise when needed. Brefic's set of particles is small and bounded.

Interjections - These are words like "Hey" that are used to express an emotion (as opposed to naming an emotion, such as the noun "anger.") I won't define specific interjections or rules for using them - my focus is on content words and particles.

Modes of Grammar
Brefic contains a wide range of possible ways to express sentences, ranging between two main "poles."

Quick mode - This is the mode of quick yet ambiguous sentences. It is based off of the long strings of nouns that can appear in English and other languages. This mode requires no particles (except the occasional conjunction like "and" or "or") but as a consequence, it isn't always clear exactly what modifies what. The only general dis-ambiguating rule is that modifiers come before the words they modify. Thus the word orders in the quick mode are:


 * Subject-Object-Verb - ex: Jo das wid = "I that see" = "I see that"
 * Subject-Verb-Object is possible in a sense, though it would translate literally to "There is/are which is/are by "
 * Adjective-Noun - ex: ruz haus = "red house"
 * Genitive-Noun - ex: jo haus = "I house" = "my house"
 * Noun-Postposition - ex: haus-in = "in the house"
 * Number-Noun - ex: tri haus = "three houses"
 * Adverb-Verb - ex: rapid cur = "fast run" = "run fast" (This is actually quite flexible, since there's no distinction between adjectives and verbs - the adverb can be used as a modal verb, and thus appear after the main verb without any sort of particle.)
 * Verb-Modal Verb - ex: ed deb = "eat must" = "must eat"
 * Relative Clause-Noun - ex: apyl ed hom = "apple-eating person" = "person who eats apples"

Clear mode - This is the mode of unambiguous yet somewhat longer sentences. Brefic contains particles which are analogous to "parentheses" and "commas" to show exactly what modifies what. These particles also reverse the word order of the quick mode - the open-parenthesis particle, dy, also functions as Brefic's only preposition, with which adjectives can appear after nouns, objects (and even subjects) can appear after verbs, and so on.

To illustrate the difference between these two polar extremes, consider an example in which letters represent content words and parentheses & commas represent particles:


 * A B C - the quick mode, with no particles. We know that B modifies C, but A might modify either B or C.  A longer string of content words would have even more permutations as to what modifies what.

The clear mode can distinguish the two:
 * C(B, A) - C is modified by B and A
 * C(B(A)) - C is modified by B, which is modified by A

And of course, it's possible to mix the two:
 * A C(B) - A and B modify C (equivalent to C(B, A) above).
 * B C(A) - same as above
 * B(A) C - equivalent to C(B(A)) above
 * C(A B) - also equivalent to C(B(A))

Content Words
To understand Brefic content words, it helps to think of them all as nouns. This isn't the only way to describe them - they can also be described as all verbs - but I'll stick with the noun-based explanation for now. "Verbs" are nouns referring to actions - the English verb "sleep," for example, can be used without modification as a noun meaning "the act of sleeping," as in "I'll get some sleep." In Brefic, every verb is like the English verb "sleep" in this regard. I will refer to this as the "gerund rule."

Brefic "adjectives" are nouns referring to states, qualities, or properties. To illustrate with an English example, English color words such as "red" or "blue" can be used both as adjectives and as nouns referring to the concepts of the colors (though one can also say "redness," "blueness," etc.) Brefic has no need for suffixes like English -ness - all adjectives have it "built in." Alternatively, think of all adjectives as verbs meaning "to be " (and they can be used as such) and turn those verbs into nouns via the gerund rule.

Brefic has no prepositions, but postpositions (technically it has a preposition dy, but that's considered a particle.) It's hard to think of pre/postpositions as nouns directly - therefore, I'll describe it as a series of part-of-speech conversions that eventually leads to nouns. First, imagine a preposition as a verb meaning "to be " ex: replace "I am in the house" with "I in the house." Then imagine this verb as a gerund, e.g. "being-in." Now we have arrived in the land of nouns.

That's all well and good, but how do we go the other way? How do we use these nouns as verbs, adjectives, adverbs, or postpositions?

First, it's important to understand the entire Brefic sentence as a noun phrase. Jo un arbor wid, "I see a tree," would mean roughly "the seeing of a tree by me" or "my seeing a tree" or "my sight of a tree," etc. Saying this noun phrase as a sentence means that you are asserting the existence/happening of the noun phrase you're referring to. This means that all nouns, even the most "nounish" of nouns, are impersonal verbs meaning roughly "There is/are " (The trouble with this English translation is that "is" and "are" implies tense, whereas what I'm trying to express here is a tenseless concept.  The "tense" of the sentence would be expressed as a modifier within the "noun phrase" itself, such as "X which happens in the future" or something like that.)  For verbs, this "reverses" the gerund rule - to say "There is a sleeping done by me" is almost exactly equivalent to "I am sleeping." This is based on a somewhat philosophical idea that "existence" and "happening" are the same thing - "happening" is "existence" for events, and "existence" is "happening" for objects.

What about modifiers? In general, the "quick mode" of Brefic grammar operates on the principle that modifiers come before the words they modify. "Modifier" in this sense refers not just to adjectives and adverbs, but even to subjects, objects, genitives, relative clauses, and postpositional phrases. In English, we can use nouns as modifiers by sticking them in front of other nouns, as in "truck tires" or "sound amplifier." There is some vague-yet-obvious-from-context meaning connecting the two nouns, but it's always clear that "sound amplifier" is a kind of "amplifier," not a kind of "sound," and "truck tires" are a kind of "tire," not a kind of "truck." (Not all English compounds work this way, but all Brefic compounds do.) Brefic modifiers operate on these exact same principles. For example, to express the concept of "angry person," one would say engo hom, which translates more literally to "anger-person." Likewise for verbs we have ed hom, "eating-person", i.e. a person who eats. In general, A B can be translated as "B which is related to A in the most direct way that makes sense, given the context" (or "B of A" if you want to be short.)

Both postpositions and relative clauses are considered specific versions of "modified modifiers." For example, in hom would mean "in-person" - a person who is inside something. But haus in hom is more specific - it means "house-in person," or a person who is inside a house. The root haus modifies in, which modifies hom.

Compounds
Compounds in Brefic mean exactly what they would mean if they had spaces separating their roots - compounds and strings of roots are interchangeable. Thus, the same left-branching rules apply to compounds - the root at the end is the "kind" of thing the compound refers to. It's possible to turn postpositions into case suffixes, and auxiliary verbs into tense suffixes, merely by compounding them. It's theoretically possible to express an entire sentence as a single, giant compound, but that wouldn't be too easy on the eyes.

Particles
Particles in Brefic are designed to help resolve ambiguities and free up the word order. Particles follow special rules different from the general rules that Content Words follow, and for this reason I intend to keep the number of particles in Brefic down to a small size.

dy - roughly "of" (though it can also mean "which is," "who," or "that") this is the only "preposition" in Brefic. It is used to indicate that the word before it is modified by a word or phrase appearing after it. For example, hom dy engo is equivalent to engo hom, angry person, and Jo wid dy un arbor is equivalent to Jo un arbor wid. This particle fills the dual role of freeing up the word order and splitting sentences up into smaller phrases to reduce ambiguity.

yc - a kind of "close-parenthesis" for a phrase beginning with dy. For example, Un hom dy engo un arbor wid would mean "There is a person who angrily sees a tree." However, Un hom dy engo yc un arbor wid would mean "a person, who is angry, sees a tree." The yc blocks the dy from "capturing" the words after it. This particle is not necessary if the dy-phrase terminates at the end of the sentence (as the first example shows.)

fy - a "comma particle" that separates multiple phrases within the "parentheses" of dy and yc. For example, "A dy B fy C fy D dy E fy F yc fy G" can be visualized as A(B, C, D(E, F), G). I believe that parentheses and commas like these show with complete unambiguity what modifies what (i.e. A is modified by B, C, D, and G, and D is modified by E and F), so the particles that translate to these symbols should create complete unambiguity as well. The particle "fy" can also be expressed as a pause in speech.

by - "by X," where X is a single root, is a shorthand way of saying "dy X dy." This can be used to turn postpositions into prepositions, among other things. A phrase that begins with by- only requires one -yc to "close" it.

Compound particles:
dy-dy - a version of dy which implies that all the individual roots within its phrase are separated by an unspoken dy. Example: A dy-dy B C D = A(B(C(D))). Only a single -yc is needed to close all the parentheses that come from this construction.

fy-dy - similar to dy-dy, but implies a fy between every root instead of a dy. Example: A fy-dy B C D = A(B, C, D)

Logical particles:
Logical particles are considered a separate class from the above grammatical particles. They include conjunctions such as "and" and "or," and the negation particle "not."

ei - means "and." For numbers, this means "plus."

mei - means "except" or "but." For numbers, this means "minus."

or - means "or"

ar - exclusive or

na - means "not." Na appears immediately after the word it negates. Ex: arbor-na = not a tree / not trees

nei - "nand," i.e. "not both"

nor - "neither...nor"

nar - "both or neither"

lei - "all" or "every" - appears before nouns as a quantifier. ex: lei hom = "all people"

lor - "some" or "at least one" - appears before nouns as a quantifier

lar - "some, but not all" - appears before nouns as a quantifier

Tense Markers
Tense markers in Brefic are nouns meaning things like "future event," "past event," etc. They can appear either as modal verbs modified by the main verb or as adverbs modifying the main verb. Some important tense words are:
 * wil = "future event," the future-tense marker.
 * did = "past event," the past-tense marker.
 * nau = "current event," the present-tense marker.

Examples of use:
 * Jo ed wil = "There is a future event of eating done by me" = "I will eat."
 * Jo wil ed = "There is an act of eating which is a future event done by me" = "I will eat."
 * Wil jo ed = "There is an act of eating done by me which is a future event" = "I will eat."

However, the verbs called "modal verbs" are different - they can't convey their meaning as an adverb, so they must be modified by the main verb. Example: Jo ed cyn = "I eat-can" = "I can eat" (The phrase ed cyn can be translated as "ability to eat.")

Comparison
Brefic has three comparison words - plu, min and tam ("more", "less", and "as much as", respectively.) These can modify words just like any other modifier:
 * plu clor = "more green"
 * min acwa = "less water"

Brefic's equivalent of "than" is to simply modify the comparison word. For example:
 * foil-plu clor = "greener than a leaf"
 * litry-min acwa = "less water than a litre"
 * mar-tam acwa = "as much water as the sea"

The superlatives are lei-plu and lei-min, meaning "most" and "least," but translate more literally to "all-more" and "all-less" (i.e. "more than all the others" etc.)

Case Marking
Brefic has optional postpositions to mark the subject and object. They are:
 * wys - the subject postposition, means "done-by," "to do," or the noun "action".
 * on - the direct-object postposition, means "done-to," "to undergo," or the noun "undergoing."

Neither of these markers are obligatory. When neither subject nor object is marked, it is assumed that the subject comes first.

Examples of use:
 * Jo-wys un apyl-on ed = "I eat an apple"
 * Un apyl-on jo-wys ed = "An apple, I eat"
 * Jo-wys ed dy un apyl-on = "I eat an apple" (Notice the particle dy - it keeps the verb ed the head of the sentence.)
 * Jo-wys ed un apyl-on = "An apple undergoes an act of eating done by me" = "I eat an apple." (This construction is allowed, but hinges on the assumption that the main verb ["ed" in this case] is another modifier to the case-marker ["on" in this case], rather than a modifier to something else.)

Between marking both of these cases and marking neither, Brefic allows the speaker to only mark one of them and leave the other unmarked. The unmarked noun should be the one closest to the main verb, but doesn't need to be. Some examples:
 * Jo-wys un apyl ed = "I eat an apple"
 * Un apyl-on jo ed = "An apple, I eat"

Likewise, if both subject and object are marked but there's a third unmarked noun phrase, that noun phrase is most likely an adverb or indirect object of some sort. Example:
 * Dis dei jo-wys un apyl-on ed = "Today I eat an apple"

Relative Clauses
A relative clause in Brefic is simply a "verb" that modifies a "noun." For example:
 * ed hom = "a person who eats."
 * hom dy ed (yc) = same as above.

This does not mean that every time a verb modifies a noun, the noun is the verb's subject or agent - rather, like all modifiers, the connection between them is "the most direct relationship that makes sense, given the context." For example:
 * ed haus = "restaurant" (eating-house, i.e. a building for eating)
 * ed apyl = "eaten apple"

However, the relationship can be clarified by inserting a content word between the verb and noun:


 * ed por haus = "house of the purpose of eating"
 * ed on apyl = "apple of undergoing eating" (i.e. eaten apple.)
 * ed wys hom = "person of doing eating" (i.e. person who eats.)

Note that the words on and wys are not acting as case-markers for ed in the above examples (i.e. they're not rendering "eating" in the nominative or accusative case.) Rather, like all content words, they are nouns, and they are behaving as part of a noun compound. If one interprets on and wys as mere stand-ins for case suffixes, then ed on apyl would mean "the apple which is done to the act of eating." But in Brefic, if there's a way to interpret a construction that makes sense (in this case "the apple which undergoes the act of eating"), then the construction is allowed, and nonsense interpretations are eliminated in favor of the interpretation(s) that make sense.

The verb can acquire subjects, objects, adverbs, postpostional phrases, etc. the same way as any other verb does - by adding them as modifiers. For example:
 * apyl ed hom = "person who eats apples"
 * apyl on ed hom = "person who eats apples" (clearer)
 * apyl on haus in ed hom = "person who eats apples in houses"

To put the relative clause after the noun, simply use the particle dy.

Example: hom dy apyl-on ed (yc)

With relative clauses that appear after the nouns they modify, it's possible to use third-person pronouns to refer back to the noun that is modified. Example:


 * haus dy ta-in pitsa-on ed (yc) = "house of eating pizza in it" = "house in which pizza is eaten"

However, this only works if the relative clause appears after the noun it modifies, since pronouns can only refer to something mentioned earlier. If one were to say ta-in pitsa-on ed haus, the pronoun ta would refer to something mentioned earlier, not haus.

=Sounds=

Where more than one sound is indicated, the pronunciation of the letter is the free choice of the speaker.

Vowels:

 * A = [a] as in father
 * E = [e, ɛ] as in great or set
 * I = [i, ɪ] as in machine or sit
 * O = [o, ɔ] as in so or sore
 * U = [u, ʊ] as in rude or push
 * Y = [ə] like a in about

Diphthongs: ai, au, ei, eu, oi, ou, ui, iu

Consonants:

 * B = [b]
 * C = [k]
 * D = [d]
 * F = [f, v]
 * G = [g]
 * H = [h, x] ([x] is not in English)
 * J = [j] like y in yes
 * L = [l]
 * M = [m]
 * N = [n]
 * P = [p]
 * R = [r]
 * S = [s, z]
 * T = [t]
 * W = [w]
 * Z = [ʃ, ʒ] like sh in show or z in azure

Only two types of consonant clusters are allowed at the ends of syllables:
 * -ng which is really the single consonant [ŋ].
 * -r + consonant, as in -rt, -rs, -rm, etc.

=Dictionary=

Main article: Brefic Dictionary

Pronouns

 * Jo = I, me, my
 * Tu = you (singular)
 * Ta = he, she, it, they(singular)
 * Nu = one
 * Se = oneself (which can mean "myself" "yourself" "themselves" etc.)
 * Jos = we (excluding you)
 * Tus = you (plural)
 * Tas = they, them
 * Nus = we (including you)
 * Ses = each other
 * Dis = this
 * Das = that
 * Dise = these
 * Dase = those
 * Cwe = what, which

=Numbers=

Digit words:

 * 0 = nul,
 * 1 = un,
 * 2 = du,
 * 3 = tri,
 * 4 = cwar,
 * 5 = cwin,
 * 6 = siz,
 * 7 = set,
 * 8 = oit,
 * 9 = nof

des = "more than one," a plural quantifier which can be used as a digit in its own right. Ex:

des haus = houses

desmil = thousands

Powers of Ten:

 * dec = 10
 * cen = 100
 * mil = 1000
 * wan = 10000

Suffixes:

 * -jon = ^2  ex: miljon = 1000000
 * -jar = ^3
 * -li = ^-1
 * -yp+number = ^(number), where number is a single digit or power of ten.

Ex: milypsiz = 1000^6, wanypwan = 10,000^10,000

When powers of ten, or words constructed from them using the above suffixes, are placed next to each other, the result is the multiplication of the two. Ex: decwan = 100,000, cenwanypwan = 100*(10,000^10,000)

Multi-digit numbers are constructed as follows:

First digit (unless it's 1) + Power of Ten + The rest of the digits listed sequentially (unless they're all zeroes.) If the first digit is 1, then saying "un" is not necessary - the power of ten will suffice.

Ex: 947 = nofcen-cwar-set

12 = decdu

20 = dudec

68,295 = sizwan-oit-du-nof-cwin

767,598 = setdecwan-siz-set-cwin-nof-oit

If there is a long string of zeros in a number, the number can be constructed using multiple powers of ten:

649,000 = sizcen-cwar-nof-mil

649,000,070 = sizcen-cwar-nof-miljon ei setdec

=Example text= ...