Hitlist for Conlangers
Up ] RPGs ] Miscellanea ] CTY ] [TASP]

 

First, I must say that I am indebted to Mark Rosenfelder for his Language Construction Kit; it has been an absolutely wonderful starting point.  That said, I have here the slightly different approach I have taken to language creation.  For a fuller explanation of these terms, go to Mr. Rosenfelder's page.  Also, much of this material is adapted from The Cambridge Encyclopedia of Language, which I think is an indispensable resource for conlangers.

Another good site is Pablo David Flores's How to create a language.

This page is a work in progress; I'll probably divide it into subpages eventually, dealing with syntax, morphotactics, and phonology separately.  I don't intend to write much about semantics...messing with that in a conlang can be painful.

Here's my list, and my apologies.  I don't know what I was thinking when I wrote this.


The first thing to do is make your syntax.  It's pretty quick and painless, and it's the bone structure of your language.  If you have any knowledge of generative syntax, some simple recursive rules can pretty quickly make a good syntax.  If not, here's a brief explanation**.

For a conlanger's purposes, there are the following classes of word: verb, noun, articles and specifiers, adjective, adverb, complementizer, inflection, conjunction and preposition.  Many of these may be familiar; complementizer and inflection are most likely the odd ones.  A complementizer is a word like "which" or "that" when used to introduce a subclause.  An inflection is not to be confused with the morphological term meaning an affix that shows certain grammatical properties of the word; in this case, an inflection is a word like English "to (Verb)" or any of the so-called modal auxiliaries: "can," "would," "may," etc. 

As for the others, a verb is an action, a noun is a thing, an article is "the" or "a," specifiers are things like "that" or "whose" when used to modify a noun, an adjective is a descriptor of a noun, an adverb is a descriptor of a verb or adjective, a preposition (which in some languages becomes a postposition) shows relative place, and a conjunction combines multiple words into one grammatical unit.

To make a syntax, all you have to do is make a few rules, like

  •  S > (NP) VP
  •  NP > DET (ADJP) N (PP)
  •  VP > V (ADV) (NP)
  •  ADJP > (ADV) ADJ (PP)
  •  PP > P NP

This series of rules gets you English syntax, more or less.  I've not included IPs (Inflection Phrases) and CPs (Complementizer Phrases) in the rules, as that gets more complex.

  •  S = sentence
  •  NP = noun phrase
  •  N = noun
  •  VP = verb phrase
  •  V = verb
  •  DET = determiner (i.e. specifier or article)
  •  ADJP = adjectival phrase
  •  ADJ = adjective
  •  PP = prepositional phrase
  •  P = preposition
  •  ADV = adverb

There are some more complex issues if you want to make a realistically plausible syntax.  Things such as headedness are straightforward, even if they require some thought.  Languages are either head initial or head final; this describes the placement of the crucial item, the head, in phrases in the language.  English is head initial; we put things like prepositions before the other words in the prepositional phrase.  Headedness doesn't make everything follow that sort of a pattern, though.  Adjectives precede their nouns in English.  This has been explained by saying that they are a different class of word, and operate more freely.  I won't go into detail here.  The point is, you can get away with some variation, as long as any given class of word is consistent; then again, languages like Italian have some adjectives that follow the noun and some the precede it.  There's room for tweaking. 

More complex because of its utility and commonness (in European languages, at least), is Movement.  When you say Who did John hit? the who is interpreted as the direct object, which in English should follow the verb.  However, Movement has taken the "important" word, the one that you as the speaker are really wondering about, and moved it to the front.  Being out of the regular order puts emphasis on the misplaced word.  However, in English, it has become normal for the interrogative noun to be Moved to the front.  Therefore, the "out-of-place" position is back after the verb, the place that is normal for non-questions.  Thus, you can say John hit who? if you want to emphasize the who.  Making rules for movement in a set like the one above is a real bugger, so it's best to describe them in their full idiomatic-ness.  Generally, Movement seems to be related to questions, but there's nothing to prevent you from giving it other uses.

If you find this interesting, find a linguistics course at a nearby college and see if you can take it or audit it.  I have no good online recommendations, but I may put a discussion of syntax up here some day, when I have the time.


So that's syntax.  Now we get to the fun part; morphotactics.  This is where morphology (word structure) and syntax meet.  This is where you determine what grammatical features are marked on words.  I have the various classes of word listed below; inflections I have not listed because they can mark only the same things as verbs.  Complementizers I have not listed because I can't think what could go on them.
  •  Verbs
      • Aspect: This describes completeness, habituality, duration, etc.  Terms like "perfect" and "imperfect" fall in this category.
      • Gender: In Indo-European languages, this tends to represent sex: masculine, feminine, neuter, etc.  But there are many languages that classify all sorts of other things here: man-made, organic, smaller than a breadbox, you name it.  It's sort of an abstract classifier.
      • Mood: Factuality, certainty, likelihood, etc.  Traditional classifications include "indicative," "subjunctive" and "imperative."
      • Number: How many there are.  English only has singular and plural, but some languages distinguish all the way up to trial, that is, three of a thing.  "Singular," "dual," "trial," "plural" and "collective" are about all you have to choose from here.
      • Person: Are we talking about me, you, it, or that other it far away?  The most that languages seem to distinguish is fourth person; English only goes to third.  Related, some languages have different forms of "we" to indicate that the addressee is included or not.
      • Tense: What time is the verb happening?  There are more options than past, present, future: you can have "near past," "far future," "hypothetical time period" et cetera.
      • Voice: Who did the action?  Did the grammatical subject perform the verb, or was the verb performed on the subject?  Was the subject the cause of the verb?  And so forth.
      • Verbs can also have case; this is getting out of my understanding, but trigger languages (q.v) have something like this.
  •  Nouns:

      • Case: Here you will find a good discussion of case as related to theta-roles (basically, the conceptual role played in the core of the sentence).  Other cases abound, but basically function as built-in prepositions.  If you don't have a case for every preposition (and I doubt any language does) then every preposition will have a case it goes with; sometimes its meaning will be changed depending what case its noun is in.  What, you ask, is a case?  Go take a Latin class.  Cases are markers on nouns that show what syntactic (grammatical) role they play in the sentence.
      • Gender: See above.
      • Number: See above.
  •  Pronouns:
  •  Adjectives:
  •  Adverbs:
      • Features: Not much that I can think of.  Any suggestions would be welcome.
  •  Prepositions:
      • Features: Sometimes, as in Italian, prepositions merge with adjacent articles.  There's not much more that they can do.

Here's the stuff to come:

  • Phonology:
      • Stress and Suprasegmentals:
  •  Orthography:
      • Native
  •  Style:
      • Politeness
      • Poetry

Contents copyright ©2002 Kit La Touche, in case anyone cares.