OTheB wrote:I have this for my language so far. It is based around monosyllabic words. I have created a list of 140 word roots to create vocabulary from, but I have no idea how to go from root to word, or how to combine roots without adding extra syllables. All the roots are between 1 and 3 of the consonants in a word (there is a maximum of 6).
How should I approach it? How should I choose what vowels to put in and where? I just can't get my head around it because it seems too arbitrary.
There are a number of ways to go from root to word. The simplest, of course, is the "root word" itself. In English, very many words are of this kind (though, mostly on account of all the once existing morphology having been lost or misplaced somewhere): see, hear, man, girl. These serve as "roots" onto which various morphological bits can be attached (sees, hearing, men, girly); and are of course "words" in their own right.
The next most simple method is semantic broadening. This is where "girl" means not just "female child" but also, by extension, "female adult", "female friend", "female lover", etc. It can also come to be applied to boys in a deprecative way: "don't be such a girl!" Here, no morphological changes have been wrought, only additions of sense & usage.
Next, consider various morphological strategies. Here we're talking about meaningful pieces that add to the basic meaning of the root. In English, these are things like -ness, -ful, -some, wan-, un-, be-, etc. So you can take the basic root "girl" and make it "girl-like" with the addition of a suffix; and then "un-girl-like" with the further addition of a prefix.
Note that in English, not all roots work with all affixes. "In-" can often be synonymous with "un-", the former being derived from Latin, while the latter is Nativeborn. We can add "in-" to corrigible, giving incorrigible; but we can't add "un-" to make *uncorrigible. We can however add "un-" to rightable, but not "in-".
Never fear homophony or homomorphology. Sometimes morphological affixes turn out identical in form, sound or shape. That's okay! Sometimes they are utterly orthogonal as regards meaning or use; sometimes they are confusingly similar. Sometimes the same word can have two absolutely opposite meanings. Take a look at these words: "see-s" and "see-s". One is a verb, with its characteristic third person present singular ending; the other is a noun, with its characteristic plural termination. Take a look at these words: "inflammable" and "inflammable". Both are adjectives, but they mean exactly the opposite of one another. One is composed of inflam- + -able and means liable to catch fire; the other is composed of in- + -flammable and means unable to be set on fire. This happy circumstance comes about because, in Latin, the external form "in-" derives from two entirely different PIE roots, "*n̥-" (not) and "*en-" (in or into).
Another method is the atom-smasher method. As the name itself suggests, you take two roots and click them together like legos. Atom-smasher, brindle-wort, man-child, tow-truck, cheese-burger. English doesn't really care what the shapes of the two roots are, we don't care what languages they come from, we just smack em together and tell them to play nice: taco-shack, water-pump, fire-hydrant. English does have some typographical rules that determine how such root compounds are spelled (.i., whether we write them as a single word, a hyphenated word or two separate words). In my opinion, this is really neither here nor there, but as far as English goes, it's something to be aware of.
Some languages will care about the boundaries, and may require extra buffering vowels or syllables. It's entirely up to you how to go about this alchemy. In Queranaran, for example, two roots can be placed together, other instances will require a liaison syllable, still others are compounded with pseudoconjunctions: marccencranna
is a kind of large, sweet black colored berry and the word is composed of marccens
, sweet and cranna
, berry (the -s is actually an inflexional marker, the root being marccen-
is a flatbread pitsa and the -ng-
in there is simply a liaison particle, a way to smoothen the transition from one root to the next; locuala an crushardo
(birth canal) - here an
is a pseudoconjunction that kind of means something like "of" or "as regards to" or "in connexion with".
As for "what vowels to put in there" -- that will largely be a matter of trial and error, paying attention to what your invented language is telling you about itself and time spent with it so that you become comfortable with what sounds right.
For example, I recently came across the noun forming prefix i-
in a more formal way just recently. Most Queranaran prefixes don't really affect the roots they are attached to. Some will experience or cause assimilation or a minor change in the following consonant such as voicing or devoicing. But i-
has rather more far reaching effects. It not only attracts the primary stress to itself, but it also causes a chain of umlaut-like effects throughout the rest of the word. For example, the word cóyahag
means "tail". Adding i-
, a kind of reductive~diminutive~almost deprecative prefix, yields ícuyâg
and means "penis" or "pigtail" or any kind of "clipped appendage". You can see that the stress shifts from the root to the prefix and also that the vowel O becomes U and further that the extended syllabic structure -aha- collapses into a simple long vowel. Depending on the vowels in question, various alchemies result. Point being, it took a pretty good understanding of how Queranaran should sound as well as some trial and error to get the results that fit the language best.
There is no cooky cutter answer here and no step-by-step foolproof method (even if the books say to the contrary). Just take a look at the above strategies, and any others you may come across and adapt them to the invented language you're working on.
And of course, feel free to show-and-tell or seek constructive criticism here!