Conlanging for Writers, Part 1

Issue 62

The Quick and Dirty Guide to

Conlanging for Writers, Part 1


by Alexis Carter

Copyright © 2011, Alix Carter, All Rights Reserved


J.R.R. Tolkien was a university professor who taught linguistics courses focusing on the Germanic languages. He also used that expertise to create some of the best-known constructed languages, or conlangs, of the fantasy genre. Inventing the sounds, grammar, and writing systems for Quenya, Sindarin, and the other languages of Middle-Earth took him many years.


You may be hoping to finish the first draft of your next story, the one for which you want to create a new language, within the next year. If so, the thought of spending many years on something your readers might only see a few sentences of probably doesn't thrill you. However, there is another way, and it's right here.


The Quick and Dirty Guide to Conlanging for Writers is designed to introduce some of the basic building blocks of language, but I will not focus on English grammar. Instead, this series is intended as a guide to help you create your own unique language, rather than a copy of an existing one. I will include some elements of basic linguistics, along with examples both from languages around the world and from Ozadai, a conlang which I have invented to use as an example for this series. This first article will focus on a couple of the big decisions you'll need to make in order to create your new conlang.


Word Order


Word order, in a linguistic sense, deals with something very specific: the relationship between the subject, object, and verb in a sentence. For example, English sentences typically have a subject-verb-object (SVO) structure. This means that in most sentences the subject comes first, then the verb, and finally the object(s).


Hindi, on the other hand, is considered an SOV language; while we would say "He went to the store," a Hindi speaker would say "Woh baazaar gaya," which literally translates to "He store went." SVO and SOV are the two most commonly-found word orders, but OVS, OSV, VOS, and VSO languages also exist.


Some languages also adhere more strictly to their basic word order than others. English is fairly inflexible. In the sentence "The boy sees the king," there is nothing to differentiate the subject (the boy) from the object (the king) except their positions relative to each other and to the verb. If you switched the words around to read "The king sees the boy," the sentence would mean something completely different even though all of the words are the same.


However, other languages have built-in mechanisms that allow the speaker or writer to switch the words around without changing the basic action of the sentence. Latin is one of them; it is typically an SOV language, but speakers and writers have more freedom to move words within a sentence around. The statement "Puer regem videt" translates to "The boy sees the king," as does "Regem puer videt." Depending on what the speaker wants to emphasize, either of those constructions may be used. "The king sees the boy," on the other hand, would be translated as "Rex puerum videt" or "Puerum rex videt." This is possible because Latin has a system of declensions that allows the listener and reader to keep track of who is doing what.


I've decided that Ozadai, my new constructed language, will be a VSO language. The strictness of the word order will be middling, with the verb always appearing near the beginning of the sentence. The speaker will have some freedom to switch the order of the subject, object, and prepositional phrases, however.


Word Formation


A morpheme is the smallest unit of meaning within a language. Sometimes, as with the word "book," it is a word on its own. Other times, several morphemes can be combined to create a single word. "Book" can also be combined with other morphemes to create new words. We could add an "-s" onto the end of it to indicate that we are referring to multiple books, or we could add "text-" at the beginning to indicate that we are referring to a specific type of book usually found in a classroom. Just as linguists have come up with different ways to group languages based on their word order, they have also come up with four different language classifications based on the fundamental characteristics of their morphemes.


Analytic languages (also sometimes called isolating languages) average very few morphemes per word. Chinese is an analytic language; most words contain a single morpheme. For example, the idea of "we" is expressed by combining two separate words, one meaning "I" and the other indicating plurality. Because every word generally contains only a single morpheme, words in analytic languages also tend to be somewhat shorter; many of them are monosyllabic.


On the other end of the spectrum are polysynthetic languages, in which many morphemes of differing parts of speech can be combined into a single word. Some Native American languages are polysynthetic, as are some found in the South Pacific. Inuktitut, which is spoken in parts of Canada, is a good example of this. The statement "She is going to the store" can be expressed as the word, "Niuvirvimmuuqtuq." There are three different morphemes here: "niuvirvik," meaning store (the final -k is changed to an -m when followed by "-muuq-"); "-muuq-," meaning "to go toward a place;" and "-tuq," meaning "s/he." As shown in the previous example, multiple parts of speech can be put into the same word. Because it can be more difficult to find information on this group of languages, I have included links to websites that discuss the grammar of specific polysynthetic languages.


The third group, synthetic languages, falls in between the polysynthetic and analytic extremes. This group is broken into two subcategories: agglutinative and fusional. Agglutinative languages like Finnish and Turkish allow the speaker or writer to simultaneously attach many different morphemes to a single root. Each morpheme has a single meaning (e.g., to indicate that a noun should be plural, or that a verb is in the past tense). In Finnish, "We speak" would be "Me puhumme," with "puhu-" as the stem of the verb and "-mme" to indicate that the subject is "we." However, if I were to make it past tense, "We spoke," the sentence would look like this: "Me puhuimme." In this case, the "-i-" morpheme indicates that the action of speaking took place in the past, although there are other morphemes that can indicate the same thing with different verb types.


Fusional languages like French and Spanish are also synthetic, but some of the morphemes that can be attached to the root words have multiple meanings. For example, "We are speaking" translates to "Nous parlons" in French, with "parl-" representing the stem of the verb "to speak" and "-ons" as an added morpheme. However, "-ons" conveys multiple pieces of information--it indicates that the action of speaking is occurring in the present, that the action includes the speaker, and that the subject is plural.


There is no clear-cut division between fusional and agglutinative languages; it is instead more of a continuum. In fact, deciding to place your language in one of these four categories does not mean that you won't be able to use anything from any of the other groups. As speakers of different languages and dialects interact, they sometimes share features of their speech. You can choose to include examples of this in your own conlang.


Ozadai, the example for this series, will be primarily synthetic, falling closer to the agglutinative end of the spectrum. However, I'm also choosing to give my Ozadai speakers longtime neighbors and trading partners who speak an analytic language. This has resulted in several analytic elements being adopted in Ozadai.


On Alphabets and Pictographs


This series is not going to discuss creating an alphabet, pictographs, or any other type of writing system. Most readers won't have the time or inclination to learn how to read an entirely different script, and it could be a turnoff for some. In the end, even if the reader won't be able to understand all of the words without an in-text translation, your story should still be accessible. Including passages or even just sentences in another script could limit that accessibility, which is the last thing you want to have happen when you're submitting to an agent or publisher.


By now, you should have the beginnings of an idea of what your new language is going to be like. However, if you've chosen to create a conlang that is radically different from what you speak, one thing you can do is find books and websites about languages that are similar to what you will be inventing. Remember to keep an eye out for the second part of The Quick and Dirty Guide to Conlanging for Writers, which will cover the sounds of language and provide some guidelines for determining which sounds you want your conlang to have.


Polysynthetic language links: Nahuatl: