Language (Conlang)

In this post we’ll discuss designing fictional languages for worldbuilding projects, and how creating an entire language doesn’t have to be a daunting task with some simple steps.

Hey everyone, my name is Matthew, at least that’s what I’m called in English, and this post is part of a series where I will be going through a science-adjacent worldbuilding process step-by-step. Last time we discussed designing fantasy races, and how creatures evolve the intelligence needed for establishing civilisations.

For today’s discussion, we’ll be looking at language, specifically designing fictional languages, how to get started, and then make some languages of our own for our newly designed fantasy races.

Language, importantly, is a structured system of communication. While many creatures are capable of communication, language in structured form is unique to humans. Proto languages are expected to have been present among our ancestors all the way back to Homo Habilis over 2 million years ago, and over the course of its history, language has evolved from our proto languages into the plethora of unique languages that exist today. This evolutionary development is a defining characteristic of a naturalistic language, and the English language today is the product of many, MANY different iterations and developments that differentiate it from the other languages on earth.

Of course, as worldbuilders, we don’t have thousands of years to progressively shape and evolve a language, which is where conlang comes in. Conlang, short for constructed language, refers to a language that is deliberately constructed for use, rather than one that evolves naturally over time. For most people however, creating an entire language is an overwhelming, borderline insurmountable task, and often isn’t necessary for the worldbuilding process. Instead, we can create what is called a proto language, a usually very simple foundational language that will have all the information needed for language expansion in the future, should you wish to follow that path. In addition to being much simpler and less arduous to create, proto languages also tend to follow an evolution that mimics a natural language’s development over time, so we can expect more authentic and believable results.

So, where do we begin? The starting point for most languages is phonetics, which refers to the sounds present in a language. To establish this, we look to the International Phonetic Alphabet, or more commonly just the IPA, which is an inventory containing all sounds that can possibly be produced by a human, and as human worldbuilders designing for human audiences, I consider it kind of pointless if we can’t at least make the sounds of the languages we create, so this is the sound inventory we’ll be sticking to. This table for the IPA is readily available on Wikipedia, complete with attached sound links that you can click on that will play any sound in the inventory.

The IPA is separated into Place of Articulation, and Manner of Articulation. Place of Articulation refers to where along the vocal tract sounds are being made. For example, the ‘b’ sound is located at our lips, which makes it labial, whereas the ‘g’ sound is located at our palate, the back part of the roof of our mouth, which makes it dorsal. The Manner of Articulation then refers to what we ‘do’ with our vocal tract in order to make the sound. For example, when we make the ‘d’ sound, sound is made that is then suddenly stopped, which makes it a plosive. All other sounds that have the same mechanics fall into the same manner of articulation, which for plosives includes sounds like p, b, t, d, k, and g. When two sounds fall into the same box, like ‘f’ and ‘v’, they refer to the same sound with the left being voiceless, and the right being voiced. When making the ‘f’ sound, your vocal cords are silent, while when making the ‘v’ sound, your vocal cords vibrate, which adjusts the sound. It is exceptionally likely that if a language includes a voiceless sound, it will also include its voiced equivalent.

So, we just have to choose the sounds we like and call it a day, right? Well, not quite. Firstly, not all sounds in the IPA are present in a language. In fact, in many languages only a handful of sounds are present, and they tend to come in groups of places and manners of articulation. For example, this is the Australian English IPA chart, and yes, different dialects and accents have their own unique IPA charts that you can find easily on Wikipedia. As you can see, there are clear groups of labial, alveolar, post-alveolar, and velar sounds, as well as nasals, plosives, and fricatives. There are some outliers to these groups, such as the ‘ch’ and ‘j’ sounds represented by these symbols, and overwhelmingly most languages follow this trend of having a group of sounds with two or three outliers. For conlanging, less is more when it comes to sound selection, and for this series, we’ll be keeping things on the simple side in order to keep things easier to follow.

Let’s create some of the phonologies that we’ll be using for our worldbuilding project. These are the four fantasy species that we created last time, the Silarin, the Urakan, Naqwuil, and… humans! We’ll talk more about humans at the end, so for now, let’s focus on the other three. For the Silarin, I generally associate reptilian-like creatures with ‘s’ and ‘th’ sounds, which are both coronal sounds, as well as fricatives, so let’s establish groups of sounds in both categories, as well as a couple of outliers for more common sounds, like ‘y’, and ‘k’, and if we’re putting in ‘k’, we should also put it’s voiced equivalent of ‘g’. Also seeing as we have the ‘t’ and ‘sh’ sounds, we can include the ‘ch’ sound, which phonetically is a combination of the two, as well as the ‘j’ sound, which is a combination of the ‘d’ and ‘zh’ sounds, which are also both present. For human languages, it’s exceptionally likely that there would be at least one labial sound, though the lips of reptiles are structured differently than humans and are less pronounced, making labial sounds harder for them to produce, so let’s exclude them entirely. Let’s also keep their vowels at a very simple five-vowel system, ‘a’, ‘e’, ‘i’, ‘o’, ‘u’.

For the Urakan, a strong trend within human languages is that the word for ‘bear’ will include a strong ‘r’ sound following a vowel: ‘bear’, ‘ursun’, ‘orso’, ‘bjorn’, ‘ilevar’, etc. In many of these languages, the ‘r’ sound is trilled, so let’s focus the language around a trilled ‘r’. If you can’t roll your ‘r’s, then a regular ‘r’ sound will work fine. From this trilled ‘r’, we can create a group of coronal sounds, though to differentiate between this group and the group within the Silarin language, let’s exclude ‘th’ and ‘sh’ sounds, which are less common in languages anyway. These creatures also do have lips that can be used for language, so we can include a grouping for labial sounds, and now that we’ve done that, it makes sense to create another grouping for nasals, plosives and fricatives. From here we can throw in the ‘w’ sound to add to the velar group, add the ‘dz’ sound because the ‘d’ and ‘z’ sounds are present, and do the same for the ‘pf’ sound. Finally, we’ll include the ‘y’ sound as an outlier, and establish a five-vowel system, though instead of ‘e’ used like in the Silarin language, the Urakan will use a ‘schwa’ instead, which is the ‘ugh’ sound used in words like ‘about’, rather than ‘A/bout’. Their language is also going to include the glottal stop, which is the sound made at the back of our throat that very suddenly stops the sound between the two syllables of ‘nuh-uh’. Bears in real-life make this sound when they are scared or in danger, which is something we’ll return to when it comes to making some of their words.

The Naqwuil have perhaps the strangest language, at least to English speakers, and I’d like their language to be inspired by otherworldly eldritch languages, the kind you would expect from a Lovecraftian adventure. Their phonology however is reasonably standard, with groupings of alveolars, velars, nasals, plosives, and fricatives. They don’t trill their r’s, include the ‘w’ and ‘y’ sounds, as well as the glottal stop, which is used in their name. They too will have the usual ‘a’, ‘e’, ‘i’, ‘o’, ‘u’ vowel system, though every vowel also has an elongated equivalent, which in IPA is noted by the colon symbol.

Now that we’ve got our phonology down, let’s quickly talk about syllables. Syllables are separated into three parts: onset, nucleus, and coda. Technically, the nucleus is the only obligatory part of a syllable, and its overwhelmingly likely that it will be a vowel. The onset and coda are then any sounds within a syllable that are before or after the nucleus respectively, which generally are consonants. Syllable structure therefore is broken up into consonant and vowel sounds, represented by C’s and V’s. For example, a language with a CVC syllable structure has syllables that are always consonant, vowel, consonant. If the C’s are in brackets like this, it means that the consonant is optional and doesn’t need to be included.

Let’s assign the Silarin language a C-V-(C) structure, the Urakan language a (C)-(C)-V-(C) structure, and the Na’quil a structure of (CCC)C-V-(CC). With simpler syllable structures, Silarin and Urakan will be easier and likely sound more familiar, while Na’quilian has the potential for some truly strange sounding words that will be harder to pronounce. Notably, while some languages do have mandatory onset consonants, codas are always optional in every language.

Now we have to establish syllable stress. Consider the words ‘SUSpect’ and ‘susPECT’. The stress placed on the first syllable compared to the second completely changes the meaning of the word. The rules for where stress is placed on a word can become complicated and are likely to develop over time as a language evolves. What we need here is not an extensive list of rules, but rather a starting point. To keep things consistent and simple, let’s say that the default stress for all of our starting languages falls on the second-to-last syllable.

With our phonology, syllable structure and stress established, we can make our proto-language proper. Importantly, we don’t need to make an entire language, remembering that our sapient creatures are going to start out very simple. Instead, here are some basic root words that we can start with. From here, we need to establish a default word order, which is simply the order that the subject, the verb, and the object appear in a sentence. English for example follows that exact order, subject, verb, object, and for simplicity we call English an SVO language. Word order can follow any of the six possible combinations, though it’s overwhelmingly most common for the subject to be placed first, followed then by languages with the verb first, and then finally languages with objects first are exceptionally rare. Let’s give Silarin a word order of VSO, Urakan a word order of SVO like English, and Na’quil a free-word-order, which means that it will require prefixes and suffixes in order to establish what is being talked about.

Finally, we need to determine whether adjectives and ad-positions are derived from nouns or verbs, which will determine where they’re placed in a sentence. For simplicity, it’s far easier to have one derived from nouns and the other from verbs, rather than both deriving from one. Let’s say that both Silarin and Urakan have adjectives derived from nouns, and ad-positions derived from verbs. In both languages, the verb appears before the object, which is the noun, and so ad-positions are ordered first which makes them pre-positions, while adjectives are ordered last. Most languages also order possession, such as “the cat’s food” in the same order as adjectives and nouns, with the possessor, in this case the cat, being treated as an adjective, while the thing being possessed, in this case the food, is treated as a noun. This gives us these final sentence structures. Na’quil will use the opposite system, with adjectives derived from verbs and ad-positions derived from nouns, though with a free-word order we don’t worry about where the adjectives and ad-positions are placed, but rather need to use prefixes and suffixes to establish connections between words and their auxiliaries.

Importantly, while we’ve discussed a basic outline for sentence structure, these are at most guidelines, and for every point there is an example of a real-life language that bucks the trend, so if you want your language to be different and unique, there’s absolutely precedent that such a language can develop. Just be aware that the further from these guidelines a language deviates, the more complex and less user-friendly it will be.

With all of this together, we have everything we need for a proto language! In the Urakan language for example, “Ra tru prod run senprak tip kril” translates to, “I am going to the river to fish for food”, which is an easy translation for us due to Urakan having the same structure as English. In Silaran however, the same phrase is “Ko tu kathochus tes tha ta sitheschus kath sath”. Hopefully, some of the personality of both languages comes through, even in this very primitive state.

From here, these proto languages can be expanded upon progressively as we move forward. When these creatures make a new discovery or develop something new, they’ll probably develop words involving that thing, such as words regarding agriculture or metallurgy. As we know from real-life, it’s exceptionally likely that a huge number of diverse languages will evolve, though keeping track of hundreds of languages across multiple species is a little beyond the scope of this series, so for simplicity, we’ll be following each species’ major language on its journey through time.

For the majority of worldbuilders, this is enough information about languages to use for a project. Using the words we already have here, we can set up naming conventions for places, like “forest”, “town” or “river”. An Urakan settlement on a river-mouth might be called something like “Senenatgru”, which might over time simplify to have “gru” be a suffix attached to any settlement on a river-mouth. This is how in English we end up with suffixes like, ‘berg’, ‘dale’, and ‘ham’ attached to many place names. Importantly, if this does occur in your language, you’ll need to establish a new word for what the old word used to mean, otherwise our Urakan that are now using “gru” to mean ‘river-mouth’ are going to have no word for ‘finish’.

And there we have it, three proto-languages ready to go and be expanded upon as we continue to worldbuild. But Matthew, you ask, what about the language your humans will use? Well, we’re deliberately going to have humans have no fictional language at all, and instead use whatever language we speak in real life. Consider it like ‘galactic basic’ in Star Wars, or ‘common’ in Dungeons and Dragons. This is a very important worldbuilding strategy, because the vast majority of viewers, readers, and players don’t want to have to learn an entirely new language just to be able to connect with your world. By providing a ‘default’ option, it makes your world more approachable, while still being exotic due to your other conlangs being present. For most people, the names of creatures, their settlements, and short phrases are more than enough to keep them immersed and provide a wonderfully rich fantasy experience.

For those of you who do want a deeper dive into creating fictional languages though, what I’ve presented in this video is only a surface look at the basics of language creation, and the extended process can be a huge rabbit hole to get lost in, including factors like vowel and consonant harmonies, cases, rule exceptions, and even more. The channel I turn to for any advanced conlanging is Biblaridion, who creates fantastic and very informative content on creating language. If you’re interested, I’ve linked his channel in the description, so be sure to go check him out.

So, to recap, constructing a language for a worldbuilding project requires a phonological inventory, syllable structure, and sentence structure. With a small lexicon of root words and some core rules, we can make up basic sentences that will get us started with enough language to provide names for places and things for our fictional species and provide them with the backbone for cultural evolution.

Join me next time where we’ll be taking our species through the stone age, detailing everything from campfires around caves all the way through to the birth of agriculture. You can find all the information for this video as well as other resources for worldbuilding in general over at worldbuildingcorner.com. And if you’ve enjoyed this video don’t forget to like and subscribe to follow the worldbuilding journey. And until next time, stay awesome!