How it works

Vulgarlang is an application for people who are completely new to conlanging and people who have been conlanging for a long time. It can quickly create a language for a story or role playing game without any input from the user, or it can be fully customized to produce a language that has every cool detail you’ve ever wanted!

For those new to conlanging, you can simply press Generate New Language — the generator will decide the sounds for the language, assign words to English definitions, and also come up with grammar rules.

You may want to click Custom Phonemes to tell the generator which consonants and vowels you want your language to have:

We often hear people say ‟I have a race of dwarves/orcs/elves in my story; I want their language to sound something like German/Latin/Mongolian, but I can’t figure out how to do it! Here is a quick hack!

  1. Click the Phonology settings
  2. Click Word structure option
  3. Choose one of the presets of a languages, e.g. German
  4. Click Generate New Language

And you should get a language that looks kind of like German!

This may be all you need from Vulgarlang. However, the most fun part about conlanging is making your own decisions, not just letting the computer decide for you! But to design a language, it helps if you to know a little bit about how languages other than English work. One of the key things about creating a fantasy language is to not just copy what English does. However, it’s hard to not copy English if English is all you know. Without having studied a range of languages from around the world, you probably don’t know what you don’t know. For example, did you know:

    • that some languages change up the order of words and end up being closer to how Yoda speaks, e.g. ‟strong is he” instead of ‟he is strong”?
    • that some languages use the same word for ‟blue” and ‟green”?
    • that some languages say a word twice to indicate the plural, e.g. ‟cat cat” to mean ‟cats”?

And that’s only scratching the surface.

This article will briefly explain some of the main things you might want to take control over, such as the sounds and spelling, as well as editing individual words. Meanwhile, our guides drill down into more detail on each topic. We also have a great YouTube channel on various topics.

Sounds and spelling

We’ve already seen two ways to control the consonants and vowels of your language. The easiest option is Custom Phonemes which tells the generator which sounds you want, but lets the generator decide where they can appear, and which sounds can appear next to each other.

The Word Structure option gives you more control by letting you decide which consonants are allowed to appear at the beginning, middle and end of a word, as well as which consonants are allowed to appear next to each other. It’s a lot more decision making, but sometimes you need this level of control to get the ‟vibe” of your language right.

Note that we’re using the word ‟phoneme” and not ‟letter” here. Phonemes are the sounds of a language. The letters we use to spell words usually represent one sound (i.e. phoneme), but often they represent many sounds! Consider the fact that E is pronounced three different ways in the word “extremely”. Or the fact that ough is pronounced differently in all of these words: rough, ought, though, slough trough, cough.

This is why linguists and conlangers often use the International Phonetic Alphabet (IPA) to represent the pronunciation of phonemes. The IPA gives us a unique symbol for every phoneme that represent the exact pronunciation of words in all languages worldwide. It can even account for regional accents, like the different between American and British pronunciation. The IPA is a standardized, objective way to pronounce words: something that is not possible to do with the Latin alphabet. In fact, the Latin alphabet only has five vowel letters, yet there are more than 30 distinct vowel sounds found in languages around the world (not to mention the dozens of variations on how they can be pronounced).

Despite the usefulness of the IPA, the documenting your conlang entirely in IPA has some drawbacks too. For one, it’s difficult to type on computers! For instance, there is not IPA keyboard layout for iPhone. Secondly, the average person doesn’t understand IPA. Therefore, most people looking at your conlang won’t actually have a clue how it’s meant to sound. For these reasons, conlangers usually create a spelling convention for their language using the Latin alphabet. This is called a ‟Romanization”. A Romanization is a good middle ground for documenting your language; it tells readers roughly what the language sounds like, perhaps without telling them exactly what it sounds like. Besides, it may not critical for your readers to know exactly how the language is pronounced. And the IPA is always there for anyone who’s interested.

In Vulgarlang, the spelling is represented as bold blue text, whereas the IPA is written between /forward slashes/.

When you let Vulgarlang decide the spelling convention, it tries to come up with a system is that roughly how English speakers would pronounce the word. In the above example iesh, /ʃ/ is the IPA symbols for sh as in ‟show”. So the generator has chosen sh too, because the most common way English spells that phoneme. But the vowel /aɪ̯/ is the sound in ‟eye”, ‟buy”, ‟lie” and ‟sky”. There is no consistent way to spelling this English. Vulgarlang happened to choose ie for this one. In reality, there may not be any great choices for this one. For instance, in English ie can also represent a different sound, as in ‟believe”.

In cases where a conlang word contains phonemes that aren’t found in English, the generator makes a spelling that is close to how it might sound to English ears. For example, the phoneme /ʈ/ is not found in English, but is found in some languages in India (e.g. Hindi). It sounds very similar to English /t/ as in “time”, but perhaps if can you imagine it done in an Indian accent! The phoneme / will be spelled with t so long as the language doesn’t also have the phoneme /t/. If it has both /t/ and /ʈ/, the latter will be spelled with some kind of diacritic mark, such as .

If you don’t like Vulgarlang’s default spelling choices, you are free to change the spelling rules in spelling options. Here is how to change /aɪ̯/ to how it’s spelled in “buy” and “guy“:

Editing the words

Before generating a language, you can add custom words to the Add and remove words. Use the format English word : part-of-speech.

Part-of-speech refers to whether the word is a noun, verb, adjective, etc., but we use the abbreviated form (n = noun, v = verb, adj = adjective, etc). If you don’t know the part-of-speech of a word, check any online dictionary.

Adding words this way makes the generator randomly generate a conlang word using the phonemes you’ve selected. To specify what the conlang word will actually be yourself, add = conlang word (in IPA):

This makes kalisi the IPA pronunciation of the word, not the spelling! Your spelling rules will determine how the word is spelled (for this particular case, the default spelling will also be kalisi since the word doesn’t have any non-English phonemes).

If you’ve already generated a language (either by loading a saved language, or by clicking Edit This Language), you can now change the vocabulary in the Add and remove words field. Remember to re-generate the language after making changes!