What are Conlangs Good For?

Yesterday (December 31, 2023), the account @lingdiscovery posted a thread on twitter titled “Here are 3 reasons why I hate conlangs”. I, like many other people, found the thread unfortunately very ill-informed, painting a picture of conlanging that does not live up to the current reality. I especially don’t like how the author of the thread is playing off conlangs and endangered natural languages (natlangs) against each other.

Like every form of art, you can like or dislike conlangs but not based on the arguments in this thread, please. So here is my reply to @lingdiscovery.

Before we start, a conlang (as @lingdiscovery defines), is a ““constructed language”, an invented language like Dothraki or Klingon or Na’vi or Esperanto”. I’m not a part of the conlanging scene. I’m a linguist who studies certain conlangs in her free time as other people study music or paintings.

This post started as a thread on twitter, which I wrote in a hurry and it is therefore full of typos 😀

1) Conlangs are less interesting than natlangs

@lingdiscovery’s first argument against conlangs is that they don’t possess the richness of natlangs:

“1) Conlangs are boring. They usually only possess features from major languages because their creators don’t realize the vast diversity of different ways that languages work in the real world.” @lingdiscovery

This is straightforwardly not true: most conlangers are extremely interested in natlangs and either use features they particularly like for their conlangs, or try to come up with new ones. The website of David J. Peterson (the linguist who created conlangs such as Dothraki and Valyrian) offers fascinating insights into how Peterson uses his knowledge of natlangs to create conlangs. Klingon has features of South American languages, the area of expertise of the creator Mark Okrent.

Trying to create new features can inform us about what languages could possibly do but for some reason don’t. This, in my opinion, is one of the most important aspects of conlanging: speculative linguistics is a valid form of theoretical linguistics. And, by the way, linguists already do that. Typologists, for example, often create typologies for linguistic features that cover all possible forms of this feature, not only the attested ones. A while ago I took a class on canonical typology where we were confronted with such an unattested expression of a feature. I immediately started to do a little bit of conlanging in order to see what a language with this feature may look like and how this feature may interact with other features.

A few weeks ago, a paper by me was published that deals with a case of speculative linguistics, pursued by nobody less than J.R.R. Tolkien. Tolkien was unsatisfied with the way English (and many other languages) uses the same words for future and past reference (e.g. after, before) and he tried to come up with an alternative system for his Elvish languages. In my paper, I offer an explanation for why the system invented by Tolkien doesn’t occur in natlangs. This is a form of historical linguistics and grammaticalization theory that wouldn’t be possible in the same way without conlanging.

@linguistcdiscovery continues his argument as follows:

“I’ve never seen a conlang designed to be spoken (or signed!) by humans which comes close to approaching the beauty and complexity of natural languages.” @lingdiscovery

Whatever is meant by “the beauty and complexity of natural languages”, I would like to point out that the conlang Esperanto has about a 1000 native speakers today. There are multi-generational Esperanto speaking families, believe me or not. But sure, tell these people that the language they raise their kids in and which is the normal means of communication in part of their culture, the Esperanto community, doesn’t come “close to approaching the beauty and complexity of natural languages”. Esperanto is proof that conlangs can get “nativized” and acquire the full range of applications that natlangs have.

Funnily enough, a similar argument was brought forward by Tolkien, one of the fathers of modern conlanging. In his lecture A Secret Vice, Tolkien argued that conlangs without a accompanying history and literature would be dead, and that attempts like Esperanto for this reason already are dead. Well, Tolkien couldn’t foretell the future. Esperanto nowadays is so much more alive than any of Tolkien’s conlangs. Conlangers nevertheless took Tolkien as a role model and just like him created fictional worlds, cultures and histories to go with their conlangs.

2) Conlangs are completely regular and therefore obscure the complexity of natlangs

@lingdiscovery’s next argument seems to be based on a lack of knowledge about contemporary conlanging:

“2) Conlangs distort our understanding of how language works. Conlangs are neat and tidy and regular. Conlangers think of them like logic puzzles.” @lingdiscovery

No, they don’t! Especially in the 17th century, but also later, philosophers tried to come up with perfectly regular and logical languages that would eradicate miscommunication once and for all. These ‘philosophical languages’ were not successful, proven by the fact that nobody who isn’t particularly interested in conlangs has ever heard about them. An example for a contemporary philosophical language is Toki Pona, a language designed to be as simple as possible by having a super small vocabulary and very simple rules of grammar. According to wikipedia, Toki Pona has about 500 to 5000 users (no native speakers). It’s an experiment that does nobody harm. In comparison, Esperanto has about 30,000 to 2 million users. Esperanto was created in the 19th century and belongs to the ‘auxiliary languages.’ These languages were created in the 19th and early 20th century to facilitate communicate between speakers of different languages. They were constructed to be completely regular in order to make them easy to learn. They are usually rather eurocentristic and, besides Esperanto, didn’t gain much following.

But the heydays of philosophical and auxiliary languages are long past. Today, conlangers often introduce irregularities in order to make their languages “more natural”. Conlanging is now mostly a personal hobby of individuals who are creating their own languages. They are often not meant to be used as means of communication. One of the early examples of a conlanger who tried to make his languages “natural” is, I mentioned him before, J.R.R. Tolkien. Tolkien saw his conlangs (best known are Quenya and Sindarin) as art and his main goal was to make them as aesthetically pleasing for his taste as possible. And, listen carefully, they have irregularities! Tolkien feigned a complete history for his languages including the irregularities that occur during language change. So yeah, Quenya irregular verbs is a thing. And often conlangers boast about how complex they made their languages.

I’d rather say that the “logic” thing is nowadays mostly found among language pedants for large national languages. One of the lasting effects of the 19th century is the believe among non-linguists that languages have to be logical and clear in order to make them an appropriate means of communication. @garicgymro recently wrote a thread on twitter about the issue.

This is actually a nice example of a widespread idea of how grammar works that's just… not well-founded. 1/ pic.twitter.com/4Wc67Gajh3

— Gareth Roberts (@garicgymro) December 28, 2023

One of the natural features of English that was ousted because it wasn’t considered “logical” enough are double negatives. Sentences such as “Nobody don’t like me.” instead of “Nobody likes me.” were frequent in English until the 18th century, and still are frequent in many varieties. Then grammarians decided it had to go.

I will now skip the next tweet by @lingdiscovery because the tweet after it fits here better:

“Worse, conlangs propagate the notion that language can be controlled, molded by pedantic scholars or authoritarian politicians to reflect their vision of how society should be. The reality is nobody really controls language.” @lingdiscovery

Again, in my experience, that’s what language pedants do, not conlangers. Pedants want to control languages, they want to allow or disallow whether people can say ain’t or whether something “is a word or not”. Language standardization efforts leave little room for variety and they are closely connected to nation building and nationalism. None of this is found in current conlanging.

Coming back to the tweet I skipped:

“Natural languages are messy, inconsistent, illogical, complex-adaptive spontaneous orders riddled with exceptions and layers of culturally-specific history that are shaped by constraints on human cognition.” @lingdiscovery

I would argue that conlangs as well are shaped by “layers of culturally-specific history that are shaped by constraints on human cognition”. But mainly by the culturally-specific history and cognition of the creator(s). It’s not a coincidence of course, that the ‘auxiliary languages’ mentioned above are mostly based on European languages – because their creators were European men. But these languages are a part of the history of European nation building, trying to counter nationalism through overcoming language borders.

It is also no coincidence that most conlangs are oral languages (produced with the mouth) instead of e.g. sign languages or olfactory languages, because most conlangers are hearing and speakers of oral languages. Nevertheless, languages that use other means of communication exist in Science Fiction. The most popular example is probably the Alien language in Arrival. The question what human cognition can create beyond what is known is extremely fascinating and an interesting field of study in itself. Cognitive linguists use experiments where participants create conlangs in order to explore how language is processed in the brain. Also think about the Wug Test that uses a nonce word (a made-up word) to study language acquisition. Conlanging is also used in the study of language evolution and many other areas of linguistics.

“Language emerges from millions of individual interactions each day. As the economist Friedrich Hayek once noted, language is “the result of human actions, not of human design”.” @lingdiscovery

True, that’s why conlangs are different from natlangs in the same way the painting of a flower will be different from the original. Unless they get nativized, like Esperanto. Native Esperanto differs in several aspects from the Esperanto devised by Zamenhof. Speakers introduced grammar rules, words and especially phonological features from their native languages, making Esperanto more variable.

3) Conlangs are an impediment for minority natlangs

@lingdiscovery’s last argument against conlangs is his most valid one, I would say, even though I disagree with him in several points:

“3) Indigenous languages need the attention more. There are 7,168 natural languages in use in the world today. Of those, at least 3,045 (42%) are endangered, meaning that they losing speakers. Fewer and fewer children learn endangered languages each year.” @lingdiscovery

It is absolutely true that natlangs, and especially endangered ones, should be our main focus as linguists and language learners. But that, in my experience, is the case anyways and I can’t see conlangs taking over linguistics any time soon. Most professional linguists have little interest in conlangs, neither do language learners. The major threat for minority languages isn’t conlangs: it’s large, standardized national languages.

“Many communities are hard at work attempting to revitalize their languages, creating dictionaries, grammar textbooks, and lesson plans. Yet these efforts receive little attention from language lovers.” @lingdiscovery

I understand the argument that conlangs may distract people who are interested in language from natlangs, but I don’t think that this makes a difference for these endangered natlangs.

“I know people who spent years creating grammars for conlangs, or entire communities that have come together to produce a dictionary (like for the Na’vi language from Avatar). Imagine this same passion put towards documenting and revitalizing indigenous languages!” @lingdiscovery

As others have pointed out, many conlangers would love to engage with endangered languages more but, surprise, not everybody can be a linguist. Conlanging is a cheap and easily accessible hobby – doing language documentation and (re)vitalization is not. Besides the fact that you can engage in both conlanging and helping minority languages, of course.

I would also make a distinction between indivuals learning a conlang and companies putting resources into conlangs instead of natlangs. For example, I’m not a fan of the Duolingo Klingon and High Valyrian courses.

I find @lingdiscovery’s line of argumentation pretty weird, especially in the following tweets:

“What to do instead of conlanging?

My advice is this: Learn a language that, by doing so, helps increase access for minority communities.” @lingdiscovery

“For example, you could learn American Sign Language (or whichever sign language is common where you live), a common immigrant language, or an indigenous language in your area, rather than make those people speak to you in English (or whatever major language you speak).” @lingdiscovery

One indigenous minority language in Germany is the Slavic language Sorbian (ca. 20,000 speakers). A couple of years ago, a friend of mine and I traveled to the largest city in the traditionally Sorbian speaking part of Germany, Bautzen/Budyšin. We went to the Sorbian museum and the Sorbian bookstore and each spent about 100€ on books in and about Sorbian. This probably helped the Sorbian community more than us learning the language and talking to speakers ever could.

“Attend language classes at the nearest tribal college. Doing so gives validation to speakers of a language who were told for generations that their language was worthless and primitive.” @lingdiscovery

Taking minority language classes can generate income for members of minority communities, you learning the language at home doesn’t. Donating money (and maybe time) to language vitalization and documentation efforts can help as well. Sure, speakers of minority languages may appreciate outsiders learning their language but I don’t believe they and their language will necassarily profit from it. Learning sign languages is an exception here because they drastically improve accessibility for deaf and HOH people.

And an English speaker speaking a minority language gives the speakers “validation”? Am I the only one who finds this statement pretty weird and problematic? Speakers of minority languages don’t need our “validation” – they need our support, and most importantly state support, to speak and pass on their languages undisturbed.

If you want to help a minority language, approach the speaker community, ask what they need or want and then LISTEN! When I’m asked why people engage in conlanging, I often reply that it is the easiest way to own a language without appropriating one. As someone wrote on twitter, the argument made in the thread sounds a lot like “throw away your conlang and appropriate a natlang instead”. I don’t believe that this is what @lingdiscovery is trying to say, though. His account is full of appreciation for minority languages.

“If you take up learning an endangered or minority language, I promise that you’ll come to appreciate language in all its intricate richness far more than you ever did studying conlangs, and put a little good into the world besides.” @lingdiscovery

Yes, learn a language that you wan’t to communicate in – that’s what languages are for. And chances are low that you will ever communicate in a conlang (besides Esperanto maybe). BUT conlanging for many people is their first insight into linguistics.

Conlanging introduces people to the richness of language and then many go on to explore natlangs and study linguistics or become lay linguists. It’s like singing in a children’s choir introduces kids to music. We should be happy about every person who wants to learn about languages and linguistics, especially since linguistics is virtually absent from schools. I myself came to linguistics via conlanging and now I’m doing language documentation. In fact, all people I know who dreamed about becoming a linguist before going to university discovered linguistics via conlanging. Many books for conlangers are actually introductions to linguistics, for example David J. Peterson’s The Art of Language Invention.

And even if they don’t study linguistics properly, many conlangers are enthusiatic about natlangs. Unfortunately, some consider them only as repositories of interesting linguistic features and don’t further care about them. But that shouldn’t lead us to demonizing the love for languages in all conlangers! Instead, we should educate the whole public about language appropriation vs. appreciation.

Conlanging ≠ conlanging

In his last tweet, @lingdiscovery recommends the book In the Land of Invented Languages by Arika Okrent. This is indeed the best book about conlangs currently available.

However, I’m wondering whether this book may also be the source of some of the misconceptions found in his thread. In her book, Okrent retraces the history of conlanging from the very beginning up to today. But she mostly tells about big conlanging projects. She says little about the many individual projects, the private languages that make up the bulk of conlanging today. Okrent didn’t visit independent conlangers, she didn’t go on reddit and followed one of the many conlanging communities there. Probably because when she wrote the book in 2009 (!), the conlanging community was very different from today. It was harder for people to make their projects public and meet like-minded people. This has changed since then. If you are interested in these smaller projects I recommend the documentary Conlanging: The Art of Crafting Tongues (2017).

If you would like to see a conlanger at work you can visit David J. Peterson’s YouTube channel where he posts about his latest projects for big TV shows etc., such as Game of Thrones or The Witcher.

Conclusion

Conlanging is an art, it has the same right to exist as every art form has (or doesn’t have). Should we use resources to study conlangs instead of natlangs? Definitely not! Should we encourage people interested in conlangs to look into natlangs too? Absolutely! Are conlangs branching off resources meant for the study of natlangs? I don’t think so.

Classes on conlangs given at universities are few. Classes that take conlangs into consideration from a theoretical perspective rather than laying the focus on creating new conlangs hardly exist at all.

I understand and appreciate the intention of the thread but I think conlanging is not the problem here. It is such a nichy hobby that it will hardly have any consequences for natlangs. Imagine what the resources spent on sports could do for language documentation instead. The seven million U.S. dollars payed on average for one 30 seconds long commercial during the 2023 Super Bowl could easily fund the documentation and (re)vitalization of more than 1000 languages. Playing off conlangs and endangered natlangs feels like whataboutism to me. Especially, since I pointed out above, conlanging and an interest in natlangs and linguistics often go hand in hand.

I also understand that people working in language documentation and (re)vitalization feel disenchanted when conlangs get more attention than the natlangs they care for. I myself sometimes feel this way. But our strategy should be to educate more people about linguistics, not going against the people who already are enthusiastic about languages.

So, in summary, what are conlangs good for?

1. They can make people appreciate natlangs. Conlanging requires knowledge about languages and linguistics and encourages people to study both.
2. They can explore the possibilities of language beyond what is documented. Knowing what could exist but doesn’t is important and conlanging is the easiest, and an already practiced, method to do so.
3. They can help us get insights into the cognition of language. Conlanging is a creative way of using language and can be employed in multiple fields of linguistics, as is already done today.

There is certainly more to be said here. The most commonly uttered sarcastic reproach below the thread goes along the lines of “stop writing fiction since there still is so much history to do”. Creating languages is different from doing language documentation or writing state policies for minority languages. Neither excludes the other. If you can, scroll through the many replies to the original thread.

As a linguist working in language documentation, I think the best thing you can do in order to help speakers of minority languages is to enable them to start their own projects. We need more people from linguistic minority groups in linguistic programs! Writing the first dictionary for a minority language is cool – helping speakers to write the dictionary themselves is 1000 times cooler.

---

---