What is a simple language?

Any language which avoids any redundant words


This an example to begin, many languages do not have separate words for poison and venom, when talking about something poisonous one must specify what type of poison and where it comes from or simply leave it to the listener to use their intuition.

Romance languages, germanic languages and English have roughly half a million words each even though most people use less than 600 words on a daily basis.

Less than 1% of the dictionary is being used, the remaining is redundant.

Courage for example has more than 17 synonymous words with the exact same meaning.

A simple language can have subjects, verbs and articles in any order and the phrase still make sense.

A simple language has words that make sense regardless of tone or accent. One shouldn't have to be a singer to be able to speak. (see French or Mandarin and Danish for example)

A simple language should not force information into grammar, in many languages one can not speak without specifying time, gender and sometimes location or quantity. The amount of information to be shared with each frase has to be optional and not enforced by grammatical rules.

Now that I have defined what a simple language means, how does one maintain a language simple over the centuries or even millenia?

This is not a question about how to freeze a language in time so it never changes but on how make to it so that the primitive proto language and the modern language will share the same simplicity.

  • 1
    $\begingroup$ Comments are not for extended discussion; this conversation has been moved to chat. $\endgroup$
    – L.Dutch
    May 27, 2021 at 4:00

6 Answers 6


Frame Challenge About Simple Languages

Some of your assumptions about "simple" languages are not as simple as you may think.

A simple language can have subjects, verbs and articles in any order and the phrase still make sense.

A simple language should not force information into grammar, in many languages one can not speak without specifying time, gender and sometimes location or quantity. The amount of information to be shared with each phrase has to be optional and not enforced by grammatical rules.

Languages with looser grammar rules require more clarifying language to make since. So lets pretend you tried to say "He has a daughter with a yellow hat" in such a language, it could be said:

  • With he daughter has yellow hat
  • Daughter has with he hat yellow
  • Has yellow hat he with daughter
  • He yellow daughter hat with has
  • etc...

As you can see, there is no way to communicate what the words mean together with grammar thrown out the window. The only way to disambiguate the language would be to make it more complex by adding a bunch of prepositions or conjugations to tell you how the words relate. So to have clear meaning the language would then look like:

  • Proposition-hat-with subject-he object-daughter verb-has adjective-hat-yellow object-hat
  • Object-daughter verb-has proposition-hat-with subject-he object-hat adjective-hat-yellow
  • Verb-has adjective-hat-yellow object-hat subject-he proposition-hat-with object-daughter
  • Subject-he adjective-hat-yellow object-daughter object-hat proposition-hat-with verb-has
  • etc...

Now you have made the sentence able to be clear without most of the time without grammar rules, but in the process, you've made the language very unnecessarily complex. Grammar in most cases simplifies languages by giving you a structure on which to make common assumptions about the relationships between words.

Grammarless language is also unpreservable because it will naturally start to form thanks to common biases. Eventually groups will settle into word orders based purely on how other people choose to order their words.

Instead what you want is actually a very strict Grammar, one without exception or irregular conjugations. So, if a noun can end in a conjugation that implies gender (like: male, female, and unspecified) it should ALWAY NO MATTER WHAT end in a gender conjugation, then you can get rid of all those pesky unmatched gender specific words like butler/maid/housekeeper. Instead you are a maido, maida, or maidum: 1 word that can be predicted and fit according to a common rule. Or you can treat gender specification as a separate and optional word that is always appended in a fixed position relative to a noun so you could have a "he maid", a "she maid", or an "it maid". The reason you don't want optional NULL conjugation is so that you don't get concatenation collisions. For example let's say you have two base words "med-" and "medum-". By making Null conjugation optional, "medum" could mean either "med-" with the "um" sufix or "medum-" unspecified. But if "medumum" was the only proper conjugation for "medum-" then you are less prone to language collisions.

As for Maintaining a Language

Historically this has proven very elusive because most people learn to speak from non-canonical sources, but today we have the ability to establish canonical language standards much better. Especially if you were to replace English teachers who each have thier own biases with a standardized language teaching AI. To freeze a language, you need an official language guide that is taught as a part of your culture's compulsory education system. When you first introduce it, you need a guide that tells you what the canonical rules and words are. A good modern example of this is Octopuses an Hippopotamuses. For a long time, these English words retained their Latin plural conjugations of Octopi and Hippopotami, but the public schools started teaching English regular conjugation of these words and within a few decades, the English regular conjugation has become the preferred method.

Basically you do this, but on a MUCH grander scale. You take the language everyone knows and you take a hack-saw to in in public schools. If a student writes "avion" instead of "bird" on his biology paper, that's -1 point. If he writes butler instead of maido, that is -1 point. Basically, you do what we are already doing in schools to homogenize language, but with the added directive to also simply it.

In this process you can also weed out same or similar sounding words by choosing cannon words based partly on their uniqueness. "Their", "there", and "they're" would be unacceptable. So, "there" could be canonized to "that place" and "they're" to "they are".

Lastly, you write down all the rules and make sure make sure the governing body responsible for language is tasked with the mission of preserving language rather than adapting to it. The problem with things like Merriam-Webster or Oxford dictionaries is that it is their goal to reflect the current meanings of words; so, they require panels of experts to constantly add & subtract words. When I was little, there was a common saying that "ain't ain't a word 'cus it ain't in the dictionary", yet today ain't can be found in many dictionaries because of its common usage. Instead whoever is in charge of your nation's dictionary on which children are taught the one-and-only acceptable method of speaking, it is their goal that when a new word starts to arises, that they will try to fit it to canonical word. So when people start using new words things like "Chillax'en" or "cellphone" the public board of language will examine the word to decide if it is infact a new thing or if it needs to be canonized to a proper word. So "cellphone" might get to become a new word because it did not exist before, but "Chillax'en" would be officially canonized to "resting" along with "chilling", "relaxing", "hanging out", "loitering", "idling" and any other word that means to spend awake time doing nothing of interest.

Under this system, colloquialisms will of course still pop-up because it is human nature, but they will also die out reverting back to the original language with each generation that passes.

  • $\begingroup$ "To freeze a language, you need an official language guide that is taught as a part of your culture's compulsory education system": Actually this doesn't work, and we know it doesn't work. For example, France has had an excellent and comprehensive public education system for about 200 years; this system teaches "official French" (a very slightly updated form of 17th century upper-class Parisian French), and it expends a lot of energy for this purpose. This did not stop the evolution of spoken French, to the point that today "street French" and "standard written French" are very different. $\endgroup$
    – AlexP
    May 26, 2021 at 22:20
  • $\begingroup$ Try going to YouTube and find interviews with ordinary French people; hearing José Bové speak (the hero of the revolt against fast food) will be illuminating on the difference of what French people actually speak and the pristine frozen perfection of official French. If you can understand a word of what he's saying. $\endgroup$
    – AlexP
    May 26, 2021 at 22:24
  • 1
    $\begingroup$ Heresy, I say! Everyone knows "here" and "there" go together. It's "their" that's irregular; obviously that should be "them's" 😁. (And "him's". At least "her's" is already correct.) $\endgroup$
    – Matthew
    May 26, 2021 at 23:58
  • $\begingroup$ @AlexP not necessarly it does not work, as govs do not have mission to prevent evolution of language. If they do not include fusion reactors in their vocabulary they are doomed. Max utilitarian purpose is to speak more or less same language at the same time/decade/epoch and establish research some logic of changes, establish some more or less uniform rules for those changes and their evaluations, reforms if required etc. To have some set of tools to navigate and for a situation to not get of hands, or solve practical problems like official italian instead of 40+ dialects etc. $\endgroup$
    – MolbOrg
    May 27, 2021 at 1:18
  • 1
    $\begingroup$ Also, 200 years ago there was no such thing as high fidelity audio recordings. With modern technology, you could have a canonical audio library that maintains, unambiguous proper pronunciations for hundreds of years because you don't need an oral tradition to pass it down, you could do a lot to homogenize a language for example if every student was taught by the same synthesized voice: think Alexa. $\endgroup$
    – Nosajimiki
    May 27, 2021 at 4:49

Well, I must say that this would not really work to begin with, it would be unable to act as a way of communication, as it would bring about great degrees of vagueness and overall lack of coherence.

As mentioned by another person, you can't completely deal away with grammer or encoded information. As this would bring about incoherence and great degrees of vagueness, and ultimately would make it fail as a language as communication through it becomes impossible.

What you ask for can't really be maintained, people would ultimately push towards a better form which allows them to better deal with the vagueness of language. Grammer would ultimately come into the language, either through unintentional shift, or through an intention addition of it in order to deal with the problem of vagueness and incoherence, and you really wouldn't be able to avoid this.

Also, what you said about Word Order is not simple. Free Word Order requires far more complex rules and inflections, and it is ultimately far more complex than a set word order. Not simple in any way, shape, or form.

Gender can be optional, that is alright. Various language do lack grammatical gender entirely. But tense and such must be encoded in some manner, even languages like Hopi have ways of showing tense, despite what some might say, just reveals it in different ways.

A complete lack of encoded information would make this fail as a language. What would be better is just a lack of inflection, as seen in analytic languages such as Mandarin, this uses articles or adjectives instead of inflections, therefore look towards it for the kind of language you seem to be looking for. Because what you describe would not function as you seem to think it would.

  • $\begingroup$ Mandarin, or indeed English. Modern English is quite advanced in its journey towards an isolating language; compared with the other related European languages, the morphology of English is extremely impoverished. (And even Mandarin does have a tiny little bit of morphology.) $\endgroup$
    – AlexP
    May 26, 2021 at 22:12
  • $\begingroup$ Yes, English is Inflective to a degree due to it having complex inflections in Old English. And you can still see some remnants of it today. Mostly in pronouns. $\endgroup$
    – Zoey
    May 26, 2021 at 22:13
  • $\begingroup$ And yeah, Mandarin does have a degree of morphology. It is just the closest language I can think of to what OP asked for. $\endgroup$
    – Zoey
    May 26, 2021 at 22:19

You've got some serious issues, and let's start with your example.

You indicate "poison" and "venom" are redundant because they mean the same thing, "toxin". Only problem, they aren't the same thing at all. To use a crude definition.

  • A venom is a poison generated by a living organism that is intended to cause harm to another organism. Snake venom, obvious example.
  • A poison is a substance that causes death, injury, or harm to an organism through molecular effects if a sufficient dose is delivered.

Among the substances that can act like poisons and kill you: water, oxygen, and vitamin D.

Substances that are required to live: water, oxygen, and vitamin D.

So while water, oxygen, and vitamin D can act as poisons, they clearly aren't venoms, even putting aside the nature of their production. There is a difference in their meaning that's sort of important. A venom is intended to harm. Something like vitamin D is necessary to live, and only becomes harmful if you take too much. That's the problem with concepts of limiting the size of vocabulary to "the most common words" and such; you by definition eliminate nuance that people need on a regular basis.

The only way of avoiding it is to use a phrase: so to use your example, where "venom" and "poison" just mean "toxin", you might use the following phrases instead when you need to specify.

  • A snake uses toxin that it produces and injects by biting to hunt.
  • The snakes were killed by human made toxin in the water.
  • The person died of toxin from having too much nutrient.

You can see the problem, yes? So instead of simply using the word "venom", you have to specify "snake biting toxin" or something similar. And eventually, humans being humans, that phrase will be shortened and compressed, just like so many words that originated in complete phrases, and it eventually become a word that means "venom". Which means you've achieved exactly net zero in minimizing the language.

Here, perhaps, is the question you might want to ask yourself: why are there so many words in languages? No one goes around forcing people to adopt new vocabulary, they do it all the time. And the reason they do it is because they see a need for it.

Let me give you another example: Let's talk about Bob. Bob is in a state of mind that very broadly, could be described as "happy". And in every example, I'll use words that Thesaurus.com declares are synonyms for "happy". Now tell me if you think they're all really describing same thing:

  • Bob was happy that Charlie failed to get the job.
  • Bon was ecstatic that Charlie failed to get the job.
  • Bob was pleased that Charlie failed to get the job.
  • Bob was gleeful that Charlie failed to get the job.

They may all be synonyms for "happy", but they are all implying very different states of mind and actions on the part of Bob. "Happy" isn't sufficient to accurately describe Bob.

Or consider a more loaded example:

  • Alice was infatuated with Bob.
  • Alice was devoted to Bob.
  • Alice was obsessed with Bob.
  • Alice lusted after Bob.
  • Alice cherished Bob.
  • Alice worshiped Bob.

A little more nuanced than a simple "Alice loved Bob". Humans do that with everything, and trying to prevent it is pure futility.

  • $\begingroup$ one organism could inject air in their prey to kill them, thus air is a venom. $\endgroup$
    – user85816
    May 27, 2021 at 9:10
  • $\begingroup$ one can eat too many snakes and get snake poisoning. $\endgroup$
    – user85816
    May 27, 2021 at 9:10
  • $\begingroup$ @furtyfur, but in that case it wouldn't be a poison, since the way injected air can kill isn't by a molecular effects but by a mechanical one, namely preventing blood flow from reaching cells by physically blocking it. $\endgroup$ May 27, 2021 at 14:14

The primary means to constrain language is to constrain thought, because new thoughts and ideas drive language to evolve and expand. If this is something you doubt, look at totalitarian regimes. Consistently, totalitarians seek to minimize and simplify language citing complexity as the tools of oppression used to fool the populace into the subjucation that the totalitarians are saving the populace from.

Similarly, Marxist theory, and its derivatives, attempt to redefine words to drastically narrow their meaning, stripping nuance from discussion, painting any topics into stark relief of right and wrong.

Want to keep a language simple, attack and vilify nuance and compromise in the culture.


Limit the number of words

First get the top 3000 words in a sample language.

remove all synonyms and all opposites.

For example, if you include the word "fast", don't include "slow" or "quick". If you include the word "big" don't include the word "small". If something is average sized it is "not big not not big"

Second, identify 32 mouth sounds.

8 consonants : B F G H L M P R

4 vowels: A E O U

So you have BA through RU to make 32 unique syllables

Then match each syllable to one of the top 32 most common words.

Then match a unique couplet of two syllables to each of the top 33 to 1056.

You now have a language with 1056 words that can describe most things. For example, you can describe The Saturn V with these words.

However, this language will have some problems.

This language will refer to computers as "number boxes". If you want to know what powers them that would be electricity, or "bright power". You get these though the internet, or "big machine group" on websites, "information places" like this one. In exchange for having a simple language that has no repeats, you now need to describe everything to make it work. Because of this this language might actually be slower in some cases. If you run a business of several types that have long descriptions you might be tempted to call one kind of business a "business-A" another a "business-B". Then this becomes common use and people will want to go to the "business-A" to get some "product-C", and now you have more words that might be synonyms.

Keeping language simple

Make it so people value speed in their language, so saying longer words that are 150% as long as a normal word is looked down upon. Because of this you will have a simple language with a small amount of words that can be used to build more complex ideas. Also, make it some people are intentionally vague with their language. If people want details they will want more complex words that are synonyms but are different in a specific way.

  • 1
    $\begingroup$ The problem with this approach is one linguists have long noted. "Not fast" meaning "slow" just means that you need to know "not fast" as a unit. And humans being humans, "not fast", being treated as a unit, will eventually transform into a compound single word notfast, then you might lose the first "t" and it becomes nofast, and then the second "t" might go, and now you have the word nofas. Which means "slow". But you still have the separate words "not" and "fast", which means your vocabulary has increased. $\endgroup$ May 27, 2021 at 5:14
  • 2
    $\begingroup$ Is this answer inspired by Randall Munroe's book Thing Explainer, which uses the "Ten hundred most common words" to describe things? It quickly illustrates why you want more words, for example, "thousand" doesn't make the top thousand words, hence the description. (Also, I think you meant 8 "consonants" rather than "constants.) $\endgroup$ May 27, 2021 at 5:43
  • 1
    $\begingroup$ The answer is inspired by Up Goer Five, a computer science lecture I once saw on cathedral vs bazaar architecture where the presenter only uses monosyllabic words or words that he defined in the lecture, and experimentation I have done in Conlang. Obviously bad conlang since the system I described is something I would consider semi-legitimate, but still. $\endgroup$
    – user64888
    May 27, 2021 at 6:43
  • $\begingroup$ Also, yes, as I mentioned you will always be pushing back against people making short cuts since stuff like business-A and business-B short hand will happen eventually. $\endgroup$
    – user64888
    May 27, 2021 at 6:44
  • $\begingroup$ And that pushing back is doomed to failure. Want proof? Always is a contraction of a phrase. $\endgroup$ May 28, 2021 at 19:48

If you want to keep it simple, limit the culture's expansion and discovery potential so they can't come up with new words.


You must log in to answer this question.