Minimality is beautiful. Terseness brings it’s own pleasure–in code, in maths, in most things conciseness is captivating. It takes skill to capture an idea and relay it, minimally. Doing so is a compromise, however, and I’ve been thinking about that compromise within the context of language.

Language

Additionally, the framework of minimality offers an idea of what humility is which is an idea I’ve been trying to capture for a while now.

You step through two levels of loss when attempting to communicate a thought to someone.

The first loss is from thoughts to language. Thoughts are abstract concepts couched in your experience and housed in your mind. Most of the time, it is impossible to truly communicate these, i.e. make someone else engage with the exact thought that you are. This should come as no surprise; your entire life provides the context for your thoughts, and so for someone else to truly, wholly understand a (suitably complex) thought you hold, they would need to be you. In practice, though, there are plenty of thoughts which can be easily, perfectly communicated. This is facilitated through language, notably vocabulary. Vocabulary (alongside some grammar structures) is enough to convey simple ideas, e.g. “That is a door.” and can convey a good deal of precision when bolstered by jargon, “e.g. that is a Gala apple.” Syntax, grammar, spelling, etc. all (theoretically) attempt to aid the transfer of knowledge through language by providing agreed upon rules, contracts which can be exploited to encode information. For example, tenses encode temporal information, and rules about plurality and such adjust count information.

The second loss is from communicating language to another person. The thought I have may be able to perfectly be encapsulated by language, e.g. “This computer has 4 GB of 3200 MT/s DDR4 RAM.”, but if you don’t understand the language someone uses, the idea is still not transported. I.e., thoughts that can be translated to language with minimal/acceptable error could incur heavy losses when trying to transport them. The canonical, extreme example would be speaking to someone in a language they don’t understand. Additionally, the problem with language that encapsulates your thoughts is that the language you use is more than just the vocabulary and grammar and so on–you encode personal meanings in language, meanings that can only be replicated by living your life. These latent meanings then may be what facilitate a good translation (thoughts to words) but undermine good transportation (words to other’s thoughts). It should be obvious that some people may share more or less lifecontext with you. As such, the same words can incur different losses at this level depending on the audience–some people may glean your latent definitions and be able to reconstruct them (consciously or not) when decompressing your language into their thoughts. This would decrease this loss.

Call the first loss “precision” and the second “accuracy.” Precision measures how well language represents a thought, and accuracy measures how well language is transported to (a particular) someone.

My idea of a minimal language would be optimizing for accuracy at the cost of precision. In this minimal language, there would be no synonymous. The vocabulary would be small, with “similar enough” words being discarded. The words that remain would have single, precise definitions. There would be no contextual overloading (“bright lamp” vs “bright person”–same “bright”, different meaning). Doing this would naturally increase the size of the vocabulary, but that is not what I think a minimal languages seeks to minimize; I think a minimal language should minimize error incurred when communicating between people. Words should have agreed upon, specific single definitions. Again, this is in pursuit of accuracy, at the sake of precision. In this regard, \we are ruining our minimal language for poetry and creative expression, but its terseness and strict vocabulary would make it very suitable for applications like legal documents or perhaps medicine where it is of utmost importance that the idea is being communicated to you at some level. That is a focus on accuracy. A focus on precision would similarly be the idea that the idea is encoded in language at the best possible level, so as to preserve as much of the thought in language as possible. Minimality naturally stands in this way by taking steps to remove gradations for the sake of accuracy. A precise language would have to be far more verbose, but would be better at encoding latent information.

Whilst this language would be poor from my perspective as a writer and poet, it appeals to me due to its proposed accuracy. I think that having an accurate language, while plain, could be beneficial as ideas could be conveyed quickly and effectively, even if the idea itself must be slightly compromised; I would rather get a fuzzy image than no image at all. I’m not sure how the syntax and grammar of this language would have to be to maximize accuracy (besides being consistent), but I think those are less important than defining the rules of the language itself. It should be easy to see how these ideas could also be interpreted in a programming language context.

Humility

I have also been thinking about humility, viz. what it means to be humble. I think that minimality offers a nice definition for humbleness; humble people speak minimally of their accomplishments. Minimality in this context is used to characterize effacement. The idea here is that one speaks about themselves only to the extent which is necessitated and no further–a humble person will not go on about their skills or such, even if pertinent to the conversation or subject at hand but not strictly necessary. They would speak only as much as needed. This is of course, optimal humbleness, and in real life humble persons needn’t speak precisely minimally as we have defined–there is room for error, and people who fit within these bounds are humble. The magnitude of that error then can be debated, and I think it best to consider it a parameter free to vary person to person.

This idea of minimality for humility was mostly inspired by my observations of the inverse. I’ve observed people add unnecessary context when they say things, which is maximal (i.e. anti-minimal). Saying something like “The rest of the homework was so easy; how do you do question 8?” is identical in effective meaning to “How do you do question 8?”. As such, the latter representation is more minimal, and so a humble person would use that instead. Of course, sometimes additional context is needed–it is just that oftentimes it isn’t. Clearly, minimality with respect to information is a nice way to look at humility.