Why a good definition is a bad definition (pt. 1)

Originally I wanted to make this a single post, after all, it is the first real post on this blog. As it happens, the post grew and grew and grew, and after having its foot in the chimney and basically flooding the screen with text, I decided it would be but a mercy to let it shrink into smaller posts. Tiny post in tiny flasks. A pretty sight to behold.

In this first part I will attempt to explain the root of the matter at hand: why a definition is good and necessary.

What is my problem?

I'm pretty sure I can't possibly be the only person with this problem: I want to know what I'm talking about, to know what I'm saying and how it materializes in others' heads. In my case it's less about over-thinking what to say and more about trying to find the essence of things, the arche of the words, if you so will. Or at least it used to be like this until I realized how futile it was to find a coherent system in one of the most chaotic and persistent systems I've worked with: language.

What does this really mean?

In reality the matter at hand is not that vague: I am talking about trying to grasp the semantics of individual words or phrases. Basically I'm talking about good old-school linguistics. The good old stuff that you learn to not pursue once you start studying linguistics in the 21st century. And yet, after dedicating good 8 years of my life to formal linguistics, I still do this. Was I a bad student? Maybe...but I would argue it's more a human thing to try to get a grasp on things and to try gain confidence by doing so. This becomes, in my opinion clearer, when we shift the focus from words to more general subjects, by this I mean, in a Wittgensteinean sense, the world, everything being the case, the totality of it and its individual parts: concepts, objects, words, situations, individuals, places. Let me try to get more explicit, more concrete, by naming some typical situations in my life:

I used to meet people I like (pre Covid, obviously) and get drinks, talk about our personal perception of things, discuss matter, point out problems with politics, society, media, the usual stuff. Those conversations sound, at least to me, fun. And most of the time they were. But time and again a big part of the conversation revolved around getting the terminology straight. And by this I mean, by trying to find a common ground for all people to agree on what we were talking about. For example: in order to discuss art a definition for art is needed. This allows the party to either acknowledge or disqualify something as being or not art and makes the conversation possible. It allows someone to reject or accept ideas like 4-D cinema, makeup, or even tik-toks being artistic expressions. Probably the a priori answer to this would be "well, it depends" and it totally does. The parameters it depends on are the very definition of what we perceive and judge, sometimes with reliable information, sometimes quite subjectively so, to be art. The definition of art. And thus, we need a definition. Sounds familiar?

During work meetings (I work as a developer) the idea for a wild new feature for a product is conceived. Someone coming from a non-technical field (of course) comes up with this new brilliant idea and tries to convey it by explaining the value this would add to the customer. Then someone else argues the idea to be great but being unnecessary because none of our customers really has that use case. Then someone else points out the persona of our customers is not yet defined, or blurry, and thus give the impression of the use case being not existent. And people start arguing about that definition and its validity. And then a technical person (of course) points out that the whole idea can't be implemented because the project we are working on assumes some parameters that are not yet the case, for example the lack of customers. So the discussion shifts to defining the difference between customers and users only to find out: that was not the subject of the meeting. Sounds familiar?

While writing an article, this blog for example, I ask myself: If I use the word arche, will I be understood? I mean, I don't know whom I'm writing to. Will they think I'm yet just another arrogant idiot wanting to feel important and writing big words? Or maybe yet other readers will think asking myself these questions is basically just wasting my thoughts on irrelevant stuff because, after all, getting a word's definition in these days is just a matter of 2 clicks/taps and whoever wants to get information is quite enabled to do so. Sounds familiar?

I happen to come across someone who, say, just dropped something on the street. I want to let this person know about this thing they lost and approach them (with proper distancing, that is). Then I ask myself: what language should I use? Living in a multicultural environment, making any assumptions about this is not necessarily playing it safe. And then the next question, if x language, then what form? Should I be formal or colloquial? And maybe, by virtue of the language I picked (assuming it to be the right one), I might have to assign a gender to the addressed person. Let's assume, for the sake of the experiment, I pick Russian and say something like "Hi, you lost (verb: past, feminine) your wallet. Here." instead of saying "Hi, you lost (verb: past, masculine) your wallet. Here." And make this assumption by virtue of my knowledge that certain body features combined with certain movement pattern tend to denote a person with a certain gender. And based on that assumption I might judge that person with certain clothes and ethnic features to speak a certain language. In a word: I am defining this person before even interacting with them. Big mistake, right? She was a Polish guy. Sounds familiar?

Maybe I'm really the only one wasting its time with these useless doubts. And maybe I'm not the only one and you too can relate to any of my examples above. And maybe, but only maybe, these aren't just doubts, but the expression of something really human that we have increasingly gotten attached to: The need for a definition.

Safe in a box of inference

"A new born child has no teeth."



"A goose has no teeth."



"A rose has no teeth."



This last at any rate—



one would like to say—



is obviously true!



-Roses And Teeth For Ludwig Wittgenstein, Matmos

A definition allows us to talk about things and gives us the certainty to be talking about the same thing, of being on the same page. The imaginary box that defines the good tone, the scope, the subject. Only if I know anything about the person I am talking to I can be sure about not assuming information I know: like the language they (and by extension we) speak, about their gender, about their preferences about politics, media, lifestyle, and their opinion on certain subjects. This would then suggest that the problem at hand would be lacking this information. And since we can not possible know the definition to all subjects (people, concepts, things) in this world, there has to be a method to deal with this. And there are, in fact, many.

A method would be, for example, to look up a definition of a concept on urbandictionary a serious encyclopedia and then tell everyone about how are the only person knowing the correct meaning of being wetter than an otters pocket. A valid method. A risky method I shall discuss an other time.

An other method would be trusting one's experience and knowledge, to infer a subject's semantics by looking at the context. We do it all the time. Last time, for example, while eating a delicious quako, it hit me. Don't look up quako. You won't find it, I made it up (it's something like a taco, but with quantum spice). But I'm sure you immediately knew had to be some food (unless you assume I am some kind crooked being eating non-edible something. And then again what is food if not something that is eaten?). This is a innate ability we humans have. This ability can be observed since early childhood and basically accompanies children during the progress of language acquisition (when children don't have a clue what those hairy apes with nursing breast are talking about, but it has to be nice because...milk) and beyond. If you feel like it, you can take a look at the famous Wug Test. But I digress. The point I am trying to make is that we humans are used to inferring information. We excel at creating drawers and putting subjects in them as a guess to make sense of the information at hand. And this has quite some advantages. It allows one to tell doctors away from nurses, the staff in a restaurant from other customers (seriously, have you ever stared at a random person in a bar and ordered a drink? It's extremely awkward and, depending on the context, quite rude), and potential partners from a potential slap (if they say 'no', accept it, trying to force your gay love onto your new homophobic crush won't get you anywhere but, perhaps, the next hospital). All these things we normally do quite instinctively and tend to be good at. They allow us to find our way through the puzzle of reality.

Does this mean that either knowing what or who we are interacting with is the safest situation, followed by being knowledgeable and being able to classify unknown objects, ideas, individuals? I would argue: it depends. But this, to be honest, I don't know. And still I shall share my thoughts with you, all the same. In the next part of this post I will try to explain why I think having a 'good definition' based on knowledge might be a problem itself.

Feel free to keep on reading the second part in which I try to explain why think that a definition is as dangerous as no definition.