The Problem with Definitions

There's a lot of fretting over definitions going on in gaming blogs, Twitter and so on, lately. It all revolves around what games are, what indie means or what art is (goodness knows what an indie art game is supposed to be). Part of the problem is perhaps we're running before we can walk. We're trying to define words that are inherently tricky, so perhaps it would be a better warm-up to have a go at defining some words that are easy.

Let's try 'table.'

At dictionary.reference.com they have this to say:

Table

1. an article of furniture consisting of a flat, slablike top supported on one or more legs or other supports.

That seems okay, actually. But just to make sure, let's see if we can think of any other articles of furniture consisting of a flat, slablike top supported on one or more legs or other supports. Wait a minute, what about a bench that doesn't have a back? The top of that could be slablike. And that's definitely not a table. Unless it's a workbench, I suppose. But that's different.

We're going to have to fix this definition of table. How about this?

An article of furniture consisting of a flat, slablike top supported on one or more legs or other supports, which isn't intended for sitting on.

Hmm. we usually don't define things by what they're not though. It's difficult to be exhaustive that way. For example, what about a bed? Nowhere does it say the flat, slablike top can't be a mattress. Let's try the other way around.

An article of furniture consisting of a flat, slablike top supported on one or more legs or other supports, intended for the purpose of sitting at or placing objects upon so that they can be used.

That seems a little better, but I keep on thinking of exceptions, still! A bird-table isn't flat. An operating table just about works for that definition, but it doesn't sit comfortably. And I can imagine a purely decorative table that isn't designed to be sat at, or have objects put on it, which means we can't define it purely by its function.

Oh dear. I'm having trouble, aren't I? And maybe that's the point.

As I go about my daily business, I am never worried about what tables are, it's never a source of confusion to me whether something is a table or not. Yet when I get right down to it, I'm completely hopeless at coming up with a precise definition of what one is.

That suggests to me that our understanding of words is not tied to our knowledge of how to define them, but instead to a much more intuitive process of pattern recognition. Coming up with definitions may be of academic interest, but has very little bearing on how words are understood. It's the conflation of definition and meaning that seems to lead to the endless and unresolvable debates and definition-policing that we have been seeing lately.

You might rightly point out that while there's no disagreement over what a table is, there's plenty of debate over art, indie and games. And that's true - at the moment there are different schools of thought on what counts as these things. That's because they're quite complex, abstract concepts unlike that of a table which is an easy enough pattern to recognise even if I can't put it exhaustively into words.

But I'm not saying that there should be no disagreement. I'm saying that attempts to come up with definitions do little to advance our understanding. Rather they hit on exactly the same nit-picking, point-missing problems that the above attempt to define 'table' does.

So let's put a stop to this nonsense that unless the definition for everything is clearly marked out, it won't be possible to have meaningful discussion about it, or everyone will be confused. If I can live with not being able to explain exactly what a table is, we can all definitely live with not being able to explain exactly what art, indie or games are.

It might not be necessary to have a quintessential definition for every term that's out there, but I think that it's important to be on the same page when you're discussing stuff. I would definitely agree that not every word/term can be precisely defined, though. Take the word "fun" when applied to games. Everybody has a different idea of what parts of a game make it fun.

Do you know what the definition of insanity is?

@nickieroonie: Oh, I agree. There are definitely situations where two people might be talking about completely different things and not realise it. But even then, the solution is rarely to come up with a strict definition they both agree on. The solution is to identify the confusion and move on from it.

@winternet: I'm aware of the supposed (dubious) quote. I don't like it, because it's misused more than it's used well. :)

Also if you're going to give examples go and use your Oxford dictionary.

Why? I mean, sure I could have done, but ... (goes to fetch it)

"A piece of furniture with a flat top and one or more legs, providing a level surface for eating, writing or working at it."

This has all the same problems as my modified definition of table from dictionary.reference.com, and also at least one new one: not all tables have legs. Some are just solid blocks (which I guess could be one leg, but it stretches the meaning), while others might be suspended, or come out of the wall, or what about a futuristic hover-table? At least dictionary.reference.com had the part about 'other supports'

So by going to Oxford, I've done nothing to advance the discussion at all. The whole point of my post was that definition gets confused with meaning, and it doesn't matter how much of an authority the person who is doing the defining is. Both the Oxford and the reference.com definitions are fine. They're good. But they both fail to convey the whole meaning because that's not what definitions do.

@kestrelpi: So let me get this straight, the definitions in the dictionary fail to convey the whole meaning of the word, but that's fine because that's not the purpose of a definition. So, what are we talking about? What's the discussion? Why does this thread exist?

I don't think a bird table is an actual table

@kestrelpi:

Defining a physical object that can be touched and observed (like a table) is a completely different than trying to define a subjective idea like art or beauty. I sort of understand what you're getting at but the table analogy is a is a poor parallel to draw between indie games, art, etc.

@winternet: I don't understand why you're in such a twist over it or indeed why the existence of this thread seems to offend your sensibilities so much. I don't mean to rile you, but it seems like a fairly simple point to me. The post is about this:

a) Lots of people are weighing into various debates about what games/art/indie are.

b) One of the things these people regularly try to do is define games/art/indie in a particular way, and then use that definition to categorise what is/isn't that.

c) I don't think that's a very good approach, for reasons listed above.

Seems pretty straightforward to me. I wasn't really trying to be controversial - just saying 'hey, maybe right now these discussions are being framed in a way that's kind of a red-herring'.

Table - a piece of furniture designed for placing objects on.

Why does it have to have a flat, slab-like top?

@ravenlight: I'm not sure the analogy is weak in the way you think it is. All I meant by it is that even easy words are hard to define precisely. In fact, your comment is kind of the point, and I sort of address it at the end: if it's not all that easy to precisely define a physical object, then isn't it going to be considerably more difficult to precisely define something abstract, nebulous and subject to interpretation? And so shouldn't we be looking for a different way to talk about these concepts rather than trying to pin them down in a really concrete way as a starting move in an argument?

@believer258: That definition also applies to a all other pieces of furniture that you put things on. Like a shelf. Or a bookcase. Or a TV stand. (Some of those still aren't ruled out even with the flat, slablike top, which is why this definition business is a bit trickier than it first appears)

The point is that I can quibble the definition of table just as much as I can quibble our definition of anything else, but I still think you know what a table is.

I'm with you man, i'll gather all dictionaries and other sources of knowledge around me and burn this filth!

Uh... :) To be clear, I think dictionary definitions are just fine. They're useful as a reference, as a quick way to get a general idea of meaning, but they're not much good at setting boundaries on that meaning, and so not much good in discussions about where those boundaries lie.

@kestrelpi: I guess I've been missing out on the "what are games/art/indie" threads.

Eh. All right. I get your point, but it's not illustrated particularly well.

Besides, no one is going to argue what is and is not a table because a table is, objectively, a table; whether or not video games are art and why they are is a far more complex and subjective topic.

I know you've already responded to this:

with this:

but exactly what "concepts" and what "other way" are you talking about? Are you talking about the approach to criticism that games get? Yeah, that needs a hell of a lot of improvement, I'll agree. But bringing table criticism to new standards as a parallel to games criticism isn't going to help you much because there's a massive difference between the objective understanding of a table (i.e. you can point to a flat slab held up by four legs and call that "a table") and the subjective understanding of a video game.

@believer258: I don't know how to explain it in any way I haven't already done, but I'll try to rephrase, so sorry - this'll be a long one.

Basically I've seen a lot of chatter lately on the internet lately, debates flaring up in one place or another about what games are, whether they're art, by extension what art itself is, and what it means to be indie.

Inevitably what'll happen is that someone will say (and I'm simplifying here) something along the lines of:

"Well, I think a sensible way to define indie is this: blah blah blah. Based on that definition, X and Y count, but Z doesn't count, which seems about right to me."

And then without fail there will be replies which say 'what about A, B and C?' And then someone else will suggest their own definition or a modification to the definition and the conversation will go on.

On the one hand this might be seen as healthy debate about the state of the medium or whatever, but it seems to me that debating definitions is a very over-simplistic way of doing that. It's like the discourse is wasting a lot of time arguing over what box various things go in that they don't get into any depth about the thing itself. Like, I was listening to a podcast segment on Proteus recently and they spent several minutes arguing about whether it was a game or not and while some interesting points were made about the nature of play, the actual thing they were talking about, Proteus, kind of got shoved aside in service of this meta-discussion. That's what I think definition-debates do, most of the time: distract from deeper analysis.

As for tables and games, I acknowledged in the original blog that they're different things in some important ways. I said: "You might rightly point out that while there's no disagreement over what a table is, there's plenty of debate over art, indie and games. And that's true - at the moment there are different schools of thought on what counts as these things. That's because they're quite complex, abstract concepts unlike that of a table which is an easy enough pattern to recognise even if I can't put it exhaustively into words."

I'm aware that the concept of a table is different from the concept of a game in lots ways. But I was using it as an example of how we approach language in daily life: we don't know what words mean because we know their definition, rather we know what words mean because we recognise and understand how they fit a pattern. The nature of the pattern is something that can be debated (and that's very likely to happen with an abstract concept like games, unlike a concrete one like tables) but attempting to settle the debate by trying to conjure an exhaustive definition usually just ends up with the same kind of nit-picking that I emulated with my table example. Except instead of 'what about a bird table?' or 'what about a snooker table?' it's 'what about this game?' or 'what about this studio?' There'll always be some section of things someone intuitively wants to call art/indie/game that nevertheless manages to fall outside any given strict definition. Seems to me the best solution is to embrace the inherent fuzziness of meanings and move on.

So, finally, and as an example, I think one way of talking about, say, Proteus is having a discussion about whether it's a game or not by trying to pin down what a game is and then seeing if Proteus fits that. But I think this is likely to devolve into nit-picking if the discussion is framed that way. I think a more interesting and productive way to talk about Proteus would be to talk about Proteus. So, for example, one could discuss what motivates someone playing Proteus to explore, and how that's different or similar to other games (or whatever you want to call them). It might not get us any closer to knowing which box to put Proteus in, but we at least might discover something interesting about the nature of play that we wouldn't get to while hung up about which definition it fits. (which, to me, seems like a far more trivial discussion. Even if it was definitively settled, Proteus forever more is/isn't a game... what now? What have we learned from that?)

@kestrelpi:

for what it's worth, i appreciate and agree with the point you're making. welcome to semiotics, language and communication theory. or as it's also known- the unspoken futility of speech :)

Thanks. My background is philosophy (in combo with music, weirdly enough) and I did a fair amount of philosophy of language among other things. While I enjoyed philosophy, I grew to sense increasingly that many so-called problems in philosophy spring from philosophical definitions being used to question everyday understanding. Much later I discovered that this was sort of in line with some of Wittgenstein's ideas.