What is good?

Ian Sales thinks he knows:

I’ve lost count of the number of times I’ve been told “good is subjective” or “best is subjective”. Every time I hear it, it makes me howl with rage. Because it is wrong.

If there is no such thing as good – because if it’s entirely subjective and personal, then it’s completely useless as a descriptive term – then how do editors choose which books to publish, how do judges choose which books to give prizes to, how do academics chose which books to study? And why don’t they all choose completely different books?

The irony here is that I’ve lost count of the number of times I’ve been told that “good is objective”. And yet, no one seems to be able to define what constitutes good. Even Ian, despite his adamant stance, describes what is good in entirely subjective terms.

It is not an exact science, and it is subject to changes in taste and/or re-evaluation in light of changes in attitudes and sensibilities. But there are certain key indicators in fiction which can be used to determine the quality of that piece of fiction.

Having established that there are key indicators that can be used to determine quality, Sales proceeds to list… approximately none of them. Instead, he talks about “taste” and “changes in attitudes and sensibilities” (both of which are highly subjective). If it’s not an “exact science”, how is it objective? Isn’t this an implicit admission that subjectivity plays a role? He does mention some criteria for bad writing though:

Perhaps it’s easier to describe what is bad – if good is subjective, then by definition bad must be too. Except, strangely, everyone seems to agree that the following do indeed indicate that a piece of fiction is bad: cardboard cutout characters, idiot plotting, clumsy prose, tin-earred dialogue, lack of rigour, graceless info-dumping, unoriginality, bad research…

The problem with this is that most of his indicators are subjective. Some of them could contain a nugget of objectivity, notably the “bad research” piece, but others are wholly subjective. What exactly constitutes “tin-eared dialogue”? One person’s cardboard cutout character is another person’s fully realized and empathetic soul.

Perhaps it’s my engineering background taking over, but I have a pretty high standard for objectivity. There are many objective measures of a book, but most of those aren’t very useful in determining the book’s quality. For instance, I can count the number of letters or words in the book. I can track the usage of punctuation or contractions. Those numbers really won’t tell me much, though. I can look at word distribution and vocabulary, but then, there are a lot of classics that don’t use flowery language. Simplicity sometimes trumps complexity. I can evaluate the grammar using the standards of our language, but by those measures, James Joyce and Thomas Pynchon would probably be labeled “bad” writers. For that matter, so would Ian, who’s recent novella Adrift on the Sea of Rains eschews the basic grammatical convention of using quotations for dialogue. But they’re not bad writers, in large part because they stray from the standards. Context is important. So that’s not really that useful either.

The point of objectivity is to remove personal biases and feelings from the equation. If you can objectively measure a book, then I should be able to do the same – and our results should be identical. If we count the words in a book, we will get the same answer (assuming we count correctly). Similarly, if we’re able to objectively measure a book’s quality, you and I should come to the same conclusion. Now, Ian Sales has read more books than me. The guy’s a writer, and he knows his craft well, so perhaps the two of us won’t see eye to eye on a lot of things. But even getting two equivalently experienced people to agree on everything is a fool’s errand. Critical reading is important. Not everyone that subverts grammatical conventions is doing so well or for good reason. Sometimes simplicity can be elegant, sometimes it feels clumsy. Works of art need to be put into the cultural and historical context, and thus a work should stand up to some sort of critical examination. But critical is not equivalent to objective.

Now, Ian does have an interesting point here. If what’s “good” is subjective, then how is that a valuable statement?

If good is subjective, then awards are completely pointless. And studying literature, well, that’s a complete waste of time too. After all, how can you be an expert in a topic in which one individual’s value judgment is worth exactly the same another person’s? There’d be no such thing as an expert. All books would have exactly the same artistic value.

Carried to its logical extreme, the notion that what’s “good” is wholly subjective does complicate matters. I don’t think I’d go quite as far as Ian did in the above referenced paragraph, but maybe he’s on to something.

So far, I have mentioned a bunch of questions that Ian asked, which I will now try to give an answer to:

  • How do editors choose which books to publish? This is a pretty simple one, though I don’t think that Ian will like the answer: editors choose to publish the books that they think will sell the most. To be sure, editors will also take a chance on something that could bomb… why is that? Because I think even Ian would concede that most readers are not even attempting to be objective in their purchasing habits. They buy what feel like reading. The neat thing about this one is that there actually is an objective measurement involved: sales. Now, are sales an indication of quality? Not really. But neither are most objective measurements of a book. The neat thing about sales, though, is that it’s an objective measurement of the subjective tastes of a given market. There are distorting factors, to be sure (advertising, the size and composition of the market, etc…), but if you want objectivity, sales can boil the subjective response to a book down to a single number. And if an editor is bad at picking good sellers, they won’t be an editor for much longer…
  • How do judges choose which books to give prizes to? My guess is that it’s their subjective taste. In most cases, there isn’t a single judge handing out the award, though, so we’ve got another case of an objective measurement of a group of people’s subjective assessments. In the case of, say, the Hugo Awards, there are thousands of judges, all voting independently. There’s a lot of room for fudging there. There’s no guarantee that every voter read every book before casting their ballot (all you need to do to vote is to pay to be a member of the current year’s Worldcon), but since there are usually around 1000 voters, the assumption is that inexperience or malice among voters is smeared into a small distortion. Other awards are chosen by small juries, one example being the Pulitzer Prize. I don’t really know the inner workings of these, and I assume each award is different. I’ve definitely heard of small juries getting together and having a grand debate amongst themselves as to who the winner should be. The assumption with juried prizes is that the members of the jury are “experts”. So if I were to be on the jury for a Science Fiction award, I should probably have extensive knowledge of Science Fiction literature (and probably general literature as well). More on this in a bit. Ultimately, an award is meant to do the same thing as revenue or sales – provide an objective assessment of the subjective opinions of a group of people.
  • How do academics chose which books to study? And why don’t they all choose completely different books? I won’t pretend to have any insight into what drives academia, but from what I’ve seen, the objective qualities they value in books seem to vary wildly. I assume we’re talking about fiction here, as non-fiction probably has more objective measures than fiction.
  • How can you be an expert in a topic in which one individual’s value judgment is worth exactly the same another person’s? I get what he’s going for with this question, but there’s a pretty simple answer here. An expert in a topic will have more experience and knowledge on that topic than a non-expert. Sales has read more books than me, both within and outside of SF, and he’s a writer himself. I would think of him as more of an expert than me. I’m just some guy on the internet. Unfortunately, one’s expertise is probably also subjective. For instance, you can measure how many books someone’s read, but comprehension and contextualization might be a little more difficult to figure out. That being said, individual experts are rarely given a lot of power, and I imagine they would suffer setbacks if they’re consistently “wrong” about things. At their most important, they’ll be a reviewer for a large newspaper or perhaps a jury member. In both cases, their opinions are smeared across a bunch of other people’s thoughts.

The common thread between all of these things is that there’s a combination of objective and subjective measurements. At some point in his post, Sales sez that objective measurement of what is good is “why some books are still in print two hundred years after they were first published.” That’s something I think we’d all like to believe, but I don’t know how true that is… I wonder what books from today will still be in print in 200 years (given the nature of current technology, that might get tricky, but let’s say I wonder what books will be relevant and influential in 200 years)? There’s a school of thought that thinks it will be the high literary stuff discussed by academics. Another school of thought thinks it will be best-selling populist stuff like Stephen King. I don’t think it’s that easy to figure out. There’s an element of luck or serendipity (whatever you want to call it) that I think plays into this, and that I think we’re unlikely to predict. Why? Because it’s ultimately a subjective enterprise.

We can devise whatever measurements we want, we can come up with statistical sampling models that will take into account sales and votes and prizes and awards and academic praise and journal mentions, whatever. I actually find those to be interesting and fun exercises, but they’re just that. They ultimately aren’t that important to history. I’d bet that the things from our era that are commonly referenced 200 years from now would seem horribly idiosyncratic and disjointed to us…

Sales concludes with this:

If you want to describe a book in entirely subjective terms, then tell people how much you enjoyed it, how much you liked it. That’s your own personal reaction to it. It appealed to you, it entertained you. That’s the book directly affecting you. Another person may or may not react the same way, the book might or might not do the same to them.

Because that’s subjective, that is.

He’s not wrong about that. Enjoyment is subjective. But if we divorce the concept of “good” from the concept of “enjoyment”, what are we left with? It’s certainly a useful distinction to make at times. There are many things I “like” that I don’t think are particularly “good” on any technical level. I’m not saying that a book has to be “enjoyable” to be “good”, but I don’t think they’re entirely independent either. There are many ways to measure a book. For the most part, in my opinion, the objective ones aren’t very useful or predictive by themselves. You could have an amazingly well written book (from a prose standpoint) put into service of a poorly plotted story, and then what? On the other hand, complete subjectivity isn’t exactly useful either. You fall into the trap that Ian lays out: if everything is entirely subjective, then there is no value in any of it. That’s why we have all these elaborate systems though. We have markets that lead to sales numbers, we have awards (with large or small juries, working together or sometimes independently), we have academics, we have critics, we have blogs, we have reviews, we have friends whose opinions we trust, we have a lot of things we can consider.

In chaos theory, even simple, orderly systems display chaotic elements. Similarly, even the most chaotic natural systems have some sort of order to them. This is, of course, a drastic simplification. One could argue that the universe is headed towards a state of absolute entropy; the heat death of the universe. Regardless of the merits of this metaphor, I feel like the push and pull of objectivity is similar. Objective assessments of novels that are useful will contain some element of subjectivity. Similarly, most subjective assessments will take into account objective measurements. In the end, we do our best with what we’ve got. That’s my opinion, anyway.