Insomnia | Commentary

How Good Exactly is Perfect?

By Alex Kierkegaard / June 1, 2009

The perfect is the enemy of the good.


One of the issues that regularly comes up among those who pretend to care about game reviews is that of ratings. Here's a tip for picking out the idiots, next time you find yourself witnessing such a quarrel: it is the guys who are against them. But why are they against them, and why are their objections idiotic? I'll start with the second question, and work my way down to the first.

Professing to be against ratings is idiotic, because ratings, like it or not, are automatically embedded in reviews by the very process of their composition, and there is no way to unembed them.

A moment's silence please, to give this profound insight time to sink in.

The issue is as simple as that. A review is an opinion, and will as such necessarily be either favorable, unfavorable, or, occasionally, ambivalent. Hence every review ever written -- and that will ever be written -- uses a rating system out of three, with, say, "1" being unfavorable, "2" ambivalent, and "3" favorable. Hence Balzac's review of Stendhal's La Chartreuse de Parme (1839), though it did not feature a rating, certainly contained one ("3"), whether Balzac likes it or not, and if he has a problem with that you can tell him that Alex Kierkegaard said he can go fuck himself.

So, essentially, the question is not whether or not to ADD ratings to reviews, but whether or not to HIDE them. Because that is what has happened whenever a review does not feature a rating -- the rating has been hidden, and the only way for the reader to discover it is to read the actual text.

The pseudo-intellectuals will of course have hit the roof by now all the while yelping that that is why reviews are written in the first place -- to be read. Certainly, but what the visible rating does, amongst other things (which we shall shortly examine), is help the reader make a more informed choice on whether or not to read a specific review, and when. I give a few examples.

1. "Universe of Goo" has just been released, and Johnny Casual, having greatly enjoyed the original, is wondering whether he should buy it. He checks in on his favorite videogamer site (say, ""), sees that it got 10/10, and then, instead of reading the review -- which, like all reviews, will necessarily spoil the game to an extent -- rushes out to buy it. After he's through with the game he sits down to read the review, and makes up his mind on how closely his own views correspond with those of the reviewer. (This, incidentally, is the correct way to treat all reviews of potentially great games, and not only of games.)

2. An alternate version of the above scenario, in which "Universe of Goo" gets 3/10 (fat chance of that happening, I know, but bear with me). This time, instead of straightaway rushing out to buy the game, Johnny decides to read the review first and make up his mind on whether to go ahead regardless with the purchase, or whether perhaps to pick up something else instead.

3. Johnny again, having just figured out how to set up MAME on his computer, is itching to try some of arcade history's greatest hits. He logs in at Softcore Gamer 101, sees that it contains a whopping 2,739 arcade reviews, sorts them out by rating and by genre, and within seconds has a full range of up-to-date lists of the site's most strongly recommended arcade games across all genres. This way he is spared the ordeal of having to read 2,739 fuckin' reviews before he can even begin playing. As he then slowly goes through the games, he can pick and choose which reviews to read and at what time.

4. Same scenario as above, but this time Johnny and his friends are pissed out of their brains and looking for some crap games to laugh at. They sort games by lowest rating and quickly get their list.

I could keep going, but I trust the above illustrations will suffice. It is essentially impossible to run any kind of decent review site without a rating system of some kind, and anyone who objects to that is either really, seriously retarded, or else runs a crappy blog with five reviews and six readers, and has therefore never had to worry about this sort of thing (though still, one would think that, as a gamer of some kind, he would have found himself in at least ONE of the above scenarios at some point or other in his life, and thus realized the helpfulness of ratings. But perhaps that is asking for too much: most people are barely able to put two and two together, let alone make such complex and intellectually demanding inferences).

When the pseudo-intellectual, therefore, or the artfag (because these are mostly the kinds of people who are against ratings), refuse to add ratings to their reviews, they are basically:

1. Forcing you to read through their entire little abortive attempts at criticism, even though, at the time, you might only be interested in simply finding out if they liked or did not like the game in question.

2. Relieving themselves of the labor of having to make up their minds on whether or not they liked the game. They simply spew out their random, incoherent ramblings on the page, and leave the job of figuring out what they are saying (if anything) to the reader.

3. Leaving open a back door to weasel out of later, in case they get blasted for incompetent criticism and/or poor taste (the size of the door depends on the vagueness of the text -- a typical NGJ text, for example, does not merely leave a door open, but demolishes the entire critical edifice and lifts the reader up to the skies, taking him on a flying adventure amidst the clouds, where nothing is of substance and where the slightest gust of wind can turn the travelers around -- without them even realizing it, since at that height there exist no features by which to navigate, this being the preferred environment of bad navigators and scatterbrained people of all stripes).

Because a rating, you see, is a sign that signifies a judgement -- though a sign which, unlike the text of the review (which is also, as mentioned, supposed to signify a judgement) does not require interpretation. Ratings are unambiguous, and that is why they are such valuable aids to the reader, whose powers (or weaknesses) of interpretation are held in check by them, and who is thus prevented from misinterpreting at least the main point of the reviewer: his ultimate judgement. And it is precisely for this quality that ratings have, their complete lack of ambiguity, that bad critics despise them, for, as Schopenhauer explained, poor thinkers love ambiguity and are drawn to it by instinct:

"It is also a characteristic of such writers to avoid, if it is possible, expressing themselves definitely, so that they may be always able in case of need to get out of a difficulty; this is why they always choose the more abstract expressions: while people of intellect choose the more concrete; because the latter bring the matter closer to view, which is the source of all evidence."

And as ratings are a bane to poor and dishonest critics, they serve as aids to the good and honest ones, for they act as a restraint, helping them stay grounded and focused on the task at hand, tirelessly working at eliminating, or at any rate minimizing ambiguity also in the text itself, rendering judgements clear and unambiguous no matter the diversity and scope of the arguments brought to their support. Thus the good critic's rating will always accord with his text, whilst when the bad critic condescends to give a rating it often clashes with it, this being among the reasons so many of them don't -- for they are well aware that there is no plainer sign of a weak and confused critic, even to the eyes of the inexperienced reader, than a disparity between text and rating (as can be seen, for example, in many NGJ reviews).

A critic's refusal to provide a rating is basically a cop-out on a fantastically wide range of levels; he is equivalent to a friend who, when asked for his opinion of a game, or a movie, or a novel, responds by telling you to hold on a few hours while he goes away to write a 3,000-word essay on the subject. How much value could that essay possibly have, coming from someone who is so out of touch with his own feelings he is incapable of summarizing them? And though the friend should perhaps be excused his incapacity and confusion, for he is presumably making no claims of authority and expertise; he who wants to pass for a critic, he who wishes his word to count for something beyond and above those of everyone else -- should never be.

But where does this modern (and as such intrinsically stupid) reaction against ratings come from? Because it clearly is a modern reaction, since all the old gaming mags employed ratings and no one ever complained.

The reaction comes, on the one hand, as mentioned, and for the reasons already described, mostly from the pseudo-intellectuals and the artfags, who are themselves a modern phenomenon -- but there is more to it than that, because even normal gamers can often be seen exclaiming against ratings nowadays. So why is that? What is their problem with them? Their objections are based on very different grounds from those of the pseuds and the artfags; they are actually a symptom of despair in the face of two modern problems:

1. The rampant dishonesty of professional reviewers, who, sometimes finding themselves unable to ignore or conceal a game's more flagrant failings in the text of the review, are forced to mention them, but who nevertheless proceed to give the game a high rating in the knowledge that this will still mislead most of their readers, and thus keep their advertisers happy. This approach also serves to contain the reaction of most of their more intelligent readers (who are anyway in the minority), who, unable to bitch about the text of the review, are confined to bitching about the disparity between text and rating -- a negligible headache for the reviewer, considering the overall advantages of this devious little scheme.

2. The inanity of the percentile and decimal rating scales, which by their nature compel even the honest reviewer (let alone the dishonest one, who in such matters requires no encouragement) to restrict himself to the upper part of the scale, so that in practice the vast majority of games end up in the 60 to 100 (or 6 to 10) range, giving as it were a thumbs up to all games, and thus rendering the entire rating enterprise pointless.

Both problems, of course, existed to an extent even in the old days, but the ever-increasing deceitfulness and cheerleading approach to criticism of today's professional reviewers have greatly exacerbated them, and it is to this development that gamers are reacting. But it is a knee-jerk reaction, a fundamentally senseless one, to blame the problem on the concept of ratings instead of on those who are misusing and abusing them. It is the equivalent of throwing the baby out with the bath water; and besides, you can't make a dishonest critic honest just by taking away his right to give out ratings; you can't cure the disease merely by treating one of its symptoms -- you might at most conceal the disease, but why would you want to do that? And besides, this "solution" is impractical, for the rabble demands ratings and wouldn't pay any attention to anyone who didn't provide them. Here is at least, for once, a case in which its instincts are correct.

But let's explain, once and for all, the problem with the percentile and decimal rating scales. Let's first of all note that scales out of ten with decimal points (i.e. 7.5/10 and the like) are still percentile scales, and scales out of four or five stars which include half stars (used mostly by film reviewers and film reviewer wannabes) are still decimal scales -- these are just silly tricks to fool the feebleminded. The only tenable rating scheme is the one out of three (favorable/ambivalent/unfavorable) I mentioned in the beginning, which is the only one that occurs to man by nature. That's how people respond when asked their opinion on pretty much anything -- no one goes "hmmm, 76,379 out of 100,000" when asked whether they liked a certain film. At most one could expand the scale by adding two more gradations: "highly favorable" and "highly unfavorable" -- but no more than that. Adding any more will always be humbug, because a person is not able to quantify his sentiments to a greater degree: we are human beings after all, not machines -- our value judgements are diffuse, uncertain, fluctuating. What is the difference between a 7 and an 8 game? Can anyone spell it out in human words? Let alone between a 72 and a 73!

The greatest drawback of these retarded rating schemes, however, is that they induce in the reviewer the delusion that what he is rating against is perfection. In the natural rating scales out of 3 or 5, nobody would suppose that the highest mark is reserved for "perfect" works, but the moment you move to a decimal or percentile scale people begin hallucinating about "perfect 10s" and "perfect 100s". You only have to observe the reaction of, say, the British gaming scene (rllmuk, NTSC-uk and the like) whenever Edge magazine gives out a "perfect 10" to realize the negative consequences of this inane delusion. There's no difference between a 9 and a 10 you fuckin' imbeciles! All those 9s the Edge morons give out are also "perfect 9s", if not in fact MORE perfect than the 10s. The 9s (and the 8s, and the 7s -- not to mention the 6s, which I will shortly mention!) are usually the more ambitious, more innovative games, which might perhaps have one or two (easily disregarded) minor problems that in the eyes of the Edge pedants keep them from "perfection", while the 10s may be less ambitious games which are however more polished, lacking these "imperfections". And since, as long as you are pedantic enough, ALL games have "imperfections", you end up randomly giving out 10s and 100s once every few years just so that no one can accuse you of... pedantry. This leads to some hilarious results, one of the most notable of which being that Edge magazine, the world's most respected, pseudo-high-brow videogame publication, gave the original Grand Theft Auto III a 6 on release, essentially advising readers to not bother with it. NOT PERFECT ENOUGH FOR THEM I GUESS LOL! LONG LIVE CONKER'S BAD FUR DAY!

To recap: Perfection is an empty concept -- a mere word, because there will always be the chance that a "more perfect" work may be produced at some future time (not to mention discovered in the past!) -- and where will your "original perfect" work be then, eh? So a rating of 3/3 or 5/5 implies no claims of perfection; such ratings merely signify that -- as things stand at this point in time, and given his gaming background -- the reviewer is strongly recommending the work in question. Who knows how things will stand in a thousand year's time? To require that a rating should be valid until the end of the universe (which is what the term "perfect" implies) is unfathomably idiotic -- so let's leave it to the idiots then.

Oh and, by the way, for reference, here are the rating guidelines I give to Insomnia's contributing reviewers:

***** Highly recommended
**** Recommended
*** Good, but has been done before, and much better
** Playable, but without much merit

One point remains to be touched on, and it's indeed, as one would expect, since I left it for the end, the subtlest and most delicate one. We must realize the difference between a review and a critique. The dictionaries are of no use here because they regard the terms as mostly interchangeable -- and nor are they mistaken: in the grand scheme of things they indeed are. Yet at this point, for reasons I will explain at a later date, we are obliged to make at least a provisional distinction. By "review" we should refer to a critical essay which attempts to place a specific work within a hierarchy whose construction is always a work in progress. Such essays must necessarily confine themselves within the scope of the hierarchy under construction, with any references to greater issues being off-topic and undesirable, because they do not in any way contribute to the hierarchy's construction. Such essays, as we have seen, should also always come with ratings, especially if they are produced in great numbers and with any degree of regularity -- as they must in order to better contribute to the construction of the hierarchy.

A critique, on the other hand, while still described, like the review, as a "critical essay", does not confine itself within a specific medium, nor does it bother with constructing hierarchies, but proceeds to place the work within a larger framework -- indeed almost the largest: that of human culture and civilization (the largest would be that of the universe, and there are indeed such ambitious critiques: we call them "philosophical critiques"). -- And it would of course be ludicrous to suggest that such essays should come with ratings. That is why, for example, George Orwell's most famous book "reviews" (some of which are themselves almost of book-length) do not come with ratings: because they aren't book reviews.

Now in videogames, to get back to our subject, the only decent equivalents to Orwell's book critiques that I am aware of are my essays On Role-playing Games and Arcade Culture, and a few others in the same vein which I am currently working on ("On Real-time vs. Turn-based Strategy", "Dungeon Crawling", and others). These essays do not confine themselves to the evaluation of a single game, but take in entire genres or design philosophies within the world of games -- always within it -- and critique their underlying fundamentals and evolution. They still, however, do not equal in scope Orwell's most extensive book critiques, and with good reason. To do that they would have to relate the significance of the games to the world outside of them, to place the game -- no longer in the context of its genre or that of videogames as a whole, but in that of culture and human civilization. But this is a decisive jump, a jump which Orwell and other literary critics were not obliged to make, since the novels they critiqued were always already placed within that context. A novel, you see, a work of narrative fiction, always refers back to the culture which produced it, and this is especially true of the significant novels (also called "philosophical novels") which serious critics are mostly concerned with. But videogames are nothing like that. The "story" or the "visuals" or the "music" or the "atmosphere", which the pseudo-intellectual gamers become so laughably enraptured with, are never ultimately valid objects of critique, since they can be easily changed without significantly altering the essence of the game. These are secondary, incidental aspects, which the pseudo-intellectuals in their ignorance and stupidity elevate to aspects of primary importance. When these secondary aspects are disregarded what then remains is a system of rules which erect and constitute a reality -- a reality which short-circuits the existing one and substitutes itself for it. How then, to relate the new reality to the old one? The old one has been abolished -- there is nothing to say for it -- except perhaps "Good-bye, you won't be missed!" The very act of game reviewing, of critically examining a videogame so as to place it within a hierarchy of videogames, presupposes that one has no interest in reality. Critiquing a specific game, therefore, or a genre or series or philosophy of game design, in the sense that we defined the term "critique" above, would be childish -- it would be a mistake that only a child would make, a child simply playing with words whose meaning he can't even begin to comprehend. For the only way to marry videogames and the concept of "critique" is to undertake a critique of videogames -- and that, as things currently stand, no one other than me is capable of doing.