Should book reviewers (like yours truly) assign grades to literature? The immediate impetus for this question is that I am contemplating creating a grading system for the books I review; yet this would seem to contradict the initial purpose I put forward for the creation of this website: to appreciate–that is, to enhance the value of—old books.

Yet obviously I don’t hold the reductio ad absurdum view that old books are to be valued merely because they are old; you won’t see me leading the McGuffey Reader revival; I may praise Jeremy Taylor’s spectacular rhetoric but you won’t find me preaching his old time religion. (Put it this way: if I swore off every old author who would have disapproved of my lifestyle, I wouldn’t be The Old Book Appreciator now, would I?) But appreciating anything means marking it out for special consideration; I can’t well run around naming all the animals and plants like naked Adam. There’s “too much and [I’m] not in the mood.” Every critic makes a selection merely by virtue of what they choose to write about; my failure to write about Nora Roberts is, whether I would cop to it or not, an unspoken message. (In this case, the message is that I have no firsthand knowledge of her oeuvre, having been warned off by passing disdainful mentions of it in the writing of individuals I read whom reputation and past experience have imbued with cultural capital, to borrow Bourdieu’s terminology.) I can however decide on a more sensible, more methodical way of singling out what I praise, bringing a slightly more empirical mien to my reviews.

While publishers must necessarily define one group of works as especially meritorious (the whole idea of a canon creates a psychological pressure to read, to stay current, so to speak), as a critic I am under no obligation to pretend that out of print authors’ books are inherently less important than the latest now striving to be sold. As Hazlitt pointed out in his essay “On Reading Old Books,” time is often the best judge of what is especially great; and yet I submit that the judgment of time has sometimes been faulty for a range of reasons, one of which is that the very act of selecting “classics” creates an unbalanced view of the quality of old literature. Fact: just as today there are many fine books that don’t win Pulitzers, so too are there many books of centuries ago that are neglected because they don’t meet the arbitrary threshold of “classic.”

In most walks of life in which grades are assigned, people cavil with them vigorously. It’s no mystery why this is: grades are evaluative. They are an (arrogant) assertion by the grader about what is important, what is better, to what should go the accolades, to whom should go the spoils, which of these australopithici gets to mate with the females. It’s a ruthless business, this grade-giving; small wonder that the world’s most approval-seeking nincompoop won’t shut up about his IQ scores. Small wonder that virtually every author bio today reads like a job resume. Who are you and what tribe of apes stamped your passport? And the result is a system of social sorting—the so-called meritocracy—which purports to be objective but is in fact deeply rooted in the caprices and prejudices of society (and not even society generally; moreso society’s most wealthy element, the 1% of people who oppose net neutrality and the estate tax, two positions it requires an immense assumption of false consciousness to imagine average people unblinkeredly supporting.)

But as teachers would argue (especially those of a progressive stamp, which in America these days is nearly all new teachers), grading doesn’t have to be just a tool of domination and social sorting. There is what is called formative (as opposed to summative) assessment, which basically means giving a trial evaluation in order to enable the student to reconstruct their knowledge and abilities before you affix their future wealth or poverty with the assessment summative. (It is notably rare for these same progressive-minded teachers or schools to forego the other, social sorting side of testing; no, the meritocracy must start grinding out worthies even  before they’ve reached the age of majority; now tell me, how many extra-curriculars do you have?)

A maxim of what might be called the science of assessment, however, is that a test can only teach you about what it has been designed to test. If you want to test the ability to memorize word definitions or historical dates, give a matching or fill-in-the-blank test. If you want to test your students’ ability to speedily handwrite in their blue books a formulaic parroting of some dates and themes they encountered in the textbook or in five weeks of the professor’s stimulating PowerPoints, give them a timed test with essay and short answer questions. (We’re getting warmer!) If you want students to reproduce in abstraction poetry the gauzy intellectual posturing of the sample academic journal articles you gave them to half-read, puzzle over, and apply the same theoretical content of—except this time to a different anachronous text—have them write essays. But if you want them to learn to write the way I do—and let me unhumbly submit that this may not be a terrible thing; or if you please, substitute my name for that of any of the great ones whose greatness a writer may aspire to—then your students’ modes of inquiry—and your modes of assessment—will need to correspondingly expand.

To grade literature—or to do it well (more on that in a moment)—is not to unpedestal these great ones and treat them with the indignities suffered by a freshman. (That’s the social sorter’s way of thinking.) It is true that any evaluation of literature must be at least partially subjective: we are deciding what to value. But my desire to create a grading system for books stems from a half-playful wish to represent my artistic values in number form; it can be informative, though not dispositive. I am not proving out my opinions, but I am clearly stating them by affixing them to numbers.

But to grade literature, I feel, a simple letter grade (or a scale of 5 dongles out of 5) simply will not do. It feels crass, like I blathered for a page and here’s a number. The starred review is little better; it says there are great books and there are not-great books. Good luck. What we need is a rubric (said Every Teacher Ever) consisting of a variety of attributes a literary work can have, and then an overall score based on their average. We will need to change the weight of different attributes in the overall score, since various qualities can predominate in importance when judging particular works of literature. Finally, the grades need to be attended by definitions for each attribute, benchmark literary texts which signal what a given score for that attribute could mean, and a description of that attribute in the given work being reviewed which justifies the score given.

This system would allow a more nuanced evaluation of literature by more accurately reflecting the variable merits and demerits of texts; we will no longer look at books in a false binary of classics and non-classics but instead be able to sift through diverse typologies of texts: your stylistic ones, your plot-dense ones, your books outstanding for dialogue, for characterization. What we are building is a beautifully detailed, almost anthropological view of literature, one which sees different novels partaking of different strengths.

Well, now I need to go build it.