
The Spirit of the Beehive, written and directed by Víctor Erice: very slow, and very much enjoyed.
How long does a film’s awfulness have to persist before offending the audience past the point of contempt and to the moment of abandon? For Roger Ebert, according to his latest online journal entry [note: Ebert's since added a second journal entry -- see the addendum after this entry's conclusion], the limit is eight minutes, or 92% of the total duration. (Ebert’s critque of Tru Loved, the eight-minuter, says all about his reasons for skipping the remainder of the film.) Or, if Ebert is reviewing Caligua, the threshold is crossed after two hours, or roughly 57% of the total film.
Clearly Ebert’s tolerance for low quality has clearly taken a severe plunge in recent years — well, when n = 2, at least.
Coincidentally, I was recently asked a similar question after modestly mentioning that my Netflix queue has grown over 340 movies. The question was: When I’m watching a movie at home, if and when do I decide that a film’s just not worth my time? When the film’s been sealed back in the dandy red Netflix envelope, was it after I watched the whole thing? Where’s the breaking point?
Heck, if Roger Ebert can’t settle on a definite demarcation, neither can I. But I don’t think either of us would bother.
When to exit the theater, either figuratively or literally, isn’t a static measure that crosses between all films — each show is judged accordingly, implicitly and unofficially, even on the basis that each film is, indeed, an entirely different experience. The medium is a constant, but bailing out on the Friday Night Film is a judgment of content, not the format.
Reading the comments and reactions to Ebert’s journal post reveals a fairly consistent opinion: one, that Ebert was right to walk out when he did, and two, walking out of films is rare but an acceptable reaction1, and three, no rule or sudden instance exists that can extrapolate the quality of an entire film from a segment of the whole show.
I like the idea of drawing a distinct line in the sand, but in my limited experience, establishing such a border is basically impossible, or at least very improbable. My favorite suggestion in the comments for an arbitrary time limit was 11 minutes, the time needed to run through a standard 35mm 24fps film reel. That standard, especially now that cellulose film is on the way out, has a kind of romance to it. But like a lot of romance, that’s basically folly: 11 minutes in a Hitchcock isn’t the same as the first 11 minutes of a Fassbinder, or a Kurosawa, Spielberg, Chan-Wook, or Un chien andelou.
Or forget entirely about just 11 minutes: the entire first act might be a snoozer, uninteresting, contrived, disgusting, revolting, or limp, but unforeseen qualities could blossom in the second or third acts. Or the actors might perform throughout like high-school stage drama bombs, but the narrative, or the set design, or the editing, framing and composition, or many other captivating aspects of moving pictures could tilt the production’s quality towards good and worthwhile, even to excellence. Or the denoument could justify a boggling climax that occurred just minutes before the end. And so on.
Of course, the film could start badly, continue badly, and end badly. Or a movie could rocket out of the gate and perform a rousing start, but having squandered the cutting room’s treasure all in one sequence torture the audience with inanity and boredom for its remaining duration.
Such extremes are unusual, but I’ve definitely seen films that start well and end poorly, and vice versa. But even if a film fails in some major respect, most of my picks from the last few years have had enough or plenty of redeeming qualities, usually far more good than bad and easily enough to justify the time spent. Once in a while I find a film that is so good — or bad — that it alters the personal bar for all past and future films, and the only way to find those gems (or bombs) is to watch a lot of movies.
Looking back in my Netflix queue, only a few rejects are sprinkled throughout a hundred or so rentals (and counting!). I pushed eject on Peter Jackson’s King Kong after an hour. Fellini’s famous Amarcord was too aloof for my tastes, and it received the fast-forward treatment after 45 minutes. And then there’s Six-String Samurai, which I began fast-forwarding after an excruciating ten minutes, just to see if the rest of the movie was as bad (it was).
Director and writer M. Night Shyamalan’s filmography is an odd, varied bunch when it comes to lasting power. The Sixth Sense was a great film; Signs was good, but not great; and The Village was terrible. That’s a rough gradient from excellence to awfulness, and unfortunately for him and his audience, the trend has apparently continued downward in recent years. Having skipped through most of The Village, my last-viewed Shyamalan, the plan now before considering another one of his films is consult Wikipedia on the plot and see if the whole shebang floats. I did exactly that when The Happening came out last summer and saved nine bucks for a ticket.
How about films that began poorly and slowly but ratcheted up as they progressed? Das Boot immediately comes to mind: the first act, which ran about an hour and a half, felt like a slow (but compelling) three hours; the second act of another hour and a half felt like ten minutes. I distinctly remembering watching Das Boot during that second act, turning to look at the clock, and wondering, Geez, where the heck did that hour go?
Another example: Two weeks ago I watched my first Yasujiro Ozu film, Late Spring; what I thought to be a very slow fifteen minutes of family drama was in actuality more than an hour, confirmed only after I took a second opinion from the clock on the wall. Garrisoned by the film’s appeal, I completed Late Spring and very much enjoyed it.
Yet another group, one that is expanding all the time, is the classification of films that offer some familiarity, whether due to genre, director, editor or some other aspect. If the film is a Shyamalan, extreme incredulity would be enforced; if the film is a Kurosawa, I would devote the entire evening to it; and if the movie turned out to be a Six-String Samurai sequel, I’d probably burn it. But even Kurosawa films have low points, and Six-String Samurai did star a guitar-playing samurai (or would that be a ronin? Doesn’t have the same ring), which even I’d admit is pretty swank.
When it comes to finding the odds and evens, the intrepid film-lover just has to exercise a smidge of caution, a touch of discretion, watch a ton of movies, and keep the remote (or aisle) within arm’s reach.
The ethereal border that bridges the lands between “play” and “skip” isn’t a consistent standard by any means, but it obviously exists for Ebert, the commenters at his journal, and myself strictly on a per-film basis. Besides, the movie walk-out should be exercised as a rare and extreme measure: film-goers should want to like these movies and, given the movie’s goodwill, accordingly give the production the greatest benefit of the doubt. Nonetheless, the line of abandon exists, but the individual never notices he’s crossed the threshold until he’s over it — and then the departure time has arrived.
But if a thug punched a gun against the side of my head and demanded that the line be clearly drawn, I’d have to go with the 11-minute mark. At the very least, 11 minutes is nearly 37.5% longer than Ebert watched Tru Loved.

Just one scene from some of the more deliberate and best 97 minutes of my movie-watching career.
Addendum: While editing the above entry, Ebert has responded to his original Tru Loved review and consequent blog entry by extending the original review to cover the remainder of the movie and a second entry that’s a post-mortem and apology of sorts.