Blood Ties (2014).

If you've seen enough movies, sooner or later you walk out of one with the feeling that the film you just saw is so deeply flawed that you're not quite sure why you don't actually hate it. I've spent a couple days tossing around the film Blood Ties, an American film directed and co-written by French filmmaker Guillaume Canet (Tell No One), in my head as a result of such a feeling.

Blood Ties is the story of two brothers, one a NYPD police officer (Billy Crudup) and the other a hardened criminal just released from prison (Clive Owen). It is set in the 1970s. The movie savvy among you can probably fill in most of the plot from there. James Caan plays their father - he favors the criminal over the cop for reasons related to their no-good mother (who makes no appearance). Mila Kunis plays the criminal's girl (and is completely wasted in the role), and Zoe Saldana plays the cop's girl. Marion Cotillard completes the key ensemble playing the criminal's ex-wife, drug addict, and prostitute that the criminal used to pimp.

In many ways the movie is a trainwreck. Canet and his co-writer, favorite-of-the-French James Gray, fail to locate the center of the story they're trying to tell and instead try to pack two or even three movies into a single two hour homage to Canet's favorite New York crime films of the 1970s. The film offers nothing new - story-wise or visually. The performances are just so-so - although that may owe more to the shoddy writing than any deficiencies of the actors.

And yet, with all that, I was still interested. For all of its flaws, I didn't hate the film or even really dislike it. If anything, I lamented what a lost opportunity it was. If only Canet had been able to focus and to avoid indulging his desire to over-pack the film, it might have been a wonderful counterpoint to the recent American Hustle.

Speaking of American Hustle, let's confront the originality issue head-on for a moment. Much of the criticism Blood Ties has faced has been complaining about the film's lack of originality. This criticism is difficult to stomach from critics that lavished praise on American Hustle just a few months ago. American Hustle didn't have an original bone in its body. That's not to say I didn't like David O. Russell's film - I would describe it as a play we've seen many times before, well performed and, with the exception of the overbearing reliance on slow motion walking shots and mostly uninspired musical choices, fairly well staged. But let's be honest - the thing was little more than Russell's 70s crime film and homage to Scorsese - apparently every major director needs one.

What kind of filmmaker would you be if you didn't demonstrate your ability to shoot someone walking in slow-motion wearing period 70s clothing set to, oh, I don't know, how about ELO or Steely Dan? Reading through the press notes for Blood Ties, Canet hardly denies the desire to do a 70s crime film because he thought it would be fun to pay tribute to some of his favorite films.

Canet perhaps chooses the wrong favorites - he cites, inter alia, Lumet and Schatzberg (I didn't see him mention French crime-film master Melville). Doesn't he know that Scorsese is the Alpha and the Omega? I can't help but wonder if some of the "it isn't original" criticism is rooted in Canet's failure to pay homage to The One (as many modern critics seem to see it).

As noted, however, originality (or lack thereof) isn't the real problem here. You can copy the old filmmakers all you want and it might just be fun to watch regardless of the lack of new ground. What you can't do is botch the story if you want your little crime drama to hold up.

Here Canet botches the story. While Crudup is the closest thing to a center the film has, Canet cannot focus his attention there. Nor can Canet focus on the key protagonist/antagonist relationship between Crudup and Owen. Canet just doesn't seem that interested in this brotherly tragedy. So we are never really allowed to dig deeply into it to discover the contours beyond "they hate each other and love each other" and filling in the blanks from the expectation that you already know this story anyway.

With his attention continually wandering, Canet ultimately tries to present six relationships - between the brothers, between the brothers and their women, and between the brothers and their father. None goes beyond skin-deep, the center of the film is continually lost, and ultimately none of their relationships are particularly compelling. It's just too much. By trying to treat them all too equally, all of them feel like they're given short shrift, and not one of them is particularly memorable. I think of the film and I wish someone could have told Canet to pick a character and cut everything that didn't tell his story (and to then beef that story up from there). It would have been interesting to see what he did.

And therein lies the rub. Because I watched this film and saw so much that seemed like it could have been a scene or a shot from a much better movie. It can actually be pretty entertaining at times. There's an interesting commitment to portraying New York with faith to the period - the film was shot there and Canet was obviously careful with how he framed shots and set up background action to try and sell it. Despite a misstep here and there, Canet demonstrates skill - he just didn't give himself the material to make full use of it. Here, the the whole doesn't add up to the sum of the parts.

The cinematography and original music are solid. I have to give Canet credit for his period musical choices. Many of the songs are actually from the 1960s (Tommy James, The Velvet Underground, Sam Cooke). That seems appropriate given that these characters are not kids - it seems likely that their music would have been from a little earlier era. His choices from the 1970s included Quincy Jones and Little Richard (Money Is), Janis Ian (At Seventeen), Ace Frehley (New York Groove), and The Isley Brothers (Take Inventory). Alas, musical selection provides the most embarrassingly on-the-nose moment for Canet - when he sets Cotillard's trek to buy heroin to, you guessed it, Heroin.

Overall, it would be easy to dismiss this film outright. It's too interesting, even if deeply, deeply flawed, to do that. I suspect that if Canet disciplines himself and develops his craft, this might end up being viewed down the road as a gem in the rough. But if he doesn't pull it together, it will give a pretty good idea of how he went off the tracks.