One of the most difficult challenges for a screenwriter of thrill-based entertainment (horror films, thrillers, murder mysteries, etc.) is to hold the audience's attention in the set-up stage, especially if the premise is complicated. Doing just that is one of the many things that Dana Stevens did right in this, her first screenplay. Although the gimmicky premise required the transmission of a lot of information, the first 25 minutes went well, transmitted the many necessary facts in a painless manner, and got the audience involved in the characters.

The overall premise could have made this just another gimmicky film in which the killer, or an important witness, or an intended victim, has a unusual medical malady. (Multiple personality disorder or amnesia, anyone?) In this case, the set-up is kind of a really complicated version of Wait Until Dark. The key murder witness is blind. Well, actually, she was blind until six weeks ago. Since her transplant, she is still kinda sorta still blind, but she can also kinda sorta see. Sometimes she sees things clearly, but more often not. Most important to the plot, she seems to have some kind of a delay in her nervous system. Stuff happens before her eyes on Tuesday, but she doesn't actually see it until Wednesday. To make matters even worse, some of her visions are flashbacks to events which happened decades earlier (before she lost her sight), so she is trapped in a world of dimly perceived sights which may or may not be clear, and may or may not be happening as she sees them.

For example, she went to a mirror and saw herself in the mirror as a child - the last thing she saw before she lost her sight twenty years earlier. Then she saw her friend in the hospital, although the friend had actually been there the day before. Given those facts, how could she know whether the criminal she saw was someone she really saw then and there, or someone from a day before, or even twenty years before?

To say the least, the police found her testimony to be lacking in credibility. In fact, they didn't even believe her at all when she first tried to report the crime - until a murder victim was found and the investigators stumbled into the formerly blind woman living in the apartment below the victim, thus supporting completely what she had reported earlier. The thrills of the thriller are generated by the woman's repeated sightings or visions of the murderer she may or may not have seen the night of the murder. Does she keep seeing him? Is he following her? Or did she ever see him at all? Since her vision is still developing, she can never really be sure what she is seeing, and she perceives most things as if in a fog or a distortion mirror.

The film was intended to function as a murder mystery as well as a thriller. The murder mystery portion of the development just doesn't work at all. When the police finally figured the case out, the relationship between the victims was something that we could not have solved from our theater seats, because the solution hinged on something hidden from us - the fact that one of the victims was killed by mistake. Furthermore, the criminal was never developed as a character, and his motivation, while eventually sufficing as a satisfactory explanation for his actions, got tossed in from deep in left field. On the other hand, if we forget about the cerebral part of the puzzle and react to the visceral portion, Blink is actually a pretty effective thriller, because director Michael Apted managed to keep the audience in the POV of the semi-blind woman, thus experiencing her paranoia and confusion.

The best part of the film, the element that lifts it above the dozens of similar films that go straight to video every year, is the character development. It is done so well that the gimmicky premise is soon fully internalized and accepted as a given. The four good characters - two cops, the blind woman, and the blind woman's doctor - are all developed, and are all real people. Like all of us, they said things they regretted, they said things that were politically incorrect, they hurt each other, they goofed off when they should have been working, and they made mistakes. The lead detective and the blind woman did fall in love, as required by movie convention, but not until after a lot of hesitation and false starts. Even after they fell in love, the screenwriter was daring enough to suggest that their mutual obsession got in the way of the bone-crunching detail work necessary to real police investigations. In time, the detective dumped the woman on a uniformed cop so he could get back to the station and get some work done. He then ducked her calls, not because he was mad at her, but simply because he had a job to do, and jobs don't stop so people can fall in love. That was pretty damned effective, because (1) it was true-to-life; (2) it made things more emotionally satisfying for us when they overcame the problems and worked things out sensibly.

The director brought something very interesting into the film - the sights, and sounds and geometry of Chicago. I used the word "geometry" with a great deal of consideration, because Michael Apted plays his own visual games with the changing geometric shapes of the city of Chicago, just as Jean-Pierre Jeunet did with Paris in Amelie. The camera angles are set up to catch the triangular symmetry of the entry stars to the El, or the unique curve of the train's approach, the diamond of the Wrigley infield, or the repeating rectangles of the skyscrapers whose details are lost in the morning fog. This is a very impressive subtlety that really invokes the feel of the city on a very deep level for those who have been there. It makes the film smell like Chicago.

I suppose the script is probably a bit too ambitious - romantic triangle, thriller, mystery, psychodrama, character study - but the overall impact is positive. It works. While not without flaws, this is a good little thriller with some complex and deep character development, and I got into it.