Boris is proposing a trial to allow cyclists to turn left at red lights (after giving way of course).
There is one very good justification for this. Being amongst traffic at a junction as the traffic lights change is very dangerous. It's much safer if you can get away from the junction before the traffic. That's why we have Advance Stop Lines (not that motorists observe them for the most part). Now you could argue that allowing a left-turn on red will encourage cyclists to take more risks. That's possibly true, but really risky cyclists will be jumping the lights anyway. Putting cyclists more in control of the risks we face has got to be a good thing, both from a philosophical and a practical perspective.
There is one question in my mind, however. What does a left turn actually mean, in reality? Let's look at a couple of scenarios:
1. 4-way cross roads. Simple enough you would think; if there is no traffic crossing and the light is red against you, you may turn left. But there's no reason why you can't turn left and immediately make a U turn, and another left turn, is there? If not, you've just gone straight on, without breaking the law. How shallow does the 'U' have to be before you've broken the law? I imagine you would have to actually leave the road by crossing the white line on the left-hand side, but of course not all junctions have such lines.
2. T junction, aproaching from the bottom of the 'T'. Simple enough, you can make a left turn against the red light, but not a right turn.
3. T junction, approaching from the left of the 'T'. In other words, you are on a major road, and a road joins from the right. There's no 'left' turn as such, so presumably going straight on is not allowed? But topologically, it is no different from the previous case. Consider if the T is not completely straight, and the major road has a slight deviation to the left - does that count as a left-turn?
Anyone got any ideas?