Well, because it just did, which is why trying to time the market is a fool's errand. As blogger Felix Salmon points out:
If anybody tries to tell you we're seeing "fears of a double-dip recession," or somesuch, ignore them. Fears of a double-dip recession do not appear overnight, and do not send markets down 3.5% in the course of a morning. When vague "fears" are cited as the prime reason for a sell-off, you can be sure that in fact there's no reason at all. Markets are volatile things, and sometimes this kind of thing happens. If you can't stand it, you shouldn't be invested in stocks in the first place.
One thing you can be sure of: all tomorrow's reports about how markets have reacted to the [July] employment report should be taken with an enormous pinch of salt. At this point, it's impossible to know what's priced in and what isn't, and in any case this kind of volatility would normally last a second day in any case. Whatever markets do tomorrow, they might well have done anyway even if the employment report hadn't come out.
If markets hadn't moved much today and instead this sell-off had happened tomorrow, it's certain that everybody would blame the employment report, no matter how good it was. It's one of the basic tenets of market reporting: if markets move on the day that non-farm payrolls are released, then there's always a direct causal relationship between the move and the report. So it's worth remembering, on days like this, that sometimes we don't know why markets have moved, and sometimes there simply is no reason.