This is an interesting question, and I think there have been a lot of good answers to it already. I'd just like to clarify one distinction - space itself does not necessarily have a temperature, instead it's the systems of particles that fill space that have temperatures. As many others have mentioned, for particles the temperature is effectively the average kinetic energy per particle. If you have one or two particles and you know their kinetic energies you could say that they have temperatures given by those kinetic energies, but that is a bit misleading. Temperature is a thermodynamic quantity, and as such it only really makes sense when you look at a large ensemble of things (it is used to describe, statistically, a distribution of particles and what happens to them). Space may be nearly a vacuum, but there are particles there. There may only be about an atom per cubic centimeter in the typical interstellar medium, but if you consider the vast numbers of cubic centimeters that fill up the space between stars you can appreciate that there are a vast number of atoms out there. If you were to plot a histogram of the kinetic energies of these atoms you'd find that they have a multi-modal distribution. Effectively you can break it down into different groups of particles with different temperatures (ranging from a very cold tens of Kelvins in "dense" million atom/cubic centimeter molecular clouds to an incredibly hot millions of Kelvins in the hot ionized medium with only 1 atom per ~1000 cubic centimeters). One thing to keep in mind is that in a given volume of space you may have more than one group of particles each with a different temperatures (you can have, for example, a very tenous hot ionized medium filling the same space as a higher density, cooler warm neutral medium) - so assigning a temperature to some place in space is a bit of a misnomer. Another thing to note is that light itself (photons) can sometimes be described as having a temperature, whic <div class="Discussion_UserSignature"> </div>