Cloe asked:

What does it mean to say that time has a direction? Does time, as a matter of fact, have a direction?

I have identified the following: psychological arrow, causal arrow, law of thermodynamics.
Concentrating on the causal arrow as possibly fundamental, can you offer me any support or
constructive criticism?

That's what I like to see: You have made a start with the problem, and identified the considerations
that you think may be relevant!

If one imagines standing outside the whole of time, from the Big Bang up to the present date, all that
exists is a series of events, ordered by the relation, 'before-and-after'. These events are like points on
a road map, or tracks on a CD, or words in a book. Looking down on the history of the universe one
can retrace it's steps from the end of the journey to the beginning. It would be like starting at the last
page of the book and reading backwards, or reversing the motor on your CD player. It makes no
difference. The same events happen in the same relative order. The only difference is the events that
you arbitrarily designate as 'first' and 'last'.

One of the events in the history of the universe was the time when as a schoolboy I accidentally threw
a cricket ball through the classroom window. How does this look when played backwards? A cricket
ball lying still on the grass jumped into the air in the direction of the broken window. As it did this,
pieces of glass lying scattered about also took into the air, and, then, as the ball passed through the
window frame into the class room, the glass pieces came together to form a window pane.

One thing you could say about this story is that it seems extremely improbable. But improbable things
can sometimes happen. It is extremely improbable, but not impossible, that the molecules composing
an area of ground would suddenly 'vibrate' in the same direction at once, giving the ball sufficient
impetus to become airborne. And so on. The law of Thermodynamics is all about probabilities. 'You
can't pass heat from a cooler to a hotter body' is not entailed by the laws describing the motions of
material bodies. There is a finite, though tiny, probability that the reverse could happen on a particular
occasion. If you allow this, then you must also allow that there is a still tinier probability that the
reverse alwayshappens. If you ignore questions of probability, it is consistent with the laws of nature
that the universe is running in the opposite direction from the direction that we think(your
'psychological arrow') that it is.

This is all mind-boggling, isn't it?

But let's press on. What I have just said only makes senseif we have some independent criterion for
the direction of time. To say that the universe is reallyrunning on the opposite direction from the
direction that we think it is running implies that there is a real difference between reading the book, or
playing the CD, backwards or forwards. What could that difference be?

This is where causation comes in. In reality, you might claim, it was the cricket ball that actedon the
pane of glass, causingit to break. When the ball thumped down onto the ground, it caused the
molecules in the ground to vibrate. The reverse did not in fact happen. The molecules in the ground
did not cause the cricket ball to jump up into the air, and so on.

According to the eighteenth century philosopher David Hume's analysis of causation, in his Treatise
of Human Nature,what I have just said has questionable validity. To say that event A caused event B
is merely to assert that there is a lawlike constant conjunctionbetween events of type A and events of
type B. In other words, statements that describe causes and effects reduce to statements about
instances that can be deduced from the laws of nature. But we have just admitted that, so far as the
most general physical laws describing the motions of physical bodies are concerned, it is logically
possible to reverse the time order in which events occur.

The conclusion is that if you want to use causation as the logical criterion which ultimately determines
the direction of time's arrow then you need a richer account of the concept of a 'cause' than the
'covering law' model that Hume's analysis entails.

Over to you.

Geoffrey Klempner