Things influence other things. That鈥檚 a basic statement of any dynamic world where things change, and things would be very dull if it weren鈥檛 the case 鈥 not that we鈥檇 exist to know about it, without a cause. Causality is the study of how things influence one other, how causes lead to effects. In the classical world we live in, it comes with a few basic assumptions.
The first big rule of classical causality is that things have causes. They don鈥檛 just happen of their own accord. If a ball moves, the likelihood is someone kicked it; if an apple falls from a tree, it鈥檚 because its weight became too great for the branch it was hanging from.
Second, effects follow causes in a predictable, linear manner. You swing your leg, make contact with the ball, and off it moves, in that order and no other.
Third, big effects grow up from little causes. A piston, for example, starts to move when a lot of individual hot atoms hit against it and push it a certain way. The laws of thermodynamics, which govern the way atoms move, then provide certain rules about what causes can precipitate what effects, and so an overall direction for causality 鈥 a flow of time.
Modern science presents a number of challenges to this naive view of causality. Einstein鈥檚 special and general theories of relativity, for example, abandon the idea of a single line of time along which causes and effects can be strung. Time is warped by the presence of large masses or when travelling at high speed, and passes at different rates for different observers.
Even so, relativity bends over backwards to maintain a sensible conception of cause and effect as we know it. The key is the speed of light, which relativity insists is a constant. It represents the maximum speed that any influence can travel in the universe. Under this assumption, the only things that can influence an object’s state lie in its 鈥渂ackwards light cone鈥 鈥 the area of space-time from which light has had time to travel towards it.
Similarly, the only things the object can influence lie in its 鈥渇orwards light cone鈥, within the area of space-time where its light can reach. As an example, if the sun were suddenly to explode, the effect of frying Earth to a cinder would not happen instantaneously. Earth would have just over eight minutes of peaceful life still left 鈥 the time it takes light to travel from the sun 鈥 before it got intimation of the oncoming calamity.
Work through the maths of relativity, and you can show that the overall effect of this stipulation of no influences faster than light is to make it impossible for an event to come before another in one observer鈥檚 frame of reference, and after it in another observer鈥檚 frame of reference. That鈥檚 why suggestions of things travelling faster than the speed of light cause such a furore: they鈥檇 destroy any sensible concept we have of what causes what, opening the door to the future influencing the past. An example was when researchers at the OPERA experiment thought they鈥檇 spotted faster-than-light neutrinos in 2011 鈥 an effect shown in the end to be caused by a loose fibre-optic cable and faulty clock used in timing the particles.
Unfortunately, matters of cause and effect get distinctly murkier in the other realm of modern physics, the quantum world. Unlike relativity, which governs the universe on large scales 鈥 the scales of stars, planets and galaxies 鈥 quantum theory determines the workings of very small things such as individual particles.
For a start, the concept of uncertainty is hard-baked into the quantum world. By setting certain limits on how accurately we can measure certain quantities at a quantum level, it also mucks with a conventional view of cause and effect. For example, one consequence is that pairs of particles can pop up at random from an empty vacuum, as long as they disappear again quickly enough not to violate the quantum uncertainty principle. As long as they exist, these 鈥渧irtual鈥 particles can have effects on the real world 鈥 an apparent case of an effect without a cause.
The quantum world is plagued by what Einstein called 鈥spooky action at a distance鈥 鈥 influences travelling seemingly instantaneously between particles, where measuring the state of one seems to influence the state of the other far faster than the speed of light. Experiments have even shown that two events can indeed seem to happen both before and after one another, opening the way on the quantum scale to what鈥檚 called 鈥渞etrocausality鈥 鈥 the future influencing the present which can influence the past.
What that means no one quite knows. At the very least it seems our conceptions of causality need to be modified in some way in the quantum world. Or perhaps causality is preserved, and it鈥檚 our conceptions of space and time that don鈥檛 quite work on that scale.
Some scientists have even argued that 鈥渂ottom up鈥 causality, where big effects at large scales grow up from lots of little effects at small scales, needs to be abandoned in favour of some kind of 鈥渢op-down鈥 causality, where cause and effect are imposed from some larger scale. And it鈥檚 even possible that the classical laws of thermodynamics, with their strict rules of what can cause what, do not have quite the same validity at the quantum scale.
That鈥檚 all very speculative, though. At the moment, the best we can say is that at the scales we work at: assume cause鈥 then effect. Whether the world truly works that way is another question. Richard Webb
