What is Freewill?

In our discussion of the ‘Not P’ machine we observed that if the machine was not told the prediction then that prediction would always be correct: but if told, always wrong. So how does this affect our understanding of freewill?

The predictor sees the machine as being completely deterministic, and so believes that they can always accurately predict the status of the ‘P’ component. But this is subject to two important caveats:

1. The predictor must have full knowledge of every aspect of the machine and its environment that could possibly affect the status of ‘P’, and
2. If the prediction is to be seen as being correct, it must be concealed from the machine until it is too late for it to respond. Otherwise, it will (quite predictably) be perceived as being wrong.

But, from the machine’s standpoint, it has control over the status of ‘P’ and can demonstrate this by refuting any attempt to predict its status. Why is this? Is it because it can do whatever it pleases? No, it is because:

1. When faced with this situation, its internal system of priorities ‘chooses’ to defy the predicted outcome, and
2. It is constructed in such a way that it has the ability to carry that choice through in spite of the wishes of anyone or anything else.

Thus we could reasonably describe the machine as having a limited form of ‘autonomy’ in these matters; although we would probably hesitate to call this ‘freewill’. So what do we mean by freewill?

Generally, when we speak of ‘freewill’ we mean our ability as humans to reach our own conclusions, make choices and carry them through without hindrance from other people or external circumstances. But if we put it that way it rapidly becomes obvious that we cannot be thinking of human freewill as something absolute and inviolable.

1. We don’t know enough to reach conclusions with absolute certainty,
2. We do not have sufficient time or mental capacity to evaluate the consequences of all our possible choices, and
3. None of us has sufficient control of our environment to prevent any outside hindrances.

It is not my wish to ‘get theological’ at this point: but when we consider the above definition of freewill, then the only being who could be described as having absolute freewill would be one having unlimited knowledge, infinite wisdom and total control – in other words, God.

For us, as humans, freewill has to be seen as a relative term; greater than the very limited autonomy of the ‘Not P’ machine, yet by no means absolute. For this reason, I am mostly going to refer to our human notion of freewill as ‘relative autonomy.’