Ready or not.

In the 1970s and 80s, Benjamin Libet, an American brain scientist, conducted a series of famous experiments. Instructed to carry out small, simple motor activities, such as pressing a button or flexing a finger, participants were placed in front of a clock with electrodes affixed to their scalps. During the experiment, the participants were asked to note the position of the arm of a clock when he or she was first aware of the urge to act. The experiments revealed an increase of electrical brain activity preceding the conscious decision to move by several hundred milliseconds. This subliminal brain activity was dubbed «readiness potential». While Libet himself has wavered somewhat in his interpretations of these findings, others unhesitatingly think this discovery has huge ramifications for our self-understanding. If volitional acts are initiated before we become aware of them, then we must be deluded when thinking our conscious «decisions» have any causal effect on what we do!

It may seem impossible to conclude otherwise. If electrical goings-on in our brans make our decisions, then in a sense we don't. But here one must not forget what a conclusion is in this context. The conclusion that we are not free is not a scientific discovery, but rather an interpretation of (or implications drawn from) certain scientific discoveries. Interpretation is a way of framing data, involving, at least tacitly, arguments based on some underlying assumptions. That human beings are mere neurological robots, therefore, is not an empirical fact, but rather the implications drawn from one possible interpretation of these facts. Am I then suggesting that this interpretation must be wrong? No. I only suggest that it is not obviously correct either. In other words, I am asking us to consider the possibility of reading Libet's data differently. How to read scientific discoveries, of course, cannot itself be a scientific question--at least not merely. This calls for philosophising.

Let us start by investigating why this gloomy conclusion may at first seem unavoidable.

Instinctively one might be alarmed to learn of the readiness potential. How unsettling that our brains decide for us! But as Wittgenstein once remarked, we often are struck by the wrong aspects of things: the important features escape our attention because of their familiarity (PI §129). Perhaps, then, the conclusion that subliminal brain activity diminishes the role of consciousness in human action seems so compelling only because we overlook something which is always before our eyes?

For example: There must be electrical goings-on in the brain all the time. Acknowledging this obvious fact should make some of the initial surprise ebb away. Electrical brain activity prior to conscious decisions is exactly what one should expect to find! After all, no brain activity is synonymous with brain death. Ok. But Libet did not merely record electrical humming in the brain—that, clearly, would be unstartling. What his experiments revealed was a significant increase of electrons firing some milliseconds prior to a conscious urge to act. Surely this warrants paying special attention to it, or am I denying that something imortant happens here? Well, not exactly. But if we take a closer look at one of Libet's graphs, will we not see the line going up and down all the time? And if that is the case, why single out this particular peak as particularly significant—rather than, say, regarding it as just another elevated stage in a normal pattern of fluctuations? Doesn't this peak seem rather randomly chosen? (Subsequent experiments have, in line with this logic, identified readiness potentials several seconds prior to any conscious urge to act.)

My intention with that rhetorical question is not to deny that a case can be made for «the compelling» conclusion, merely to emphasize that the case must indeed be made. One cannot simply take Libet’s assumptions for granted. Once his reading of the graph is seen for what it is, namely a reading, and the assumptions built into that reading are made explicit, then, my point is, we are positioned to see that other readings must be possible too.

But let us, for argument’s sake, grant Libet his reading of the graph's ultimate peak. Let us assume, then, that «readiness potential» was an established fact. Would this render conscious decisions causally ineffective in volitional human action?

Before stampeding towards that conclusion, one thing we should notice is how anemic are the concepts of action and decision with which Libet operates. A severely limited understanding of these human capacities is buit into his research design, making it unclear how much of human action actually is illuminated by Libet's research. Purposeless finger flexing and random button pushing are (at best) special instances of human action--if indeed «bodily movements» is not a better term for them than «human action». However we categorize it, the behaviour of Libet's subjects is quite unlike much of what we otherwise think of as human action. Furthermore, Libet’s experiments were purposely designed so as to make the timing of the movement irrelevant. The participants were to have no reason for preferring sooner over later. Some human actions may very well be like that—at least they were in Libet's artificial settings—but more often than not do we care about what to do and when to do it. To make a long story short: It will take a considerable amount of philosophical work on Libet’s part to make this convincing as a paradigmatic picture of human behaviour. (Which is needed for the implicit general rejection of conscious decisions to be plausible.)

And what does it mean to say that an action is «initiated»? Being clear on that point is of course crucial when reflecting on the implications of Libet's findings. The readiness potential is conceived as a brain activity initiating human action at a particular moment in time. But does it make sense (and if so, what sens does it make) to think of all volitional actions as subliminally initiated some milliseconds before they occur?

Sometimes we say that it took us time to make up our minds on what to do, suggesting a picture of a long inrun before a sudden take-off. A typical case would be the child who hesitates for a long time before suddenly ripping off a band-aid. Now, it seems plausible that certain happenings in the motor center of the brain—i.e. a «readiness potential»—can explain the sudden motion. But could it not also be argued that these electrical goings-on only mark the final stage of the initiation process? That the initiation of the initiation, as it were, began when the child first formed the intention to rip the band-aid off? Consider also planned actions. How much light does the readiness potential shed in such cases? Take, for example, someone who finally asks for his girlfriend’s hand in marriage. A lot of preparation has led up to this moment: For months he has deliberated on how best to approach it; he has considered different options for time and place—should he propose at midsummer and out at sea, or rather wait until her birthday when her favorite flowers blossom?; he has rehearsed the question (the exact formulation of which he has actually written down on a piece of paper in his pocket); he has booked a hotel suite; bought a ring, and today he has picked up a lovely bouquet of roses. If someone were to explain the time and manner in which he proposed in terms of electrical signals firing in his brain, then there are reasons—not empirical reasons perhaps, but certainly philosophical ones—for being sceptical about their approach. Of course this man would never have performed as he did but for electrical firings in his brain—but, this surely is a very thin explanation, and miles away from what we under all but very limited circumstances would consider an answer to the question of what prompted him to propose when and how he did.