I’ve just completed “reading” (listening to) The Black Cloud, by astronomer Fred Hoyle. It has been refreshing to discover a book – science fiction at that – which I can enthusiastically recommend to almost any reader. Hoyle was a successful astronomer who turned to writing science fiction later in life. Hoyle’s science fiction emphasizes the science more so than most authors of the genre, and in particular almost all modern authors. The Black Cloud describes the proper process by which scientific discovery and inquiry advances. In particular, for those interested and knowledgable in astronomy, it is a wonderfully accurate depiction of the science as it was practiced in the 1950s. My fascination with the book also derives, in part, from being old enough to remember at least some of the technology that Hoyle describes, though I was experiencing it much later, in the 1980s, not the 50s.
The story in the book (which I won’t spoil here) does get sufficiently unusual to easily classify as science fiction, but continues to bring in fascinating philosophical sidelights. Unlike most science fiction, I’d even go so far as to say there was no obvious lack of plausibility in the story line – though I’m sure if I thought about it harder there would be plenty of mud to throw, it is fiction afterall. For the faint of heart, I’ll warn you that many people die, though the description of the calamity is not particularly detailed – I might even classify it as “callous”. But for Hoyle, the story is only a vehicle to make some very deep points about the philosophy of science, the nature of information, the nature of life, and even voice some frustrations about government.
What I want to discuss here are the thoughts he provides on the nature of science – I may very well come back to other thoughts he expresses in this wonderful book in later discussions. The point he makes (only stated openly twice in the book, but demonstrated continuously) is that observations of correlations or coincidences are not a proper basis for science, and do not represent causality. This is certainly not a new sentiment presented by Hoyle, but it is one that is essential to grasp, and which is of increasing relevance in what passes today as “science”.
The purpose of science is to determine the causes of phenomena. By a cause we mean an antecedent particular entity or event whose existence results in the existence of the phenomenon. The determination of cause is a structured activity, involving observation, formulating a hypothesis, prediction, and verification.
The correct progression in the development of a scientific theory starts with observation of the phenomenon under study. The researcher may then find correlation between the phenomenon and other entities or events. This must lead to the formation of a hypothesis which explains the connection between these antecedents and the phenomenon, and this explanation must be formulated within the context of existing knowledge. That is, the hypothesis cannot be arbitrary, whimsical, or rely upon unknown forces or entities. The hypothesis must arise from inductive reasoning from the observations made of the phenomenon and its antecedents. Any valid hypothesis must further be capable of making predictions which can be subsequently verified.
Hoyle’s commentary specifically deals with the critical step of when a hypothesis becomes a theory – when it is accepted as an explanation, and the cause of a phenomenon established. Hoyle points out that the mere observation of correlation is never sufficient to establish a connection; furthermore, that attempting to “prove” the hypothesis by deductive reasoning using the correlation and hypothesis itself as a starting point leads to completely invalid chains of logic, and cannot bring additional validity to the hypothesis. The only way to validate a hypothesis is to use it, in conjunction with other observations if necessary, to make predictions of future observations, and then to make those observations and confirm the predictions.
Let’s have a look now at the actual text from The Black Cloud. It is additionally interesting how Hoyle manages to present his ideas in very short passages. He further shortens them by using the Russian character “Alexandrov”, who speaks very tersely. After a few comments from Alexandrov that only later become interesting (and to which I’ll return), the first example of Hoyle’s attack on poorly constructed science arises after the astronomers observe a strange behavior in how the Cloud reacts to block radio transmissions. After reviewing the actual observations, the scientists begin hypothesizing a feedback mechanism, but without explaining how it would work. Here is the relevant passage:
“Let’s go into this in a bit more detail,” … “It seems to me that this hypothetical ionising agency must have pretty good judgment. Suppose we switch on a ten centimetre transmission. Then according to your idea, Chris, the agency, whatever it is, drives the ionisation up until the ten centimetre waves remain trapped inside the Earth’s atmosphere. And — here’s my point — the ionisation goes no higher than that. It’s all got to be very nicely adjusted. The agency has to know just how far to go and no further.”
“Which doesn’t make it seem very plausible,” said Weichart.
“And there are other difficulties. Why were we able to go on so long with the twenty-five centimetre communication? That lasted for quite a number of days, not for only half an hour. And why doesn’t the same thing happen — your pattern A as you call it — when we use a one centimetre wave-length?”
“Bloody bad philosophy,” grunted Alexandrov. “Waste of breath. Hypothesis judged by prediction. Only sound method.”
The next outburst by Alexandrov follows a comment about ESP experiments:
“I know this is rather a red herring, but I thought these extra-sensory people had established some rather remarkable correlations,” Parkinson persisted.
“Bloody bad science,” growled Alexandrov. “Correlations obtained after experiments done is bloody bad. Only prediction in science.”
…“What Alexis means is that only predictions really count in science … It’s no good doing a lot of experiments first and then discovering a lot of correlations afterwards, not unless the correlations can be used for making new predictions. Otherwise it’s like betting on a race after it’s been run.”
The final passage uses an example that can easily be remembered. This both reinforces the theme discussed so far, and suggests another variant of the theme. The situation leading to this passage requires some explanation. The governments of Earth had launched nuclear missiles into the cloud. These had been redirected by the Cloud to return to their points of origin, with “random perturbations”. Three major cities had been destroyed.
“It looks to me as if those perturbations of the rockets must have been deliberately engineered,” began Weichart.
“Why do you say that, Dave?” asked Marlowe.
“Well, the probability of three cities being hit by a hundred odd rockets moving at random is obviously very small. Therefore I conclude that the rockets were not perturbed at random. I think they must have been deliberately guided to give direct hits.”
“There’s something of an objection to that,” argued McNeil. “If the rockets were deliberately guided, how is it that only three of ’em found their targets?”
“Maybe only three were guided, or maybe the guiding wasn’t all that good. I wouldn’t know.”
There was a derisive laugh from Alexandrov.
“Bloody argument,” he asserted.
“What d’you mean ‘bloody argument’?”
“Invent bloody argument, like this. Golfer hits ball. Ball lands on tuft of grass — so. Probability ball landed on tuft very small, very very small. Million other tufts for ball to land on. Probability very small, very very very small. So golfer did not hit ball, ball deliberately guided on tuft. Is bloody argument. Yes? Like Weichart’s argument.”
“What Alexis means I think,” explained Kingsley, “is that we are not justified in supposing that there were any particular targets. The fallacy in the argument about the golfer lies in choosing a particular tuft of grass as a target, when obviously the golfer didn’t think of it in those terms before he made his shot.”
The Russian nodded.
“Must say what dam’ target is before shoot, not after shoot. Put shirt on before, not after event.”
“Because only prediction is important in science?”
“Dam’ right. Weichart predicted rockets guided. All right, ask Cloud. Only way decide. Cannot be decided by argument.”
I will return to discuss the new aspect which I believe Hoyle has introduced in this particular passage in my next post.