Rose begins with a description of an ordinary day in the life of his laboratory. Here he tackles head-on the thorny issues raised by experimentation on animals — his research involves killing baby chicks. He then presents an account of his own academic and intellectual history, tied in with the disciplinary histories of biochemistry, neurochemistry, and neurobiology.
One chapter is devoted to the history of ideas of memory, from the oral traditions that preserved epics through to modern computers. Here Rose loses his footing, and his portrait of artificial intelligence research is a distorted strawman. He rightly complains about attempts to replace neurobiology with computer science, but doesn't realise that doing the opposite is just as silly; computer memory may be different to biological memory, but it is just as "real". It is particularly annoying that both Searle's debunked Chinese Room thought experiment and Penrose's dubious indeterminacy/uncomputability of consciousness thesis are presented without any criticism at all, and Rose's classification of the symbolists as holists and the connectionists as reductionists is just confused.
The next chapter is an introduction to the phenomenology of memory, which looks at different kinds of memory, people with unusual mnemonic abilities (such as eidetic memories), and diseases and brain injuries which affect memory. We also get a brief introduction to the anatomy of the brain and the scanning techniques used to investigate it. Rose then moves on to consider memory in animals, and in particular the experimental tests used to verify the occurrence of associative and conditioned learning and phenomena such as sensitization and habituation. This is tied in with the history of psychology. A whole chapter is devoted to describing the dead end of attempts to find "molecules of memory" and to reduce neuroscience to molecular biology. (Rose is vehemently anti-reductionist, as is apparent throughout the book.)
After all this background is covered, he gets down to describing the latest work in neuroscience. One chapter describes recent work on sea-slugs and the hippocampus, then he finally returns to his chicks, with two chapters describing his recent research in some depth. This contains enough technical detail to be of interest to biologists, but is still presented clearly enough for the ordinary reader to follow. The penultimate chapter returns to the sociology of science, with a good description of what actually happens at scientific conferences and of how scientific papers are "fictions". The final chapter attempts to synthesise everything.
The description above is much neater than the reality. The Making of Memory actually deals with topics and issues as they arise, and as a result is a rather hodge-podge kind of work. Everything that is included is there to help the reader understand what is going on, however, and the result is an extremely readable and very accessible book, whatever its formal failings. In particular it does an outstanding job of portraying what is actually involved in doing scientific research; this is a book that only a sociologically informed practising scientist could have written.
Readers should be warned that Rose is a Marxist and that politics do feature, but they should not let this put them off. The most important thing is that the science itself is interesting and well presented, and I have only one reservation about recommending The Making of Memory — everything Rose writes about artificial intelligence should be taken with a grain of salt.
April 1994
- External links:
-
- buy from Amazon.com or Amazon.co.uk
- Related reviews:
-
- books by Steven Rose
- books about neuroscience
- books about philosophy of science
- more popular science