Customer Reviews

Searching for Certainty: What Scientists Can Know About the Future

5 star:		(0)
4 star:		(2)
3 star:		(1)
2 star:		(0)
1 star:		(0)

 

 

It is a study of some diverse fields attempting to give a measure of how well we understand a field and how well we can predict the future. I am not sure why various fields were selected but overall they give a good range of concepts - weather, biology, stock market, outbreak of war and mathematics.

The writer then at the end of the discussion gives his own grading for how well we can predict an event and how well we can explain it. These figures are purely subjective and I often disagreed with them.

There are often two problems in trying to predict something using science. The first is how good is your scientific knowledge of something occurs. The second is how good is your data. For example imagine a rocket being feed wrong data. Although our knowledge of mechanics is excellent, it will probably crash.

Three of them are actually fields that I have spent a considerable period of time and study on and can comment on them in detail.

The weather I found very interesting. It is a case where we have very good data and our theories look good. The major problem seems to be our ability to process the data.

He then goes into predicting climate. Here we have good data but lousy theories. We really know very little about this field.

Stock market - I found his study starting to being very good on the theoretical points of the merits of different theories. Unfortunately just as he started to get interesting he stopped. Possibly as his study is to short or he lacks the experience in this field.

Here again we have a field where we have good data but lousy theories.

However many studies of various theories of the market are examined. Some that the writer states show that the market is not perfect. But he seems a bit dubious about this at times. Quite rightly I feel. For example he shows over time that smaller companies are more profitable. This maybe due to because they are riskier they are undervalued by the market. This argument is true of even large companies with low P/E that the writer misses altogether. There is often a very good reason why the market gives them a low P/E. But the logical point now is that if this is so what does it say about the market. If the market is overcompensating for risk then it cannot be perfect.

The other main problem that in my experience from many years in the stock market there are three main variables on a share one how well the company does, how well the market does and how well the shareholder does out of it. They can and often are very different.

If the market is near perfect as some theories suggest because the public know enough of the facts. As people do not know or use the information equally well some people must be using the information better then others. Some are showing over the years repeatedly better performance then the average. Buffet is a huge one and I am a tiny one. Now why? This is not explained. Buffet does not buy heaps of shares but a few selected ones.

The next discussion was on trying to predict war. This is even harder to predict. Here we have little data. For example how much knowledge did the US have on the North Vietnamese government thinking processes at the start of the Vietnam war or how much did the Egyptians of the Israeli government in 1967?

Here the writer claims that after the event our theories can offer scientifically defensible schemes for explaining how any given war came about. This is nonsense. Historians are still debating the causes of just about every war I know of! Often there reasoning seems like rationalisations for what happened.

He does go though a few theories that just shows that we really know almost nothing about predicting war. Almost all historians would probably admit that even knowing what they know now, very few war could have predicted say a month before it happened. One historian I read stated that Germany attack on Russia was remarkable as it's about the only war that you could predict. Interesting though I would argue that Stalin did not.

Above all he claims to make war you need cash. This is true but it predicts little as in today's world a country can often by a third party be given the cash.

This is the same problem with the master variables discussed - population, technology and resources. Where he states that you need technology to make war. This maybe true but predicts nothing, as a country may be able to get it without having it. For example Egypt under Nasser had the population. Russia supplied the technology and never got paid as Egypt had little resources.

The last section is about mathematics and I am a bit confused as to why he put it in. There he goes though Godel's principal. I found this whole section a rather confused mess. Mathematics is more a study of human thought then science as such.
Perhaps the writer was trying to say that we could never really be sure of anything even an idealised system.

He then gives it a B for predictability that is quite arbitrary.

Overall I found the book interesting as one-person attitude to this concept of looking at certainty.

This is a fun and thought provoking book. I only give it four stars, because I had questions, and they did not all get answered. Not fair, but neither are the grades that Casti gives science.

My original questions were as follows:

1. What is chaos theory? Casti provides a clear and concise answer, although this is not the main theme of the book.

2. Does chaos theory have anything to say about the philosophy of determinism? This is not addressed in the book.

3. Does chaos theory provide any guidance on when to stop spending time and money analyzing a chaotic system? This is not addressed.

Some questions that were provoked by the book follow:

1. Are climatologists still using the same modeling techniques to predict climate change as used to predict weather? Casti made the point that predicting climate and predicting weather are two different endeavors, yet they were both using the same modeling techniques, as of the time that this book was written. I don't think he explained why.

2. What is Godel's Theorem of Incompleteness? Casti provides a good (as far as I know), non-rigorous explanation in chapter 6.

3. Does the fact that biological forms (or any other things) exist, imply that science can eventually describe a means of creating them? This controversial question is probably only implied by the full context of the book. Casti seems to want his audience to keep an open mind, because he cleverly avoids the creation versus evolution debate. Instead, in chapter 3, he focuses on what science could tell us (in 1990) about how cells differentiate into tissues in an embryo, and then form organs of appropriate size, shape, and arrangement. As in all chapters, he ends by jokingly giving science a letter grade on its ability to enlighten us.

Much later in chapter 6, he includes a humerous description of a chocolate cake machine (CCM) that he would like to invent, so that he could have any conceivable kind of chocolate cake. He then worries that Godel's Theorem of Incompleteness implies that there may not be a recipe for every conceivable type of chocolate cake. The implications with respect to my question about creation are not directly addressed, but perhaps the conclusion is obvious.

If you find the title of the book intriguing, you'll probably like the book.

Overall I agree with the other reviewer on this page. The opening chapters of Searching -- those treating science and uncertainty from a theoretical perspective, and those devoted to 'case studies' of the weather and the stock market -- were concisely and strongly argued. Together they formed a solid introduction to scientific reasoning and method. The chapter on the prediction and explanation of war was the least compelling. Casti is at his best when discussing the hard sciences. Like the other reviewer, I am also at a loss to understand Casti's discussion of mathematics (Goedel's theorem, Turing machines, etc.) in a book on scientific uncertainty. That said, it was a useful overview of the topic (albeit a more or less verbatim excerpt from Casti's slim volume on Goedel [Goedel: A Life of Logic]).

On balance, Searching is a very thoughtful and thought-provoking book. I particularly appreciate the attention Casti lavished on his annotated bibliography.