Sunday, June 17, 2007

On prediction and testability

It has been brought to my attention by smokey in the “5 point summary” thread, that my predictions regarding ID (in the thread "Science of ID") are nothing but restated hypothesis.

While it is true that the first stated prediction does contain the hypothesis restated, it also contains more specifics as to which information system (life) and which possibly intelligent system (information processor which causes the program of the universe) of which it is a necessary programmed result. However, whether those are genuine predictions or not, I HAVE included an idea of an experiment in the last two points, which consists of two computer simulations, the prediction being that the hypothesis is further verified and the aforementioned cycle is further verified as a law of nature if the first computer simulation works and the second does not. However, if at any time, the second simulation accidentally produces an information processor from a set of random laws, then Intelligent Design theory suffers a critical blow and is falsified.

This is similar to the experiment that had provided further confirmation of the theory of general relativity. If relativity is true, mass will bend “space/time” thus causing electromagnetic radiation to follow the bent path. Therefore, relativity is further verified if the light from the star is bent by the sun (as in the famous experiment), but suffers a critical blow if it isn’t. Likewise, if Intelligent Design theory is true, programming by intelligence is a necessary cause of information processors thus information processors will not randomly generate themselves from a random set of laws. Therefore, ID theory is further verified if a program can be fine tuned by intelligence to develop an information processor but suffers a critical blow if any random set of laws which make up a program will accidentally generate an information processor.

Furthermore, now that I’m on the topic of scientific prediction, I’d like to discuss it a bit. What is necessary in order for a theory to have prediction potential and thus be scientific? In order for prediction to be possible, the theory must follow a law and create a pattern.

For example, the evolution of life is briefly defined as a continual generation, storage, and use of new information (CSI). The pattern is a continual increase in information, in a pattern identical to learning where noise is filtered and useful information is added to previously stored useful information. Thus, a prediction (and restatement of the theory) that new complex specified information will be observed to be produced by life is compatible with and a verification of the theory of evolution.

Likewise, ID theory proposes that the aforementioned cycle (pattern) that I discussed in the thread “Science of Intelligent Design” is a law of nature and can therefore be applied to any observed phenomenon which is at any part of the cycle. The prediction that follows is that, since the cycle can not be broken (although may be started by yet another phenomenon), ...

1. whenever information (CSI) is seen, upon further investigation, intelligence will be discovered to be the cause, and
2. that intelligence will necessarily produce information, and
3. information/processing systems will never arise independent of the programming of an intelligent system.