4 ms·
You can say: if X and Y happen, then with a great degree of certainty Z will follow. That's explaining things, despite not being able to predict if Z will occu
by lottin 9y ago
You can say: if X and Y happen, then with a great degree of certainty Z will follow. That's explaining things, despite not being able to predict if Z will occur or not. There are many ways to test the validity of a model, for example, you can "forecast" past values (using known values of the explanatory variables) and check the accuracy of the predictions.
- arethuza 9y ago"if X and Y happen, then with a great degree of certainty Z will follow" How is that different to normal scientific predictions?
- lottin 9y agoIt's not different. What I'm saying is your ability to predict Z is contingent on your ability predict X and Y. If X and Y happen to follow regular patterns, then you can predict Z, if not you're out of luck. The point is that claiming that something is not science on the grounds that it can't predict Z makes no sense.
- Pamar 9y agoThere is a little problem here, though. In other scientific disciplines the "laws" are not time dependent. I.e. Chemical reactions in 1894 work just the same as in 2017 or 3251. On the other hand discovering that your country is sitting on a big oil field in 678 b.C. will not help your state budget at all, while in 1940 it could be enough for your neighbours to finance a war of invasion. So being able to "explain" past economical phenomena is not very useful if you then cannot use the (now validated) model to guide present and future decisions.
- lottin 9y agoIt's true, although it's still useful in many instances if we take a more general approach. So, let's say that instead of focusing on the effects of discovering oil on the national budget, we could focus on studying the economics of natural resources in a more general sense. Here historical evidence could be perfectly useful.