Originally Posted by Antarctica!

"Synthesis certainly plays into it," Risu confirmed, nodding, "but a scientific theory is more than just a synthesis of knowledge. It doesn't just answer the question, 'Why?' That's of course necessary for the theory to work, but it's not all of it. Every good scientific theory has a second part to it, a very important one. Testable predictions."

He raised the fifth finger of his right hand. "Testable predictions. To stay within our example, we have concluded, through our experiments, that ice melts faster in warmer climates. The first part of our theory says something like, 'Ice melts faster in warmer climates because...'," Risu paused, then shrugged. "Something. It has to do with temperature and the strength of chemical bonds, and an exact explanation would take us too far afield. But our theory fulfils the first important function, which is to explain the experiments we've run."

He got up with a heavy sigh and stood before the class. "Now, the second part - does our theory make testable predictions? It doesn't yet, but we can amend it to do so. Let's assume..." he smiled again, "that for the maths of our theory to work out, ice melts twice as fast for each degree the temperature rises. That's what our theory predicts. On the basis of that prediction, we can now go back and run a new series of experiments investigating that specific claim. If the experiments show that ice does indeed melt twice as fast when you raise the temperature by one degree, then our theory has been confirmed and we can be proud of ourselves. If the experiment shows that, for example, ice melts three times as fast, there is something wrong with our theory and we need to tweak it."

He paused again. "Questions?"