Science, as it is known in the mainstream, is very seriously flawed and few people talk about it. The flaw is requirement to use faith in accepting your hypothesis no matter how strong the evidence supporting it; the problem is that science touts itself as the disciplined practice of using measured evidence to make a statement about the observable world. In reality science can only disprove statements on a very limited and measurable set of variables.
In the maximum likelihood statistical realm there are two hypotheses: the null hypothesis and the alternate hypothesis. The null hypothesis is evaluated against emperical evidence and can therefore never be accepted. It can only be rejected or there is a failure to reject because only measurable evidence can be used to evaluate the null hypothesis.
If the null hypothesis has a failure to reject, a leap of faith is generally made to accept the alternate hypothesis. This leap of faith is based purely on negative, or disproving emperical evidence around the null hypothesis. Let's say there is a hypothesis, 100 measurable variables that influence the outcome, and 1000 unmeasurable variables that influence the outcome. The observer can only do one of two things: [1] reject the null hypothesis based on any one of the 100 variables or fail to reject based on the 100 variables. They are entirely unable to accept the null hypothesis because there are an additional 1000 unmeasurable variables. The observer typically will then accept the alternate hypothesis on a statement of faith that they believe (completely absent of emperical evidence) the other 1000 variables are not influencing the outcome.
The problem is dramatically compounded by the fact that those 1000 extra variables are actually the multiplicative product of every conceivable combination of treatments within each variable; these are also known as interactions and there could literally be billions, quadtrillions, etc. of them for more complex systems.
I rarely meet scientists who understand this. They don't realize that everytime they perform an experiment and accept any hypothesis they do so on faith. No different than the religious people they often belittle.
sallystrothers: Science, as it is known in the mainstream, is very seriously flawed and few people talk about it. The flaw is requirement to use faith in accepting your hypothesis no matter how strong the evidence supporting it; the problem is that science touts itself as the disciplined practice of using measured evidence to make a statement about the observable world. In reality science can only disprove statements on a very limited and measurable set of variables. In the maximum likelihood statistical realm there are two hypotheses: the null hypothesis and the alternate hypothesis. The null hypothesis is evaluated against emperical evidence and can therefore never be accepted. It can only be rejected or there is a failure to reject because only measurable evidence can be used to evaluate the null hypothesis. If the null hypothesis has a failure to reject, a leap of faith is generally made to accept the alternate hypothesis. This leap of faith is based purely on negative, or disproving emperical evidence around the null hypothesis. Let's say there is a hypothesis, 100 measurable variables that influence the outcome, and 1000 unmeasurable variables that influence the outcome. The observer can only do one of two things: [1] reject the null hypothesis based on any one of the 100 variables or fail to reject based on the 100 variables. They are entirely unable to accept the null hypothesis because there are an additional 1000 unmeasurable variables. The observer typically will then accept the alternate hypothesis on a statement of faith that they believe (completely absent of emperical evidence) the other 1000 variables are not influencing the outcome. The problem is dramatically compounded by the fact that those 1000 extra variables are actually the multiplicative product of every conceivable combination of treatments within each variable; these are also known as interactions and there could literally be billions, quadtrillions, etc. of them for more complex systems. I rarely meet scientists who understand this. They don't realize that everytime they perform an experiment and accept any hypothesis they do so on faith. No different than the religious people they often belittle.
As for science, its weakness is that it is inductive, not deductive. It's a very fine tuned prediction machine, but it is still a prediction machine. Predictions are made by people who are uncertain of the future, and if there is no certainty, there is no knowledge. I define knowledge as certainty.
So many people say they know this or that, and how often are meteorologists wrong, for example? All the time! Then people say, "Well, it's not an exact science." No science is an exact science in that it offers certainty. Science is not knowledge. It is inductive assumption, no matter how finely tuned. The weakness of it lies in empiricism, which has no knowledge of the future. It tries to predict the future, and while it often does so accurately, it is always and forever uncertain.
So, in short, I agree with you, though I'm not certain if due to the same reasons. What do you think?
I don't think science is necessarily inductive or deductive but can be both. Adademic science tends to be inductive and industrial science tends to be deductive. I strongly agree with you that all science can produce is an assumption, which is what I was trying to convey. The assumptions can be accurate but they are always blind faith assumptions, so the question "does God exist" is very much in the realm of science just the same as "does man cause global warming" or "how many neutrons are in a carbon molecule." All of those questions can be answered by science but in all cases the observer makes a leap of faith due to the difficulty of rejecting null hypotheses.
As for predictive models go I think their accuracy is completely independent of time, but rather on the dependent variables. When all of the controlled variables are fixed and the response outcome behaves as a random variables it says the observer is missing factors in their prediction model. The price of a market index is the outcome of thousands, millions, or billions of main effects and exponentially more interactions. If we use 4 factors to predict a market index the randomness appears large because we only represent a fraction of a fraction of a fraction of a percent of the independent variables.
A lot going on here, and there is really nothing I can add other than saying you have a point. Though I don't feel too much like arguing, getting into any depth, or disecting what you're saying because it's too much work right now.
I think this is why we have to seriously look more into Teutonic styled thinkers such as: Wittgenstein,Stirner, Heidegger, Austrian Economists, Nietzsche, Weber, Husserl, Freud, etc.
We tend to be at severe odds with the more "Anglo-American" tradtion - and they are right to a degree, only subserviant to anything that we can really say. In many cases while the "start and finish" of their theories may be correct, they tend to usually be fine on the bulk which is the "middle" - as their methodology works as a decent enough method to get the "gist"for the mechanics of things.
"As in a kaleidoscope, the constellation of forces operating in the system as a whole is ever changing." - Ludwig Lachmann
"When A Man Dies A World Goes Out of Existence" - GLS Shackle
all science can produce is an assumption,
I don't think you're using the word "assumption" in its ordinary sense. The obvious question is: assumption compared to what? No human knowledge is absolute, that much we know. In that sense, everything is an assumption. But not all knowledge is created equal and some knowledge is definitely of a higher quality than other.
In any case, the fact that human knowledge is not absolute is irrelevant because we are still compelled to act. Mises points out that knowledge, too, is a category of action. To put it in plain language, the need to know is driven, ultimately, by the need to eat.
they are always blind faith assumptions,
I don't think you are using the phrase "blind faith assumption" in the ordinary sense. If you ask "what does the inside of a black hole look like?" any answer to this question is, necessarily, a blind faith assumption. But if you ask "is this car red or blue?" the question has a definite answer that cannot at all be described as "blind faith assumption."
so the question "does God exist" is very much in the realm of science
It is connected to any other question of human knowledge in that it is a question that is a subject of human inquiry. I don't go in for the whole "God is not science" positivist crap that a lot of the Reason/Dawkins/Hawking types go in for. In other words, reject epistemological dualism.
just the same as "does man cause global warming" or "how many neutrons are in a carbon molecule." All of those questions can be answered by science but in all cases the observer makes a leap of faith due to the difficulty of rejecting null hypotheses.
I don't think that your OP description of "science" is really accurate. I had that class in college, too, and it explains one method of scientific experimentation. But the fact is that science - being just one chapter in the book of human knowledge - is actually a "bag of tricks" - to be more precise, it is a bag of domain-specific tricks. By far the most important trick of all is the Gedankenexperiment and all of modern science of the "null hypothesis" variety lives and breathes within the walls built by Gedankenexperiments.
Clayton -
I've read enough research papers to know that nobody follows the supposed rules. It's more of a "gotcha" in the establishment arsenal if they disagree with you.