A
AndyT_81
Guest
Hi NowAgnostic,
Thanks for the interesting comments.
My second point would be that I don’t really agree that science has a monopoly on the discovery contingent truth, even in the sense you are using the word. For instance, I know that I love my wife, and I do this without anything formal enough to be called observation or inference, it’s just something that I know. Yet this is a contingent truth. It is a contingent truth that I went to Church this morning, yet I don’t know this contingent truth by any formal obervations or inferences i.e. my mind doesn’t consciously break the question into any formal argument such as:
However, parsimony will not be invoked when a less complex model does not have the same predictive power as a more complex model. In such a case the less complex model will be abandoned. So I’m not sure what you mean with regards to “correct one”, as when parsimony is invoked to choose between models, all of these models will, with regard to utility, be “correct”, i.e. they all have the same or similar predictive power. However as I have already stated, to move from utility to truth (the way things actually are) something else is needed (not just more empiricism).
Let me try to discuss what I mean by this a bit further, I guess I am trying to get at two points. First, generally scientific theories are not just sets of mathematical theorems or rules, but rather, they are mathematical theorems based around objects or entities, such as atoms. What does the “work” in a scientific theory is the mathematical structures, it is the math that produces the predictions that are confirmed or denied via experimentation. These entities serve a purpose in that they are relatively easy to conceptualise. However, can these conceptualisations be considered to be ontologically real? Take the example of quarks. Here are sub-atomic particles that can never be “seen” directly by themselves in experiments, but only can be inferred via the behaviour of hadrons which they compose. Are quarks ontologically real entities, or are they simply an easy way of conceptualising something and building the mathematics that are required to explain sub-atomic particles? This is not a question science can really answer and shouldn’t really care about, science should only care about what works and the idea of quarks seems to.
The second idea I want to present is that scientific theorems come with no guarantee that they are unique, and this is especially true of the entities invoked in such theories. The idea of quarks seems to be successful and scientists are happy to use that conceptualisation, but I think it is naive to say that, given time, some other form of conceptualisation of entities could give rise to the same succesful mathematics (strings anyone?).
cont…
Thanks for the interesting comments.
I’m not sure I completely agree here, with your way of defining science or with you notion that science is the only way of discovering contingent truth. If we consider science to be mere observation and inference, then we would need to class trivial observation-inference pairs, such as infering that a car is travelling faster than you on a highway when it goes past, as science. This type of mental activity seems to me to be a standard part of human cognition, and I would hesitate to call it science. I think a more formal process of observation and inference is needed for science to be rightly called such.Science is the only way of discovering contingent truth. I am using “science” loosely here to denote a method constituted by observation and inference.
My second point would be that I don’t really agree that science has a monopoly on the discovery contingent truth, even in the sense you are using the word. For instance, I know that I love my wife, and I do this without anything formal enough to be called observation or inference, it’s just something that I know. Yet this is a contingent truth. It is a contingent truth that I went to Church this morning, yet I don’t know this contingent truth by any formal obervations or inferences i.e. my mind doesn’t consciously break the question into any formal argument such as:
- I remember going to Church this morning
- My memory is generally reliable, and I have no reason to doubt it
- Therefore I went to Church this morning
This seems right to me, but I’m not sure what this shows. The fact that science has been successful does say something about the world, but it seems to me that what is says is that (among other things) matter is mostly predictable, and it’s behaviour can be described my mathematical formulisms. Note I say it’s “behaviour”, not its essential ontology.The contingent truth that science has been successful is only a result of the contingent truth that the world is such that it will be successful.
No, I never suggested it was a willy-nilly selection. What do you mean by “correct one”? My understanding of the principle of parsimony is that if a number of competing theories/models have the same predictive power, the least complex model will be chosen. This makes sense, because, as I have alluded to, science is nominally utilitarian and only cares about what “works”. Why use a complicated spanner when a simpler one does the job?Again, I’ll agree in part, but science also seeks the most parsimonious model consistent with the evidence, since from induction the most parsimonious model is the most likely to be the correct one. So it’s not a case of willy-nilly selecting anything that works.
However, parsimony will not be invoked when a less complex model does not have the same predictive power as a more complex model. In such a case the less complex model will be abandoned. So I’m not sure what you mean with regards to “correct one”, as when parsimony is invoked to choose between models, all of these models will, with regard to utility, be “correct”, i.e. they all have the same or similar predictive power. However as I have already stated, to move from utility to truth (the way things actually are) something else is needed (not just more empiricism).
When we are testing a model via experimentation, the more experimental results that align with said model, the more likely the model is to be accurate with respect to predicting the behaviour of matter in a wide range of situations. However, this is not to say that the model is more likely to be true, where by truth I mean the way things actually are.If the science is done properly, the model is very likely to be true, although never 100% because you never arrive at 100% through induction. What you’re trying to say here maybe is that the model is not all of the truth?
Let me try to discuss what I mean by this a bit further, I guess I am trying to get at two points. First, generally scientific theories are not just sets of mathematical theorems or rules, but rather, they are mathematical theorems based around objects or entities, such as atoms. What does the “work” in a scientific theory is the mathematical structures, it is the math that produces the predictions that are confirmed or denied via experimentation. These entities serve a purpose in that they are relatively easy to conceptualise. However, can these conceptualisations be considered to be ontologically real? Take the example of quarks. Here are sub-atomic particles that can never be “seen” directly by themselves in experiments, but only can be inferred via the behaviour of hadrons which they compose. Are quarks ontologically real entities, or are they simply an easy way of conceptualising something and building the mathematics that are required to explain sub-atomic particles? This is not a question science can really answer and shouldn’t really care about, science should only care about what works and the idea of quarks seems to.
The second idea I want to present is that scientific theorems come with no guarantee that they are unique, and this is especially true of the entities invoked in such theories. The idea of quarks seems to be successful and scientists are happy to use that conceptualisation, but I think it is naive to say that, given time, some other form of conceptualisation of entities could give rise to the same succesful mathematics (strings anyone?).
cont…