Is science possible at all if the future does not behave like the past? That is the question before us, and it would appear to be obvious to many that science does indeed require nature to behave in a uniform manner. However, my opponent approached me with a desire to be Con on this issue, so it is my duty to initiate the debate with a defense of the resolution.
Science - The observation, identification, description, experimental investigation, and theoretical explanation of phenomena. (http://www.thefreedictionary.com...)
Require - To have as a requisite; need. (http://www.thefreedictionary.com...)
Uniformity of Nature - A doctrine or principle of the invariability or regularity of nature; specifically one that holds identical antecedent states or causes to be uniformly followed by identical effects—called also principle of the uniformity of nature. (http://www.merriam-webster.com...)
Induction - The process of deriving general principles from particular facts or instances. (http://www.thefreedictionary.com...)
Nature - The material world and its phenomena. (http://www.thefreedictionary.com...)
Here is my argument in syllogism:
1. Induction requires the future to behave like the past. (Necessary Precondition)
2. Science is inductive. (Definition of Science)
3. Therefore, Science requires the future to behave like the past. (From 1 & 2)
4. Science studies nature. (Observable Fact)
5. Therefore, Science requires that nature's future behaves like its past. (From 3 & 4)
6. Therefore, Science requires the uniformity of nature (Definition of Uniformity of Nature)
1. Induction requires the future to behave like the past.
The principle of logical induction is that we can derive universal truths by observing particular truths. For example, we touch a fire once, burning ourselves, and inductively conclude that all fire is very hot. Said another way, fire is expected to be hot in the future based on our past experience with the hotness of fire. If induction did not expect the future to behave like the past, then we would have to check if fire was hot every time we encountered it. After all, this time it might be cold, or edible, or nurturing.
The point is this: any time we derive a generality from one or more particular facts, we are in essence assuming that the future behaves like the past. If the future does not behave like the past, then there is no point in generalizing anything. There is no point to generalizing that fire is hot if tomorrow it might be cold, or edible, or nurturing. Therefore, the necessary precondition of induction is that the future will behave like the past.
2. Science is inductive.
Science is the process of observing then testing particular phenomena in controlled conditions in order to derive general truths about our reality. Science discovers how the universe operates by observing it operating in specific instances. Thus, science is inductive.
3. Therefore, Science requires the future to behave like the past.
If science is inductive and induction requires the future to behave like the past, then science also requires the future to behave like the past.
4. Science studies nature.
By nature, I mean our material existence as a whole, which also includes what we call the natural sciences, such as biology, botany, etc. Science is not in the position to study anything we cannot observe. We can only observe our cosmos, so science is limited in practice to the cosmos. Theorizing about metaphysics is certainly permitted, though such theorizing is still based on the generalities inductively discovered in our own universe.
5. Therefore, Science requires that nature's future behaves like its past.
If science studies nature and requires the future to behave like the past, then science requires nature to behave the same way. Thus, science requires nature's future to behave like its past.
6. Therefore, Science requires the uniformity of nature
The uniformity of nature is the notion that nature's future will behave like its past. Since science requires nature's future to behave like its past, science therefore requires the uniformity of nature.
The syllogism is logically sound. The resolution, "Science requires the uniformity of nature" is affirmed. I look forward to my opponent's rebuttal.
Introduction and framework
I’ll start off my opening round with two main points. The first point is how Hume’s problem of induction has created problems for any epistemology based on inductive reasoning. The second point is an explanation of how a scientific epistemology can be formed that does not depend on inductive reasoning and inductive generalization. The first point is for historical context and isn’t an argument about the resolution. The second point, however, negates the argument presented by Pro in support of the resolution.
Hume and Induction
David Hume demonstrated the huge problems with inductive reasoning when he wrote A Treatise of Human Nature . Hume pointed out that all inductive reasoning contains a missing premise which has gone unjustified. This premise is what Hume called the ‘’uniformity of nature’’. This unjustified premise, according to Hume, states that “instances of which we have had no experience, must resemble those of which we have had experience, and that the course of nature continues always uniformly the same”. Most philosophers agree that Hume decisively proved this assumption could not be proven deductively or causal reasoning, nor can it be proved inductively, as that would be circular. 
There’s a comical criticism of inductive reasoning that can show its flaws. In all of my experience, when an object in motion gets pushed, it travels faster. Therefore, I can make an inductive generalization that an object in motion goes faster when it’s pushed. Imagine a book traveling at the speed of light. According to my inductive generalization, if I push the book, it will travel faster than the speed of light. This means that my inductive generalization has proven that an object can travel faster than the speed of light. Therefore, my inductive generalization has refuted the theory of relativity, one of the most proven theories in all of science.
Thus begins the problem with scientific epistemology. Hume proved that induction was not a valid method of reasoning. Yet science is supposed to be inductive! How can the two be reconciled?
In comes Karl Popper. Karl Popper agreed with Hume and his supporters on the problem of induction. He even rejected the argument that, despite the flaws in inductive reasoning, it can still give us reliable or probable results. His goal was to form a scientific epistemology that did not depend on inductive reasoning. He outlined this new epistemology in The Logic of Scientific Discovery . He writes that ‘’The theory to be developed in the following pages stands directly opposed to all attempts to operate with the ideas of inductive logic.’’. This new scientific epistemology is known as the theory of falsificationism.
The first step is to form a hypothesis. It should be checked to see if it is internally consistent (not self-contradictory). If it is, then the hypothesis should be abandoned. If it is not, then you move onto the next step.
‘’From a new idea, put up tentatively, and not yet justified in any way--an anticipation, a hypothesis, a theoretical system, or what you will--conclusions are drawn by logical deduction. These conclusions are then compared with one another and with other relevant statements, so as to find what logical relations (such as equivalence, derivability, compatibility, or incompatibility) exist between them.’’ 
The second step is to form what Popper calls an ‘’axiomatized system’’. This is when you state the necessary assumptions of the hypothesis, and nothing more (you can’t state extra and unnecessary assumptions).
“This, I believe, is the reason why the form of a rigorous system is aimed at. It is the form of a so-called ‘axiomatized system’—the form which Hilbert, for example, was able to give to certain branches of theoretical physics. The attempt is made to collect all the assumptions which are needed, but no more, to form the apex of the system. They are usually called the ‘axioms’ (or ‘postulates’, or ‘primitive propositions’; no claim to truth is implied in the term ‘axiom’ as here used). The axioms are chosen in such a way that all the other statements belonging to the theoretical system can be derived from the axioms by purely logical or mathematical transformations.” 
This axiomatized system must meet four requirements, according to Popper.
‘’(a) The system of axioms must be free from contradiction (whether self-contradiction or mutual contradiction). This is equivalent to the demand that not every arbitrarily chosen statement is deducible from it.1 (b) The system must be independent, i.e. it must not contain any axiom deducible from the remaining axioms. (In other words, a statement is to be called an axiom only if it is not deducible within the rest of the system.) These two conditions concern the axiom system as such; as regards the relation of the axiom system to the bulk of the theory, the axioms should be (c) sufficient for the deduction of all statements belonging to the theory which is to be axiomatized, and (d) necessary, for the same purpose; which means that they should contain no superfluous assumptions.’’ 
The final step is to experiment on testable consequences of your hypothesis. If your hypothesis contradicts the evidence, then you hypothesis has been falsified. If it is confirmed by the evidence, then your hypothesis is not falsified. A hypothesis gains more support as new empirical evidence confirms it without falsifying it, and confidence in that hypothesis increases. Hypothesis that have not been unfalsified are compared by their falsifiaility. If there are more ways to falsify hypothesis A, for example, if it makes more precise predictions, then it is a better hypothesis than B, which makes less precise predictions. According to Karl Popper's falsificationism, 'Rather than progressing through theories being inductively confirmed, science actually progresses through theories being falsified via deductive reasoning'. 
Consider the hypothesis 'all rabbits are white'. If I go out and find a rabbit that is brown, my hypothesis has been falsified. However, if I see a rabbit that is white, my hypothesis has not yet been falsified. Further sitings of white rabbits, and no sitings of non-white rabbits, will increase the confidence in my hypothesis. The hypothesis is better than a hypothesis with less specific predictions, such as 'all rabbits have a color'.
Karl Popper, side-stepping Hume's problem of induction, created a philosophy of science that does not depend on inductive reasoning. This philosophy of science is employed by most scientists writing popular-level books. For example, Stephen Hawking frequently endorses Karl Popper's philosophy in his books. 
Therefore, P2 of the argument is false. Science is not inductive. Any scientific epistemology that is inductive has been refuted by Hume.
3: The Logic of Scientific Discovery, pg 6
‘’My own view is that the various difficulties of inductive logic here sketched are insurmountable. So also, I fear, are those inherent in the doctrine, so widely current today, that inductive inference, although not ‘strictly valid’, can attain some degree of ‘reliability’ or of ‘probability’.’’
5: The Logic of Scientific Discovery, pg 6
6: The Logic of Scientific Discovery, pg 9
7: The Logic of Scientific Discovery, pg 50
8: The Logic of Scientific Discovery, pg 51
9: Visual Reference Guides: Philosophy, pg 186
10: A Brief History of Time, pg 10
I appreciate the lengths to which my opponent went in explaining how science does not require induction. In so doing, he helped me realize that my first argument was needlessly specific. Like my opponent's rebuttal, my argument was focused on the means, but the real argument only cares about the ends.
The goal of science is to generate predictive statements about nature, and both induction and falsification lead to that end. Induction starts with observation to derive universal truths which are inherently predictive. "All fire is hot," predicts that the next fire we encounter will be hot. Falsification on the other hand starts with the predictive statement "All fire is hot," then tests the hypothesis for failure with each encounter of fire.
As gross an oversimplification as my example may be, it is no less true that falsificationism relies upon prediction, and I'm sure my opponent would agree. In his own words, "If there are more ways to falsify hypothesis A, for example, if it makes more precise predictions, then it is a better hypothesis than B, which makes less precise predictions." Indeed, a hypothesis is by definition a prediction: a guess about the behavior of nature in the future. So, although falsification seems to sidestep Hume's problem of induction, it does not escape the implications of prediction, which leads me to my revised argument.
1. Predictive statements about nature require the uniformity of nature.
2. Science makes predictive statements about nature.
3. Therefore, science requires the uniformity of nature.
1. Predictive statements about nature require the uniformity of nature.
In order to predict nature, nature must be predictable. A nature that is not uniform is not foreseeable in any possible sense. It does not matter if the predictive statement about nature turns out to be true, for it is the predictive quality of the statement itself that assumes the uniformity of nature, not its validity.
Using Con's own example, consider the hypothesis "all rabbits are white." This is a predictive statement since it predicts that the next rabbit we encounter will be white. It also assumes that rabbits will still be rabbits tomorrow instead of turnips and that whiteness is a property that will persist uniformly into the future. If whiteness and rabbit-ness will only exists today, what purpose is there to this prediction whatsoever? Moreover, if there is no value in this prediction, there is no value in any prediction.
Even an obviously false prediction like "all unicorns have horns" presumes uniformity. The notion of unicorns, though imaginary, presumes that our brains will be able to think "unicorn" tomorrow in the same way that we think "unicorn" today. The same goes for the horn. Better yet, if nature is not uniform, then even if this statement is false today, it might be true tomorrow, then false again the day after. Thus, falsification will never end, even when a hypothesis is falsified.
I submit that is necessary to conclude—due to the impossibility of the contrary—that this premise is true.
2. Science makes predictive statements about nature.
A hypothesis is a predictive statement about nature. It might be true or false, but it is predictive nonetheless. If the hypothesis is generation through induction, then its prediction of the future is based on observations of the past. If falsification is employed, then the hypothesis' prediction of the future is based on other known generalities. Either way, the hypothesis is beyond all doubt a predictive statement about nature.
3. Therefore, science requires the uniformity of nature.
The conclusion follows inescapably from the premises. If predictive statements require the uniformity of nature, then science, which relies on predictive statements, also requires the uniformity of nature.
The new argument is agnostic to the means by which theories and hypotheses are acquired, whether by induction or falsification. Methodology is not the problem, it's the end goal. Just ask yourself, if the universe was not uniform, if things could pop in and out of existence, if properties might be here today then gone tomorrow, then what purpose would science serve at all? Knowledge cannot advance if there is nothing to build upon. If Con is correct, then we must begin at square one every second. You just discovered the gravitational acceleration of the earth? Well, you better do it again because it might be different.
Con's position reduces to absurdity. Therefore, it is a very reasonable to affirm the resolution.
In my last round, I provided an overview of Hume’s problem of induction and its relationship with scientific epistemology. The consensus in academic philosophy is that Hume refuted any scientific epistemology based on inductive logic. Pro didn’t comment on this overview, so I will assume that he agrees with this.
I outlined Karl Popper’s theory of falsificationism, which is a scientific epistemology based on deductive reasoning. The steps in falsificationism go like this:
1) Form a hypothesis.
1A) Check for internal consistency.
1B) Find what logical relations exist between your conclusions.
2) Form an axiomatized system.
2A) The axioms must not contradict themselves or each other, they must not contain any axiom deducible from the remaining axioms, they must be sufficient for the deduction of all statements belonging to the theory that is being axiomatized, and must not contain any unnecessary assumptions.
3) Experiment on testable consequences of your hypothesis.
3A) If empirical evidence contradicts your predictions, the hypothesis has been falsified.
3B) If empirical evidence supports your predictions, it has not been falsified.
3C) Confidence in a hypothesis increases as it always confirmed but never falsified.
3D) A hypothesis is never proven with certainty, as a future experiment can always falsify it.
Pro agrees that science need not be inductive. This entails that P2 of his R1 argument is false. Pro has created a new argument to replace his now refuted first argument. I’ll spend the rest of my round criticizing the new argument.
A Critique of Pro’s Argument
I think it's false to say that predictions require the uniformity of nature. I will explain, using the theory of falsificationism which I have outlined, why this is so. 1: A prediction does not depend on the uniformity of nature to say that things won't be drastically different in the future. It depends on the principle of falsifiability. 2: The nature of a hypothesis is that a future experiment can falsify it. This means that it's inherent in any hypothesis that the future could be different, the opposite of the uniformity of nature.
Do predictions rely on the uniformity of nature to say that things won't be different in the future?
Consider the principle of uniformitarianism. This is the geological principle which states that geological processes happening now also happened for the Earth's entire history. If nature is not uniform, then how can this principle be true? It's possible that the half-life of carbon isotopes was different in the past! The principle is not true because the uniformity of nature is true. We have high confidence in the principle because of Popper's principle of falsifiability. Remember, Popper said that a good hypothesis has to be falsifiable. There has to be some experiment that can be done to demonstrate the hypothesis as true. The hypothesis which says 'Geological processes might have been different in the past' is unfalsifiable. I can't go two billion years into the past and see if Potassium-Argon dating still works like it does today. Geologists shouldn't reject the idea that geological processes were different in the past because nature is uniform, but because the hypothesis is fundamentally bad, as it cannot be falsified. I think the same line of reasoning applies to concerns Pro brought up like 'maybe rabbits will be tulips tomorrow' and 'maybe our brains will be different'. Scientists don't ignore these concerns because nature is uniform. Like our geologists, they ignore it because it violates the principle of falsifiability. Predictions don't rely on the uniformity of nature to make them coherent. They rely on the fact that wild concerns are unfalsifiable by their very nature.
We could always find a brown rabbit!
In fact, it's inherent with any hypothesis that a future experiment can always falsify it. Let’s take the hypothesis which states that all rabbits are white. Suppose that experimenters find that, out of one thousand rabbits observed, all of them are white. No non-white rabbits have been found yet. This gives us high confidence, but not certainty, that the hypothesis is true. One of the central tenets of falsificationism that is that a hypothesis is never proved with certainty. As Stephen Hawking notes when discussing falsificationism, ‘’No matter how many times the results of experiments agree with some theory, you can never be sure that the next time the result will not contradict the theory’’. On falsificationism, the future can always be different and falsify a hypothesis. Far from assuming that nature will be uniform into the future, falsificationism actually says the future could be different from the present. (This doesn't mean we have to pay attention to wild concerns about the future being drastically different, because as I talked about in the first paragraph, these concerns are unfalsifiable.)
Given the truth of the two statements, I think it's clear that P1 of Pro's new argument is false. Perhaps it's logically valid, but since the first premise is wrong, it's not a sound argument.
A Brief History of Time, pg 10
The resolution before us presents a difficult epistemological dilemma for my opponent, and while his argumentation regarding the nature of falsification is learned, he seems to be missing the point. The point is not that falsification operates as intended without the uniformity of nature, it's about the heavy implications of Con's thesis. If there is no uniformity of nature, there is no scientific knowledge to be gained whatsoever.
After Con's latest argument, I am able to see precisely where my opponent's mistake lies: step 2 of falsificationism. As my opponent has said repeatedly, every hypothesis must be accompanied by a system of axioms with the following constraints per my opponent:
"The axioms must not contradict themselves or each other, they must not contain any axiom deducible from the remaining axioms, they must be sufficient for the deduction of all statements belonging to the theory that is being axiomatized, and must not contain any unnecessary assumptions."
This seems reasonable enough, so let's hypothesize once more that all rabbits are white. Among other things, the axioms must include the notion of a rabbit and the notion of color. In my earlier round, I stated that if nature was uniform, then rabbit-ness and color-ness cannot be relied upon. They could exist one second and not the next. So, is it reasonable to even include the axiom of a rabbit and the axiom of color without also presupposing the axiom that nature is uniform? Absolutely not, since the hypothesis relies on the ability to examine rabbits and color without having to re-falsify those notions every second, or millisecond, or picosecond.
Consider how my opponent said statements like "rabbits might be tulips tomorrow" can be ignored because they are not falsifiable. This is simply untrue if nature is not uniform! Indeed, if the laws of nature are not consistent, then the hypothesis that "rabbits will be tulips tomorrow" is very much falsifiable. Moreover, if my opponent is going to be consistent with his thesis, he better take hypotheses like these seriously as scientific claims or he will not be doing science according to his own preferred scientific epistemology.
In truth, however, no scientist would take such claims seriously because it is a waste of time because every scientist ultimately prefers to rely on the most fundamental of all axioms: the future will behave like the past. So while falsification doesn't require the uniformity of nature from a theoretical standpoint, scientists will always presuppose it, particularly in their axioms.
In fact, any hypothesis whose axiom is another falsifiable hypothesis is making my case for me. As soon as one hypothesis axiomizes another, it is presupposing that the axiomized hypothesis will persist into the future. That is how science advances. We build upon previous knowledge. The computer I am using now is possible because engineers relied on the laws of electro physics persisting in to the future.
Now, my opponent will insist that it's all okay because if electro physics is falsified, we'll just start over with new falsifiable theories. This would be a rather trite response to win a debate and would miss the implication. If nature is not uniform, science should start over every day, even if it just so happens not to change. My opponent doesn't realize that anytime falsificationism does not start over, it is presupposing nature is uniform and there is no need to start over. Every theory must be retested moment to moment. Of course, that doesn't happen.
In closing you, the reader, are being asked to accept that science does not require the uniformity of nature even though the uniformity of nature will continue to be assumed in all scientific endeavors. My opponent wants you to accept his epistemology even though he will most assuredly not live it out in practice. He, and all scientists like him, will be a living contradiction to his own thesis, which I am confident I have reduced to absurdity.
I thank both the readers and my opponent for their time and thoughtful consideration.
I apologize in advance if anything I say is hard to understand, or I make a spelling error. I started school recently, and with that, I have the combined work of one AP class and two honors classes. This forces me to type out what I’m saying in one sitting.
Pro made an argument in R1 to affirm the resolution. One of the premises of this argument was that science is inductive. I outlined a scientific epistemology which makes this premise false. Pro agreed, and dropped the argument in place of a new one.
I responded to this new argument by attacking the first premise, which states that predictions require the uniformity of nature. I made two arguments in response to this premise. The first argument is that predictions don’t rely on the uniformity of nature to dismiss claims like ‘’rabbits might be tulips in the future’’. Rather, they rely on the fact that such concerns are unfalsifiable. The second argument states that it’s actually inherent within any hypothesis that the future could be different. It seems that Pro dropped his second argument as well. From what I gather, his new argument is that the implications of falsificationism are too absurd for it to be a sound epistemology.
Please keep all of this in mind when you vote on the debate. Pro made an argument, and then he dropped it. Pro made another argument, and then he dropped it. I think this fact alone is sufficient to award Con argument points.
I think there some flaws in his new, and third, argument that need to be pointed out. On falsificationism, a hypothesis can always be wrong in the future. The best we can have is high confidence that a hypothesis is true. There is no certainty about the truth of any hypothesis, because we could ‘’always find a brown rabbit’’. For some inexplicable reason, Pro thinks that this entails we can’t have scientific knowledge. There is a missing premise in this argument that Pro hasn’t justified. Why should we think that scientific knowledge is impossible because nature could be different in the future? Scientific knowledge is just the hypotheses (plural of hypothesis, as I learned today) that we have a high amount of confidence in. Having knowledge does not require you to have certainty that your knowledge is true. It only means that your hypothesis is always confirmed and never falsified.
Pro says that axiomatizing your hypothesis depends on the uniformity of nature. If you pay close attention to his argument, you’ll notice that this is, for all practical purposes, the same exact argument Pro gave in his last round, but with different words. I already explained why these statements don’t assume the uniformity of nature. Scientists don’t say that rabbits are white and color exists because nature is uniform, but because the hypothesis that color won’t exist tomorrow or that rabbits will be some new color is unfalsifiable. I gave a very lengthy explanation of this using uniformitarianism as an example. (Please feel free to skip the italics. I just put it there for reference.)
“Consider the principle of uniformitarianism. This is the geological principle which states that geological processes happening now also happened for the Earth's entire history. If nature is not uniform, then how can this principle be true? It's possible that the half-life of carbon isotopes was different in the past! The principle is not true because the uniformity of nature is true. We have high confidence in the principle because of Popper's principle of falsifiability. Remember, Popper said that a good hypothesis has to be falsifiable. There has to be some experiment that can be done to demonstrate the hypothesis as true. The hypothesis which says 'Geological processes might have been different in the past' is unfalsifiable. I can't go two billion years into the past and see if Potassium-Argon dating still works like it does today. Geologists shouldn't reject the idea that geological processes were different in the past because nature is uniform, but because the hypothesis is fundamentally bad, as it cannot be falsified. I think the same line of reasoning applies to concerns Pro brought up like 'maybe rabbits will be tulips tomorrow' and 'maybe our brains will be different'. Scientists don't ignore these concerns because nature is uniform. Like our geologists, they ignore it because it violates the principle of falsifiability. Predictions don't rely on the uniformity of nature to make them coherent. They rely on the fact that wild concerns are unfalsifiable by their very nature.”
To my surprise, Pro insists that hypotheses like ‘’color won’t exist tomorrow’’ are falsifiable! This claim doesn’t make any sense. Falsifiability is the property of being able to be disproved. There is no way to disprove the claim that color won’t exist tomorrow. I can’t even begin to think of an experiment that would disprove that. It’s about as falsifiable as saying ‘’I’m really a one hundred foot giant, but I appear to you as six feet tall’’.
Pro portrays falsificationism as saying that we always need to check our hypothesis, and that we can never have any confidence that they’re true: ‘’if nature is not uniform, science should start over every day, even if it just so happens not to change. My opponent doesn't realize that anytime falsificationism does not start over, it is presupposing nature is uniform and there is no need to start over. Every theory must be retested moment to moment. Of course, that doesn't happen.”
I think this is an extremely inaccurate portrayal of falsificationism. It ignores the Bayesian and probabilistic nature of falsificationism. Falsificationism allows you have to have high confidence that a hypothesis is true. The more something is confirmed and never falsified, the more confidence you have. This also increases the amount of evidence you would need to rationally discard the hypothesis. There is a relationship between past experiments which confirm a hypothesis and expectations of falsificationism in the future.
Pro says some things about how scientists and I, despite our academic insistence on falsificationism, will still keep assuming the uniformity of nature anyways. Of course, this is all just speculation and opinion, so I feel safe in ignoring it.
The winner of the debate should be the person with a preponderance of the evidence. Pro has dropped two out of three of his arguments, and the third one has serious flaws. This is why the preponderance of the evidence flows towards Con.
Thanks to Pro for taking part in this debate. I learned a lot about scientific epistemology, because accepting this debate forced me to research falsificationism. I appreciate that Pro didn’t forfeit or concede the whole debate. He presented a reasonable case, but unfortunately, it can’t stand up to criticism. I look forward to debating him again one day.