Heh. It was not rushed. We *did* get an R&R. Boghossian is a philosopher. Singer is a bigtime philosopher. You (or anyone) of course can still disagree with anything in there but the argument "you should have had some philosophers involved in the writing and reviewing" is wrongheaded, because we did. Of course, you can still say "we had the wrong philosophers!" or "they did not do their jobs well!" or "they should have had MORE philosophers" or some such. But you really can't say we should have had philosophers as if we didn't because we did.
On the merits, I will do one more pass at an explanation that is, perhaps, a slight retreat. What I care about is whether some claim is true or not. And I subscribe to (philosopher!) Huemer's Trivial Theory of Truth: "“Sometimes people say things. When you say things, sometimes stuff is the way that you say it is. Other times, it isn’t. When stuff is the way that you say it is, we call your statements “true”.
That’s it. I told you it was a simple and basic concept. I shall call this ‘the Trivial Account of Truth.’”
I really do not care much whether unfalsifiable speculations can sometimes be considered science. I do care a lot about whether unfalsifiable speculations are being presented as if they are true. Systemic racism comes to mind. It *can* be defined in such a way as to evaluate whether a claim that is true is being made, but it seems to me very often it is not. It is often not defined. It is often claimed without citation regularly, as if evidence isn't even needed. It is a catch-all "explanation" for gaps, but it is very obvious that not all gaps come from discrimination, so one needs more than a drumbeat of "systemic racism!" to conclude that *any particular gap results from *any racism here and now whatsoever.* However, its advocates will rarely accept *anything* as refuting "systemic racism here and now!" including extreme evidence of anti-White, anti-Asian and pro-Black discrimination at UNC and Harvard presented in the recent SCOTUS case. Rather than recognizing that evidence of East Indian, Carribean Black, East Asian, and Cuban success (especially higher levels of college degrees and incomes than White people) disconfirms the "White Supremacy!" narrative, excuses are made. So if you reject all evidence of non-White people outperforming White people as inconsistent with a White Supremacy!/Systemic Racism! analysis, then this is a political argument, not a scientific one. The easiest way to cast this is as unfalsifiable. If one wishes to cast it in some other way to treats it as political nonsense that one should take seriously, that's fine with me.
To respond with edge cases of scientific speculation as "hey, life in the universe is science and its not falsifiable!" is fine but misses the point. If Delgado and Crenshaw and Kendi and DiAngelo want to recast everything theysay as "well, here are some speculations, we do not really know if any of this is true and no one should act like it is true" (as, say, scientists speculating on life in the universet) that's fine with me. Academia would be greatly improved by such clarification.
I don't see much to disagree with anymore here. I continue to think that some parts of your argument missed the mark, and had there been a different writing/editorial process that mark wouldn't have been missed by as much. But I shouldn't have speculated about the process that led to publication.
And I definitely agree that there are massive epistemic defects with how some critical social justice views are supported! (I said on the podcast that I'd have appreciated a bit more steel-manning, and I stand by that, but Spencer was right to push back because the dumb versions of the views you're attacking are depressingly popular.)
I'm glad you take my point that "falsifiability" might not be the best lens through which to view those defects. But to concede the point back to you a bit: confirmation bias is genuinely a bad thing, as is dogmatic thinking. Calling them out is entirely appropriate, and it's not obviously wrong to use the word "falsifiability" when you do, although other conceptual schema would be better.
Heh. I admit it, I do not get the objection. First, we did not spend the first third of the paper on falsificationism. In fact, we hardly address the issue.
The first page and a half or so is on how great science is. Then we get to our thematic point:
"Fulfilling this responsibility, however, is being hindered by a new, alarming clash between liberal epistemology and identitybased ideologies. Liberal epistemology prizes free and open inquiry, values vigorous discourse and debate, and determines the best scientific ideas by separating those that are true from those that are likely not. The statuses, identities, and demographics of scientists are irrelevant to this great sifting of valid versus invalid ideas.
In contrast, identity based ideologies seek to replace these core liberal principles, essential for scientific and technological advances, with principles derived from postmodernism and Critical Social Justice (CSJ), which assert that modern science is “racist,” “patriarchal,” and “colonial,” and a tool of oppression rather than a tool to promote human flourishing and global common good."
There is this single sentence on falsificationism:
"CSJ is not an empirical theory, because its tenets are maintained despite their being either
demonstrably false or unfalsifiable." I stand by that. Astrology is not a scientific theory because it is demonstrably false. God created the universe is not a scientific theory because it is not falsifiable and, in at least the Biblical variant, has led to falsified predictions (e.g., the Earth is the 6000 years old). There is life somewhere in an infinite universe is absolutely an edge case. As you say, it can be verified by not falsified. However, what makes it an edge case (in my view) is that It does lead to falsifiable predictions (e.g., about life on Mars or exoplanets). Find some claim that is not merely unfalsifiable, but which has not been falsified and which is incapable of leading to any falsifiable predictions, and I'd consider changing my view.
Good discourse on auxiliary assumptions, though. Totally agree there. Auxiliary assumptions permit scientists to write apologia defending their failed predictions and never having to surrender their cherished ideas.
But let's put all that aside. I plead innocent to the charge of "philosophical duffer" but, hey, perhaps The Court of Serious Philosophy convicts me anyway.
Its a single sentence of the paper. Lets review how we continue:
"The existence of objective reality, for example, which CSJ denies, is attested to by every successful engineering project, from bridges to satellites, from cell phones to electric cars, ever conducted. The fallibility of “lived experience” is attested to by a wealth of psychological research demonstrating errors and biases in self reports."
Here, we argue that certain CSJ claims are not unfalsifiable, but false. Every successful engineering project works because the engineers understood objective reality, thereby *falsifying* claims that it does not exist. "Lived experience" -- which is often put on some sort of bizarre pedestal by many CSJ perspectives and advocates -- is not completely useless, but so fallible that it does not deserve a (dare I say it?) "privileged" place is social science. Those claiming or implying otherwise are promoting falsified claims when they argue that "lived experience" should be privileged over other sources of social science knowledge.
We continue:
"Below we discuss publications making unsupported claims of systemic injustices and attacking merit. Such publications rarely, if ever, provide evidence that observed disproportionalities in the race or gender distribution of a scientific field are the result of present day structural or systemic racism."
Here, we are making an implicit falsificationist claim, or, at least, that's my view (no idea what my co-authors think on this). Our claim is that modern academic discourse routinely makes evidence-free assertions about present "systemic racism" causing inequality and gaps. Implicitly, (imho), our argument is that if you make an evidence-free assertion you do not require the assertion to be based on evidence. If you do not require evidence, your assertion may be many things, but it cannot possibly be falsified. That is, its dogma (and I won't repeat all our arguments about the politicization of science here because that is the core of the paper).
In the end, we (and certainly I) probably care less about whether its science or not; the bigger problem is that its a very particular political strain of dogma claiming the mantle of science. And if "systemic racism" (or any of central claims of CSJ, e.g., regarding sex, [trans[gender]], colonialism, or anything) cannot be falsified, generates few falsifiable predictions, and is still claimed even when its predictions are falsified, then, in my view, its really silly to treat this as any form of science at all.
-On the larger point about whether your paper is good and whether the view you're attacking is worthy of attack: I don't disagree with you at all! I like the paper quite a bit, I think your conclusions are substantially correct, and my concerns are really, in the end, nitpicks. Still, I feel like these are nits that need picking because they are real flaws in a paper that would have been better served by not having those flaws. And the nature of those flaws relates to long-standing pet peeves I have about the attitude many scientists have towards the philosophy of science. So really, I have an axe to grind, and I'm using your paper as an excuse. In this post, I'm focusing on falsificationism not because that's central to the case you're making (it isn't) or because you talk about it a lot (you don't), but because I think the problems with falsificationism are really quite pronounced and it frustrates me that scientists (including you and your co-authors) seem stuck on the idea. (To get a sense of my frustration, imagine if philosophers talked about psychology from time to time (which we do), but every time we did, we did so from a Freudian perspective. And perhaps a relatively well-informed Freudian perspective, like it's clear we paid attention in class on Freud day, but... why FREUD? Sure, he's a historically important psychologist and still has a handful of adherents. But the field moved on, long ago, for good reason! This is how philosophers feel whenever a scientist quotes Popper at them.)
-The point about falsificationism is really minor, but there's a half dozen other small issues I have with the way you lay things out in the first third or so of the paper. A further illustration of the kinds of concerns that I had, which I didn't bring up in the podcast: I appreciate the citations of Merton and Rauch, but they seemed somewhat irrelevant to the point you're trying to make. Merton is doing positive sociology, describing what scientists DO. Merton's norms are norms in the sociologists sense: these are rules that scientific communities DO follow. It's a further question about what norms scientific communities SHOULD follow. Does following Mertonian norms make for better science? Or could deviation from those norms in some respects improve the quality of scientific output? That's a hotly debated question in the philosophy of science!
-I think the editorial process really failed on this paper. Initially rejected for transparently absurd reasons, then rushed into JCI just because JCI wants to be the kind of place to publish papers like this, but without enough attention paid to tightening up some of the stuff towards the beginning of the paper. If I were peer-reviewing the paper, it would have given a recommendation of R&R: good, important paper, should definitely be published, but the authors should take another pass at the theory section, maybe bring someone else on board who can help with that part. Maybe this is just me being a grumpy philosopher, upset that you're walking in my territory without being able to say the proper shibboleths. But I sincerely think that these theoretical issues about what makes for good science are important, and it seems you do so as well, and philosophers have spent A LOT of time and effort on that question over the last century, and the paper just doesn't demonstrate familiarity with that literature. I don't want to suggest that you or any other particular person are unfamiliar with the relevant philosophy. Papers with this many authors tend to be camels and so perhaps things just got turned around somehow in the drafting process. But I think it's telling that three professional philosophers who are all on your side on the big picture question had the same reaction to the paper's treatment of philosophical issues.
-So to the question of falsificationism itself: The claim that life exists somewhere in the universe is unfalsifiable, and it does not "lead to" any falsifiable predictions. Consider the logical relation between the claims "There is life on at least one other planet" and "There is life on Mars." The latter entails the former; Mars is another planet, so if there is life there, then there is life on at least one other planet. But the former does not entail the latter. If there is life on at least one other planet, that other planet doesn't have to be Mars, it could be any other planet in the universe. So "There is life on at least one other planet" does not entail, and thus does not "lead to" (in any obvious sense) the falsifiable prediction that there is life on Mars. Nor does it lead to the falsifiable prediction that there is life on Titan, or LV-426, or anything else. So there's your counter-example of a clearly scientific claim that is not falsifiable, nor does it lead to any other falsifiable predictions. Perhaps I'm misunderstanding your way of thinking. Can you say a bit more about the sense in which "there is life on other planets" leads to falsifiable predictions?
- Let's take a step back. What are we trying to get at here? You say that your CSJ opponents put forward claims that aren't supported by evidence. Perhaps! I'm inclined to agree with that, at least in part. So perhaps the central point is that CSJ claims aren't supported by evidence. Well, what is evidence? (Again: this is a HUGE question in epistemology and the philosophy of science.) Do we insist that all evidence must be empirical? That seems self-undermining. "All evidence must be empirical" is not a claim that is supported by any empirical evidence. So we're stuck, right off the bat. One way forward is to try to give a more expansive account of what evidence consists in, which covers both empirical evidence and some hard-to-articulate kind of theoretical evidence. If we do that, then maybe CSJ stuff is supported by the hard-to-articulate kind of theoretical evidence. (This is what many philosophers who are sympathetic to CSJ would say.) Another way forward is to restrict the range of claims that we say must be supported by empirical evidence. So not "all evidence must be empirical" but "all evidence for scientific claims must be empirical." This is the route Popper took, and it's why he considered the question of what makes a claim scientific to be so important. We also need to ask what it means for a claim to be supported by empirical evidence. We have a vague, intuitive sense of this, perhaps. Scientific claims need to be able to be TESTED by experience. Ok, but what does that mean? Tested how? There's lots of possible answers to that question that have been defended by various philosophers over the years. Falsificationism is one (bad) answer to that question. There are plenty of related-but-distinct questions here. And I think you're right to say that there are big problems for CSJ in this area. So I think I agree with your larger point. The evidential support that has been provided for various CSJ claims is severely lacking, and part of the problem is that it's not clear how some of those claims can be tested against the tribunal of experience. But in the paper and in your response here, this concern is summarized by saying "these claims are unfalsifiable." That's just not a good way to summarize that point, because it imports Popperian assumptions which really don't hold up to scrutiny.
I’m not very familiar with the work of Karl Popper; I found his writing somewhat incomprehensible, a commonality I find for philosophy books translated from another language. I will, however, defend the importance of falsifiability to the scientific method. Though my version may differ from Popper’s.
Life on Other Planets
I believe, as is most often the case in philosophical debate, that at least part of the disagreement here is due to confusion between different definitions of the term science. The term I believe most often is used to refer to the body of empirical knowledge of reality, especially–but not exclusively–knowledge acquired using the scientific method. It also includes knowledge we have from making simple observations: facts like the Earth has one moon, humans have many bones, volcanoes can eject lava when erupting, etc.. It also includes knowledge from measurements: the density of gold is 19.3 g/cm3 and the melting point of tin is 449.47 °F.
But the word science can also refer to the scientific method itself. It is this sense of the word we are using when we say a scientific claim must be falsifiable. What we really mean is that in order for a claim to be tested with the scientific method, it must be falsifiable.
The claim that there is non-terrestrial life somewhere in the universe is not falsifiable, at least not without some crazy new technology that allows us to scan every object in the universe simultaneously. However, the null hypothesis of this, that is the claim that there is NOT non-terrestrial life anywhere in the universe, is falsifiable. You just have to find one instance of extraterrestrial life and the claim is falsified. To test the hypothesis, you could do a survey of a few hundred million planets and if you still find no life, your hypothesis would be supported. But not verified. Even if you survey every planet, moon, star, comet, asteroid and nebula and find no life, you have still not verified your hypothesis. There is always the possibility that the life you were looking for found a way to evade your detection or that the life was so foreign to you that you weren’t able to recognize it as life. There is always the potential for unknown unknowns.
We need not jump through these hoops, however, if we directly observe life from other planets. Any observation of reality can be entered into the body of knowledge we call science. It’s just that those observations are open to interpretation, and the relevance of observations made using the scientific method is much clearer.
Conjunctions
It is true that every experiment has assumptions, but these are not conjunctions of the main hypothesis. If it were something being tested, we wouldn’t call it an assumption. If you don’t believe the assumptions are justified, then you should not trust the results of the experiment.
If you were confident enough that the hypothesis were true, you could treat that as an assumption and then perhaps the experiment could be said to be a test of another assumption that you aren’t confident is true, but then you’ve just changed it to a hypothesis. Either way, we can only evaluate one claim at a time and it must be falsifiable.
In summary, [if A, then B] =/= [A and B].
Or if I’ve made some mistake there, I see no reason not to say that each part of your hypothesis must be falsifiable AND all your assumptions must be falsifiable.
Clarification
Again, I am not necessarily trying to defend Karl Popper’s version of falsifiability. I’m sure I would have many issues with his proposal of the idea. And to be clear, when I say a claim must be falsifiable, it doesn’t mean that a negative result means we reject the claim forever. But if there’s no way to get a negative result for a claim, then just by doing lots and lots of experiments to test the claim I will eventually get some positive results. Then I would be able to present the positive results and no one could counter the negative results. In this way, every unfalsifiable claim will be supported. And since there are an infinite number of potential unfalsifiable claims, we would end up believing an unlimited amount of nonsense.
Heh. It was not rushed. We *did* get an R&R. Boghossian is a philosopher. Singer is a bigtime philosopher. You (or anyone) of course can still disagree with anything in there but the argument "you should have had some philosophers involved in the writing and reviewing" is wrongheaded, because we did. Of course, you can still say "we had the wrong philosophers!" or "they did not do their jobs well!" or "they should have had MORE philosophers" or some such. But you really can't say we should have had philosophers as if we didn't because we did.
On the merits, I will do one more pass at an explanation that is, perhaps, a slight retreat. What I care about is whether some claim is true or not. And I subscribe to (philosopher!) Huemer's Trivial Theory of Truth: "“Sometimes people say things. When you say things, sometimes stuff is the way that you say it is. Other times, it isn’t. When stuff is the way that you say it is, we call your statements “true”.
That’s it. I told you it was a simple and basic concept. I shall call this ‘the Trivial Account of Truth.’”
I really do not care much whether unfalsifiable speculations can sometimes be considered science. I do care a lot about whether unfalsifiable speculations are being presented as if they are true. Systemic racism comes to mind. It *can* be defined in such a way as to evaluate whether a claim that is true is being made, but it seems to me very often it is not. It is often not defined. It is often claimed without citation regularly, as if evidence isn't even needed. It is a catch-all "explanation" for gaps, but it is very obvious that not all gaps come from discrimination, so one needs more than a drumbeat of "systemic racism!" to conclude that *any particular gap results from *any racism here and now whatsoever.* However, its advocates will rarely accept *anything* as refuting "systemic racism here and now!" including extreme evidence of anti-White, anti-Asian and pro-Black discrimination at UNC and Harvard presented in the recent SCOTUS case. Rather than recognizing that evidence of East Indian, Carribean Black, East Asian, and Cuban success (especially higher levels of college degrees and incomes than White people) disconfirms the "White Supremacy!" narrative, excuses are made. So if you reject all evidence of non-White people outperforming White people as inconsistent with a White Supremacy!/Systemic Racism! analysis, then this is a political argument, not a scientific one. The easiest way to cast this is as unfalsifiable. If one wishes to cast it in some other way to treats it as political nonsense that one should take seriously, that's fine with me.
To respond with edge cases of scientific speculation as "hey, life in the universe is science and its not falsifiable!" is fine but misses the point. If Delgado and Crenshaw and Kendi and DiAngelo want to recast everything theysay as "well, here are some speculations, we do not really know if any of this is true and no one should act like it is true" (as, say, scientists speculating on life in the universet) that's fine with me. Academia would be greatly improved by such clarification.
I don't see much to disagree with anymore here. I continue to think that some parts of your argument missed the mark, and had there been a different writing/editorial process that mark wouldn't have been missed by as much. But I shouldn't have speculated about the process that led to publication.
And I definitely agree that there are massive epistemic defects with how some critical social justice views are supported! (I said on the podcast that I'd have appreciated a bit more steel-manning, and I stand by that, but Spencer was right to push back because the dumb versions of the views you're attacking are depressingly popular.)
I'm glad you take my point that "falsifiability" might not be the best lens through which to view those defects. But to concede the point back to you a bit: confirmation bias is genuinely a bad thing, as is dogmatic thinking. Calling them out is entirely appropriate, and it's not obviously wrong to use the word "falsifiability" when you do, although other conceptual schema would be better.
Heh. I admit it, I do not get the objection. First, we did not spend the first third of the paper on falsificationism. In fact, we hardly address the issue.
The first page and a half or so is on how great science is. Then we get to our thematic point:
"Fulfilling this responsibility, however, is being hindered by a new, alarming clash between liberal epistemology and identitybased ideologies. Liberal epistemology prizes free and open inquiry, values vigorous discourse and debate, and determines the best scientific ideas by separating those that are true from those that are likely not. The statuses, identities, and demographics of scientists are irrelevant to this great sifting of valid versus invalid ideas.
In contrast, identity based ideologies seek to replace these core liberal principles, essential for scientific and technological advances, with principles derived from postmodernism and Critical Social Justice (CSJ), which assert that modern science is “racist,” “patriarchal,” and “colonial,” and a tool of oppression rather than a tool to promote human flourishing and global common good."
There is this single sentence on falsificationism:
"CSJ is not an empirical theory, because its tenets are maintained despite their being either
demonstrably false or unfalsifiable." I stand by that. Astrology is not a scientific theory because it is demonstrably false. God created the universe is not a scientific theory because it is not falsifiable and, in at least the Biblical variant, has led to falsified predictions (e.g., the Earth is the 6000 years old). There is life somewhere in an infinite universe is absolutely an edge case. As you say, it can be verified by not falsified. However, what makes it an edge case (in my view) is that It does lead to falsifiable predictions (e.g., about life on Mars or exoplanets). Find some claim that is not merely unfalsifiable, but which has not been falsified and which is incapable of leading to any falsifiable predictions, and I'd consider changing my view.
Good discourse on auxiliary assumptions, though. Totally agree there. Auxiliary assumptions permit scientists to write apologia defending their failed predictions and never having to surrender their cherished ideas.
But let's put all that aside. I plead innocent to the charge of "philosophical duffer" but, hey, perhaps The Court of Serious Philosophy convicts me anyway.
Its a single sentence of the paper. Lets review how we continue:
"The existence of objective reality, for example, which CSJ denies, is attested to by every successful engineering project, from bridges to satellites, from cell phones to electric cars, ever conducted. The fallibility of “lived experience” is attested to by a wealth of psychological research demonstrating errors and biases in self reports."
Here, we argue that certain CSJ claims are not unfalsifiable, but false. Every successful engineering project works because the engineers understood objective reality, thereby *falsifying* claims that it does not exist. "Lived experience" -- which is often put on some sort of bizarre pedestal by many CSJ perspectives and advocates -- is not completely useless, but so fallible that it does not deserve a (dare I say it?) "privileged" place is social science. Those claiming or implying otherwise are promoting falsified claims when they argue that "lived experience" should be privileged over other sources of social science knowledge.
We continue:
"Below we discuss publications making unsupported claims of systemic injustices and attacking merit. Such publications rarely, if ever, provide evidence that observed disproportionalities in the race or gender distribution of a scientific field are the result of present day structural or systemic racism."
Here, we are making an implicit falsificationist claim, or, at least, that's my view (no idea what my co-authors think on this). Our claim is that modern academic discourse routinely makes evidence-free assertions about present "systemic racism" causing inequality and gaps. Implicitly, (imho), our argument is that if you make an evidence-free assertion you do not require the assertion to be based on evidence. If you do not require evidence, your assertion may be many things, but it cannot possibly be falsified. That is, its dogma (and I won't repeat all our arguments about the politicization of science here because that is the core of the paper).
In the end, we (and certainly I) probably care less about whether its science or not; the bigger problem is that its a very particular political strain of dogma claiming the mantle of science. And if "systemic racism" (or any of central claims of CSJ, e.g., regarding sex, [trans[gender]], colonialism, or anything) cannot be falsified, generates few falsifiable predictions, and is still claimed even when its predictions are falsified, then, in my view, its really silly to treat this as any form of science at all.
Hi Lee, thanks for such an in-depth reply.
-On the larger point about whether your paper is good and whether the view you're attacking is worthy of attack: I don't disagree with you at all! I like the paper quite a bit, I think your conclusions are substantially correct, and my concerns are really, in the end, nitpicks. Still, I feel like these are nits that need picking because they are real flaws in a paper that would have been better served by not having those flaws. And the nature of those flaws relates to long-standing pet peeves I have about the attitude many scientists have towards the philosophy of science. So really, I have an axe to grind, and I'm using your paper as an excuse. In this post, I'm focusing on falsificationism not because that's central to the case you're making (it isn't) or because you talk about it a lot (you don't), but because I think the problems with falsificationism are really quite pronounced and it frustrates me that scientists (including you and your co-authors) seem stuck on the idea. (To get a sense of my frustration, imagine if philosophers talked about psychology from time to time (which we do), but every time we did, we did so from a Freudian perspective. And perhaps a relatively well-informed Freudian perspective, like it's clear we paid attention in class on Freud day, but... why FREUD? Sure, he's a historically important psychologist and still has a handful of adherents. But the field moved on, long ago, for good reason! This is how philosophers feel whenever a scientist quotes Popper at them.)
-The point about falsificationism is really minor, but there's a half dozen other small issues I have with the way you lay things out in the first third or so of the paper. A further illustration of the kinds of concerns that I had, which I didn't bring up in the podcast: I appreciate the citations of Merton and Rauch, but they seemed somewhat irrelevant to the point you're trying to make. Merton is doing positive sociology, describing what scientists DO. Merton's norms are norms in the sociologists sense: these are rules that scientific communities DO follow. It's a further question about what norms scientific communities SHOULD follow. Does following Mertonian norms make for better science? Or could deviation from those norms in some respects improve the quality of scientific output? That's a hotly debated question in the philosophy of science!
-I think the editorial process really failed on this paper. Initially rejected for transparently absurd reasons, then rushed into JCI just because JCI wants to be the kind of place to publish papers like this, but without enough attention paid to tightening up some of the stuff towards the beginning of the paper. If I were peer-reviewing the paper, it would have given a recommendation of R&R: good, important paper, should definitely be published, but the authors should take another pass at the theory section, maybe bring someone else on board who can help with that part. Maybe this is just me being a grumpy philosopher, upset that you're walking in my territory without being able to say the proper shibboleths. But I sincerely think that these theoretical issues about what makes for good science are important, and it seems you do so as well, and philosophers have spent A LOT of time and effort on that question over the last century, and the paper just doesn't demonstrate familiarity with that literature. I don't want to suggest that you or any other particular person are unfamiliar with the relevant philosophy. Papers with this many authors tend to be camels and so perhaps things just got turned around somehow in the drafting process. But I think it's telling that three professional philosophers who are all on your side on the big picture question had the same reaction to the paper's treatment of philosophical issues.
-So to the question of falsificationism itself: The claim that life exists somewhere in the universe is unfalsifiable, and it does not "lead to" any falsifiable predictions. Consider the logical relation between the claims "There is life on at least one other planet" and "There is life on Mars." The latter entails the former; Mars is another planet, so if there is life there, then there is life on at least one other planet. But the former does not entail the latter. If there is life on at least one other planet, that other planet doesn't have to be Mars, it could be any other planet in the universe. So "There is life on at least one other planet" does not entail, and thus does not "lead to" (in any obvious sense) the falsifiable prediction that there is life on Mars. Nor does it lead to the falsifiable prediction that there is life on Titan, or LV-426, or anything else. So there's your counter-example of a clearly scientific claim that is not falsifiable, nor does it lead to any other falsifiable predictions. Perhaps I'm misunderstanding your way of thinking. Can you say a bit more about the sense in which "there is life on other planets" leads to falsifiable predictions?
- Let's take a step back. What are we trying to get at here? You say that your CSJ opponents put forward claims that aren't supported by evidence. Perhaps! I'm inclined to agree with that, at least in part. So perhaps the central point is that CSJ claims aren't supported by evidence. Well, what is evidence? (Again: this is a HUGE question in epistemology and the philosophy of science.) Do we insist that all evidence must be empirical? That seems self-undermining. "All evidence must be empirical" is not a claim that is supported by any empirical evidence. So we're stuck, right off the bat. One way forward is to try to give a more expansive account of what evidence consists in, which covers both empirical evidence and some hard-to-articulate kind of theoretical evidence. If we do that, then maybe CSJ stuff is supported by the hard-to-articulate kind of theoretical evidence. (This is what many philosophers who are sympathetic to CSJ would say.) Another way forward is to restrict the range of claims that we say must be supported by empirical evidence. So not "all evidence must be empirical" but "all evidence for scientific claims must be empirical." This is the route Popper took, and it's why he considered the question of what makes a claim scientific to be so important. We also need to ask what it means for a claim to be supported by empirical evidence. We have a vague, intuitive sense of this, perhaps. Scientific claims need to be able to be TESTED by experience. Ok, but what does that mean? Tested how? There's lots of possible answers to that question that have been defended by various philosophers over the years. Falsificationism is one (bad) answer to that question. There are plenty of related-but-distinct questions here. And I think you're right to say that there are big problems for CSJ in this area. So I think I agree with your larger point. The evidential support that has been provided for various CSJ claims is severely lacking, and part of the problem is that it's not clear how some of those claims can be tested against the tribunal of experience. But in the paper and in your response here, this concern is summarized by saying "these claims are unfalsifiable." That's just not a good way to summarize that point, because it imports Popperian assumptions which really don't hold up to scrutiny.
I’m not very familiar with the work of Karl Popper; I found his writing somewhat incomprehensible, a commonality I find for philosophy books translated from another language. I will, however, defend the importance of falsifiability to the scientific method. Though my version may differ from Popper’s.
Life on Other Planets
I believe, as is most often the case in philosophical debate, that at least part of the disagreement here is due to confusion between different definitions of the term science. The term I believe most often is used to refer to the body of empirical knowledge of reality, especially–but not exclusively–knowledge acquired using the scientific method. It also includes knowledge we have from making simple observations: facts like the Earth has one moon, humans have many bones, volcanoes can eject lava when erupting, etc.. It also includes knowledge from measurements: the density of gold is 19.3 g/cm3 and the melting point of tin is 449.47 °F.
But the word science can also refer to the scientific method itself. It is this sense of the word we are using when we say a scientific claim must be falsifiable. What we really mean is that in order for a claim to be tested with the scientific method, it must be falsifiable.
The claim that there is non-terrestrial life somewhere in the universe is not falsifiable, at least not without some crazy new technology that allows us to scan every object in the universe simultaneously. However, the null hypothesis of this, that is the claim that there is NOT non-terrestrial life anywhere in the universe, is falsifiable. You just have to find one instance of extraterrestrial life and the claim is falsified. To test the hypothesis, you could do a survey of a few hundred million planets and if you still find no life, your hypothesis would be supported. But not verified. Even if you survey every planet, moon, star, comet, asteroid and nebula and find no life, you have still not verified your hypothesis. There is always the possibility that the life you were looking for found a way to evade your detection or that the life was so foreign to you that you weren’t able to recognize it as life. There is always the potential for unknown unknowns.
We need not jump through these hoops, however, if we directly observe life from other planets. Any observation of reality can be entered into the body of knowledge we call science. It’s just that those observations are open to interpretation, and the relevance of observations made using the scientific method is much clearer.
Conjunctions
It is true that every experiment has assumptions, but these are not conjunctions of the main hypothesis. If it were something being tested, we wouldn’t call it an assumption. If you don’t believe the assumptions are justified, then you should not trust the results of the experiment.
If you were confident enough that the hypothesis were true, you could treat that as an assumption and then perhaps the experiment could be said to be a test of another assumption that you aren’t confident is true, but then you’ve just changed it to a hypothesis. Either way, we can only evaluate one claim at a time and it must be falsifiable.
In summary, [if A, then B] =/= [A and B].
Or if I’ve made some mistake there, I see no reason not to say that each part of your hypothesis must be falsifiable AND all your assumptions must be falsifiable.
Clarification
Again, I am not necessarily trying to defend Karl Popper’s version of falsifiability. I’m sure I would have many issues with his proposal of the idea. And to be clear, when I say a claim must be falsifiable, it doesn’t mean that a negative result means we reject the claim forever. But if there’s no way to get a negative result for a claim, then just by doing lots and lots of experiments to test the claim I will eventually get some positive results. Then I would be able to present the positive results and no one could counter the negative results. In this way, every unfalsifiable claim will be supported. And since there are an infinite number of potential unfalsifiable claims, we would end up believing an unlimited amount of nonsense.