Monday, December 12, 2016

Science is not the only way to know things: a rebuttal to Lawrence Krauss

Epistemological naturalism, otherwise known as scientism, is the view which states that science is the source of all knowledge. To assess the merits of the view, one must get clear on what one means by science as well as what one means by knowledge. More than just an exercise in nitpicking, the task of sorting out adequate definitions is important to making advances in philosophical debates. In this essay, I will target some recent philosophical claims advanced by the physicist Lawrence Krauss and argue that he defends a view which understands science too broadly, and knowledge too restrictively. Krauss’s claims are not merely semantic; they are controversial claims about the nature of science and knowledge, claims that have been widely discussed and examined by both philosophers of science and epistemologists.

Krauss is a strong proponent of two closely related theories of knowledge; empiricism and scientism. Empiricism states that sense experience is the source of all knowledge, whereas scientism states that science is the sole source of knowing. While these two views are closely related, one could be an empiricist without endorsing scientism and vice versa. In a series of recent debates and discussions, Krauss has made a number of philosophical claims with little to no argumentative support. Among several others, Krauss has stated that “There are no such thing as non-empirical facts” and that “science is the sole source of knowing”. One might argue that Krauss's statements on these issues should not be taken as representative of his actual views. It's possible that Krauss might come to recognize that many of his statements were sloppy or mistaken, and that he actually holds views that are much more plausible. I don't buy this. Krauss has been very consistent in how he answers questions about knowledge and science over the years. His recent statements should, therefore, be taken to express views he sincerely believes. Furthermore, his statements about the nature of science and philosophy have a wide audience. He has written several popular books and has appeared in dozens of public debates and discussions with scientists and philosophers like. As a theoretical physicist, Krauss rightly recognizes the importance and usefulness of doing science, but I think he gets several things seriously wrong when he starts talking about knowledge and how the practice of science should be defined. 

For instance, take Krauss’s definition of science; “rational thought applied to empirical evidence”. Empirical evidence is clearly important for doing science, and so is rational thought, but science is not the only discipline that meets this description. For instance, most contemporary philosophers would be scientists under Krauss’s definition. One might object by saying that his definition is too broad as it captures many other academic disciplines thought to be distinct from science. But the vagueness of the definition also captures countless other forms of activities and practices. If we accept his definition, plumbers are regularly doing science because they need to make observations about pipes and faucets and make rational decisions while troubleshooting problems. Artists would also be doing a lot of science because they need to rely upon sense experience and rational thought to determine how they are going to create their artwork. In Krauss's own words, "we all do science every single day". Such consequences demonstrate the inadequacy of his definition. After all, a good definition should not be open to an endless number of counter examples. An easy way to remedy this problem would be to give a much more specific and accurate set of conditions for what 'science' is. 

If one understands Krauss to be stipulating a definition, then his claim is trivial. One could just as easily stipulate that philosophy should be defined as the exercise of critical thinking or careful reflection. Since we need to think and reason about what we know, philosophy is indispensable for knowing things about the world. Therefore, philosophy is the source of all knowledge. But I take it that these disagreements are not merely semantic, but a disagreement about what ‘science’ actually is. Like in his recent book on how the universe came from “nothing”, Krauss illegitimately defines his terms and then (ironically) accuses his critics of just playing with semantics. Playing with semantics to reach a philosophical conclusion happens to be a common trend amongst scientists writing popular books lately (Harris 2011, Krauss 2012, Wilson 2014).

With the difficulties facing Krauss’s definition of science, it is unsurprising to find that his claims about knowledge are subject to many of the same criticisms. Krauss seems to think of the activity of doing science as the source of knowledge. But science, understood as a complex intellectual activity, can be broken down into more basic components. When epistemologists talk about sources of knowledge, they usually have in mind things like introspection, memory, perception, reasoning, and testimony. Scientific knowledge is comprised of a complex combination of all of these. 

Scientists rely upon their perceptual faculties when they make observations of the world. They use reason to make inferences and deductions about what they’ve observed. And they frequently rely upon the testimony of their colleagues and scientists of previous generations. Once one realizes that knowledge can be derived by more basic sources (than scientific investigation) it is easy to see why countless instances of knowledge that are by no means scientific. Here are five examples:

v  1: I know whether I am experiencing pain, hunger or thirst (via introspection).
v  2: I know that I turned off my television set before I headed out to the store (via memory).
v  3: I know that my good friend recently got engaged (via testimony).
v  4: I know that “There are an infinite amount of prime numbers” (mathematical reasoning).
v  5: I know that it is wrong to torture children for fun (moral reasoning)

In considering such examples, Krauss would likely point out that they all involve sense experiences, to some degree, providing good support for his belief in empiricism. The problem here is that Krauss misunderstands the debate between empiricism and its opposing view, rationalism. Rationalism states that you can come to know certain things are true through reasoning. Rationalists do not think that you can conduct reasoning without having experiences (e.g. introspection). That would be absurd. They are claiming that the warrant or justification for believing certain propositions comes from the chains of reasoning themselves, rather than something you have to experience or observe. For instance, take logic and math. If you know that Tom is a bachelor, you can come to know that Tom is an unmarried man. To know this, you don’t need to go out and do an investigation of Tom or look at statistics on bachelors. It’s a conclusion that you can come to know just by thinking about the meaning of the concept ‘bachelor’. Tellingly, Krauss disagrees (timestamp: 20:50). He thinks one needs to look out at the world to see whether all bachelors are in fact married. We can observe that all bachelors are actually unmarried men, therefore, it is empirical evidence, that grounds that fact. I will say more on why this is deeply confused later.

Similarly, with mathematics, you just have to sit back and reflect upon the nature of ‘prime numbers’ and ‘infinity’ in order to work out why it is true that 'there is no largest prime'. To make the point more finely, if experience was what really gave you the truth of mathematical propositions, then you could encounter things in the world that might falsify basic mathematical truths like “2+2=4”. Strangely, Krauss, in misunderstanding empiricism, bites an unnecessary bullet on this point (1). In conversation with Peter Singer, Krauss has stated that if you were to encounter a situation where, say, two pairs of apples are put into an empty box, to yield five apples, you would have grounds for thinking that 2+2=5. Singer (rightly!) says that the rational thing to do in such a situation would be to assume that some kind of magic trick was performed, rather than to revise your mathematical beliefs. Your experience of the world has nothing to say about the answers to mathematical equations or basic logic questions.

The irrelevance of empirical evidence to fields that deal largely in the abstract explains why mathematicians and logicians have made great progress with just pencils and paper. (Side note: I tend to think that philosophers occupy a kind of middle ground between math/logic and science. Some are interested in questions where empirical evidence is mostly irrelevant (e.g. metaethics) while others, like myself, are interested in issues (e.g. how the mind works) that are intimately connected with scientific investigations.)

One might argue that since facts about mathematics and logic aren’t about the physical world, they aren’t bona fide facts. They’re just relations of ideas or conceptual truths. This is what some empiricists, such as David Hume, argued. But why believe that facts have to be about the physical world? What makes propositions about the physical world privileged? Is there a sense of objectivity that comes with science that is not realized in mathematics or logic? If so, what exactly is the relevant difference between claims about the world and claims about formal systems? A possible answer would involve appealing to yet another philosophical view, scientific realism. Unlike in mathematics or logic, the entities and postulates described by scientific theories are objectively real, independent of human thought. Mathematics and logic are just useful human inventions and do not bear on claims of truth. Judging by what Krauss has said about the nature of science, he strikes me as an ardent scientific realist. For instance, Krauss has stated “It is nature that determines what facts are, not people” (my emphasis).

While I do not have the space to defend such claims, I want to say that any proposition—a statement that is either true or false—is a candidate for knowledge, and that a fact is to be understood as a proposition that is true (e.g. 2+2=4). Therefore, there can be facts and knowledge about mathematics, logic, and perhaps even morality and metaphysics. In a future post, I will try and grapple with some of the deeper questions raised in these closing sections. Specifically, I will discuss scientific realism and determine whether or not the view is strong enough to justify claims regarding science as a privileged epistemic standpoint. 

Perhaps the take away message from Krauss's philosophical ventures is that he nicely illustrates how not to do philosophy, and demonstrates the need for philosophical rigor and understanding. Philosophy has applications not only within domains of study that are mostly nonempirical (e.g. ethics, aesthetics, metaphysics), but for both doing and reflecting upon science. As the American philosopher Daniel Dennett has aptly put it,

“There is no such thing as philosophy-free science; there is only science whose philosophical baggage is taken on board without examination." (Dennett 1996)

Works cited:

  • Dennett, D. C. (1996). Darwin's dangerous idea: Evolution and the meanings of life. New York: Simon & Schuster.
  • Harris, S. (2011). The moral landscape: How science can determine human values. Simon and Schuster.
  • Krauss, L. M. (2012). A universe from nothing. Simon and Schuster.
  • Wilson, E. O. (2014). The meaning of human existence. WW Norton & Company.

Thursday, December 1, 2016

Applied critical thinking: Expert testimony

We often take the testimony of experts for granted. When reading the newspaper or watching a documentary program, it seems reasonable to accept certain claims, provided that they are stated by individuals who have the right sort of credentials. But for areas of genuine controversy, it would be unwise to just accept expert testimony at face value. If we are to take sides in an area of controversy, we ought to be able to explain why a given expert is right and why others who disagree are wrong. There are plenty of instances where experts strongly disagree. For instance, there are doctors who believe acupuncture is an effective treatment for muscuoskeletal pain, and others who believe it doesn’t work, nothing more than an elaborate placebo. In such cases, it might not be obvious whose testimony we ought to trust, especially if one knows little to nothing about medicine. Figuring out who to trust can be complicated and will likely take some time. Through personal experience, I have encountered many friends and family who are disposed to throw their hands up in the air whenever areas of controversy are brought up. How can we—as nonexperts—ever decide who is telling the truth? How can we know? Instead of adopting agnosticism regarding all areas of controversy, most of us, in practice, do listen to some experts and ignore others. For instance, we are more likely to accept the testimony and advice of experts who share our own views. Instead of trying to confirm the beliefs we already hold, or engage in wishful thinking, we ought to critically evaluate competing expert testimony to the best of our ability. In what follows, I will expand upon a proposal developed by the philosopher Alvin Goldman, aimed to help one decide which experts to trust.

First step: Sift out the pseudoexperts
Before discussing the problem of how to choose between experts, we need to determine who has expertise in the first place. The thought is that once we eliminate all of the phony experts, we can move on to the harder questions of how to decide between the genuine experts. Perhaps a sufficient condition for being an expert in some domain of study would be the possession of an advanced degree—in the relevant field—awarded by a recognized academic institution. For instance, an expert in physics would be expected to have a PhD in physics. However, the knowledge set of your typical PhD is likely to be highly specialized. For instance, a scientist with a PhD in physics may be an expert regarding particle physics, but know very little about astrophysics or applied physics. In some cases, it may be unclear which area of study is the most relevant to the issue at hand. Here’s one example. For dietary advice, one might consider a nutritionist to be the most relevant authority to consult. While nutritionists might know a fair amount about dieting and nutrition, a better source would be a registered dietician. Dieticians tend to have much more training in science and medicine than a nutritionist, and have to pass a comprehensive exam to become certified. Overall, they are more qualified to be making judgments about dietary claims than nutritionists. Therefore, with regard to claims about dieting, the relevant experts are dieticians, not nutritionists. Individuals who purport to be experts on certain matters, but whom lack the relevant qualifications and/or training should raise red flags. In summary, before assessing expert testimony, one must try to answer the following questions:

What does X’s expertise consist in?
Is X’s expertise in any way relevant to the issue at hand?

Once one establishes that they have found a genuine expert and more importantly, one whose expertise is relevant to the question at hand, one must determine whether or not this expert is trustworthy. Are there reasons to doubt his or her testimony?

Steps for analyzing the testimony of true experts

It can be unhelpful to look at the testimony of one expert in isolation. From the point of view of a layperson (nonexpert) most experts tend to be highly persuasive. To get a better sense as to how reliable their testimony is, try to find an expert who disagrees with them, preferably, an expert that has had comparable training and experience. After finding two experts that disagree, it’s time to compare what each of them has to say. The philosopher Alvin Goldman proposes five ways to determine which expert is more trustworthy. I will deal with each one in turn and list some of the problems these guidelines face.

“(1) Read or listen to arguments and counter-arguments offered by the two experts, whether in a published exchange of views, an oral debate, or separate defenses of their respective positions.” (1)
Difficulty 1: The evidence and/or arguments discussed may include esoteric terminology. One may try and listen to the arguments and counter-arguments but fail to understand or even misunderstand them. Goldman makes a distinction between esoteric and exoteric terminology. Esoteric terms are not only unfamiliar to non-experts; they are inaccessible to them. This may because they involve unfamiliar concepts and theories. Exoteric terms are unfamiliar to non-experts but can that can be learned and understood by novices without any specialized training. Grasping exoteric terms may involve a subject to do some extra reading on the subject, whereas understanding esoteric terms may require one to become an expert in that field.

Difficulty 2: Superficially convincing arguments could be made to support one side of the debate, but these arguments may turn out invalid or contain false premises. If one tries to assess the arguments and counter-arguments of two disagreeing experts, one better have a decent working knowledge of informal logic. The expert committing more logical fallacies possesses fewer solid reasons for their belief. But a valid argument is not necessarily a sound one. Going back to the first difficulty, non-experts might not be able to tell whether a premise is true or not. To an expert, a given premise may be obviously false and contradicted by lots of evidence they are aware of. But to a layperson, it may seem to be plausible.

“(2) Find out what the opinions of other (putative) experts on the topic in question.    If most of them agree with expert A, then identify A as your best guide.  If most choose expert B, identify B as the more trustworthy one.  In short, go with the numbers to guide your choice of favored expert.”
Caveat: When it comes to issues where the vast majority of relevant experts are in agreement, going with the numbers is a good rule of thumb. For instance, the vast majority of climatologists accept that the planet is currently going through a warming trend and that this is primarily due to recent human activity. Should we accept the testimony of climatologists solely on the basis of consensus? In short, no. There are some possible scenarios where it would be rational to doubt consensus opinion (e.g. Nazi scientists during WW2). As long as the consensus position seems to be backed by valid arguments and independent sources of evidence, it is rational to side with the consensus.

“(3) Consult "meta-experts" about experts A and B.  Try to find out which of them is the superior expert by asking people in a position to compare and contrast them.  Or people who trained them or have worked with them.”

The idea is looking for additional experts other than the two you initially found. They might be in a good position to tell whether or not experts A or B have compelling arguments, especially if they have nothing to gain or lose in the debate. Where can one find meta-experts? You’re likely to find tons of meta-experts if you look through the peer-reviewed literature, reputable periodicals, or even personal blogs. In today’s age, many scientists blog to try and educate the general public about their work. One could even directly contact meta-experts at a local university or one could reach out to experts through email or on specialized web forums. And if you're lucky enough, you may personally know some meta-experts that could weigh in on the debate.

 “(4) Obtain evidence about the experts' biases and interests, which might lead them to self-serving answers of dubious veracity (whatever their underlying competence).”

In some cases, there is obvious bias and conflicts of interest at play with a given expert. Whether it is some political or religious agenda, these factors need to be taken into account. But one should be cautious when discrediting certain experts. It could be the case that some of the testimony is perfectly valid, or that their bias or conflicts of interest played no role in the formation of their beliefs about the issue at hand. Furthermore, an expert may be extremely biased but turn out to be right. Therefore, before one discredits an expert on grounds of bias or a conflict of interest, one must have some independent reason to think that their claims are wrong. Conspiracy theorists frequently misuse such a guideline when they discredit all experts who testify against their favored theory. Finding a potential source of bias (e.g. government funding) for some expert, and then completely disregarding everything they have to say is intellectually lazy and dishonest. The implication is that all such experts are lying or saying misleading things. If it can be shown that either is the case, that would be the reason to seriously doubt the expert’s testimony, not from just from the possibility of a conflict of interest.  

Another source of bias can be uncovered by carefully studying the behavior of a given expert. Experts who dismiss alternative positions out of hand should raise red flags. I say this because most experts tend to exhibit a certain kind of psychological profile. They tend to be fairly humble, well-integrated into their epistemic communities, and they seem to be genuinely interested in the truth. Experts tend to be very cautious when making controversial claims and admit that their own favored hypotheses could be mistake. They tend to actively engage their peers in academic journals and at conferences, try and test their hypotheses, and compare the success or failure of their own predictions to rival theories. Commenting on the small group of scientists who endorse the 9/11 conspiracy theories surrounding the tower collapses, Noam Chomsky has noted that, “They are not doing what scientists and engineers do when they’ve think they’ve discovered something" (2). Having studied the phenomenon of conspiracy theories quite extensively, I can vouch for Chomsky. Many of the “experts” who promote conspiracy theories lack the psychological profile found amongst genuine experts, and seem to share a number of opposing psychological traits. For instance, many manifest an excessive degree of pride or intellectual superiority, they tend to keep within a closed circle of peers who share their views, and they don't even attempt to convince the general scientific community about their "findings". These individuals might even possess the relevant knowledge to assess claims of interest, but their psychological profile and behavior casts serious doubt upon their capacity to seriously engage criticism and to critically evaluate their own positions. It is for this reason that their testimony should be taken with a grain of salt.

“(5)  Gather evidence of their past track-records and apportion trust as a function of these track records.”

Gathering evidence of an expert’s track record may turn out to be a challenge. Good track records of experts may come in the form of accurate and specific predictions. One might also try to look through their publication history or get a sense of their reputation amongst peers. Another way would be to look at an expert’s track record is to see if they have subscribed to controversial or fringe views in the past. Some experts can just be contrarians, while others seem to be suffering from crank magnetism. For instance, James Fetzer, a well-respected philosopher of science who taught critical thinking for most of his career, believes just about every conspiracy theory. He not only believes that 9/11 was an inside job, but that no children were killed at Sandy Hook, that we didn’t land on the moon, and that Paul McCartney died in the 60’s and was then replaced by someone with the same physical appearance, personality, and musical talents. In the case of Fetzer, there seems to be some kind of systematic misapplication of critical thinking going on (at least when he is theorizing about certain historical events). Those like Fetzer not only routinely appeal to pseudoexperts, but they accept many demonstrably false or highly questionable claims, make a number of unwarranted assumptions about human nature, and are apparently unskilled at making inferences to the best explanation. Since critical thinking is really a set of skills, knowledge of certain concepts and strategies related to the subject is not sufficient for knowledge how.

I believe that if one closely follows all of the advice outlined above, one will be more likely to acquire true beliefs about the world. But it is one thing to propose a strategy that makes sense in theory and quite another to have the strategy actually work for most people. Certain facts about our psychological limitations (e.g. confirmation bias, cognitive dissonance) may prevent us from being objective enough to really follow through with such advice. Nonetheless, we ought to at least try to be as objective as we can. Given the complexity of the world around us, we all need to appeal to experts at some time or another, whether it's to find out about our personal health, or how the world works. If we are going to appeal to experts, and are genuinely interested in discovering the truth, then it's a good idea to try and track down the right ones.

 1. Goldman, A. I. (2001). Experts: which ones should you trust?. Philosophy and Phenomenological Research, 63(1), 85-110.

2. Tuskin, B. (2013). Noam Chomsky has no opinion on building 7. Retrieved December 05, 2016, from