Is Science Failing Us? (RJS)

I came across an interesting article in Wired this week written by Jonah Lehrer, Trials and Errors: Why Science Is Failing Us. In this article Lehrer explores the complexity and false steps in the pharmaceutical and biomedical industries. He begins with a story about a promising drug being tested by Pfizer in 2006. This was supposed to be a real breakthrough (and money maker)  – but wound up pulled from clinical trial when it correlated with increased mortality. We spend, through industrial investment, government funding, and private foundations, an enormous amount of money each year on health related research. Most of it predicated on the idea that disease has a cause. Lehrer suggests that the fundamental flaw here is not in the specific research, but in the underlying assumption that “cause” is a real phenomenon capable of investigation. From Lehrer’s article:

This assumption—that understanding a system’s constituent parts means we also understand the causes within the system—is not limited to the pharmaceutical industry or even to biology. It defines modern science. In general, we believe that the so-called problem of causation can be cured by more information, by our ceaseless accumulation of facts. Scientists refer to this process as reductionism. By breaking down a process, we can see how everything fits together; the complex mystery is distilled into a list of ingredients. And so the question of cholesterol—what is its relationship to heart disease?—becomes a predictable loop of proteins tweaking proteins, acronyms altering one another. Modern medicine is particularly reliant on this approach. Every year, nearly $100 billion is invested in biomedical research in the US, all of it aimed at teasing apart the invisible bits of the body. We assume that these new details will finally reveal the causes of illness, pinning our maladies on small molecules and errant snippets of DNA. Once we find the cause, of course, we can begin working on a cure.

There are a number of problems with this paragraph, or so it seems to me, the first being that the search for cause is not merely reduction to the separable constituents. A “list of ingredients” is nothing without an understanding of the intricate multicomponent interactions and feedback mechanisms that occur only when they are all present together. Complexity does not point to a lack of causes, nor does it undermine the entire scientific enterprise.

Is science failing us?

or more concretely:

Does the complexity inherent in biology undermine the search for cause and effect on which science is based?

The article is long, but well worth reading. Lehrer makes many good points about the danger of oversimplifying and about the mistaken ideas and claims that have been made. He relates a story about the swing in medical practice from back pain as something that heals when left alone, to something that requires surgical intervention as indicated by new information available through MRI technology, to the current realization that many of the “abnormalities” observed by MRI were not the cause of back pain at all, but are as common among control populations as among the patients complaining of back pain. There are very important lessons here. Lehrer, however, goes a good step further – taking this and other similar examples of mistaken assessment as evidence that undermines “cause” as a real phenomenon.

The problem with this assumption, however, is that causes are a strange kind of knowledge. This was first pointed out by David Hume, the 18th-century Scottish philosopher. Hume realized that, although people talk about causes as if they are real facts—tangible things that can be discovered—they’re actually not at all factual. Instead, Hume said, every cause is just a slippery story, a catchy conjecture, a “lively conception produced by habit.” When an apple falls from a tree, the cause is obvious: gravity. Hume’s skeptical insight was that we don’t see gravity—we see only an object tugged toward the earth. We look at X and then at Y, and invent a story about what happened in between. We can measure facts, but a cause is not a fact—it’s a fiction that helps us make sense of facts.

The truth is, our stories about causation are shadowed by all sorts of mental shortcuts. Most of the time, these shortcuts work well enough. They allow us to hit fastballs, discover the law of gravity, and design wondrous technologies. However, when it comes to reasoning about complex systems—say, the human body—these shortcuts go from being slickly efficient to outright misleading.

Lehrer implies in his article that all causal reasoning is merely useful fiction. Sometimes it works well, even in complex biological and environmental systems. The increase in lifespan over the last century or so is an indication of how well this “fiction” can work. But…

And yet, we must never forget that our causal beliefs are defined by their limitations. For too long, we’ve pretended that the old problem of causality can be cured by our shiny new knowledge. If only we devote more resources to research or dissect the system at a more fundamental level or search for ever more subtle correlations, we can discover how it all works. But a cause is not a fact, and it never will be; the things we can see will always be bracketed by what we cannot. And this is why, even when we know everything about everything, we’ll still be telling stories about why it happened. It’s mystery all the way down.

It will probably come as no surprise that I think that Jonah Lehrer’s premise and reasoning is incorrect. Biological systems are incredibly complex – and the complexity plays out in ways that we are only just beginning to understand. Nonetheless causes are real and capable of investigation, from the Higgs boson, to global warming, to the origin and progress of cancer.

Perhaps you have a different idea – perhaps it is worth some conversation.

Do you think that causal beliefs are merely a useful fiction?

Are causes real or is “it” mystery all the way down?

If you wish to contact me directly you may do so at rjs4mail[at]att.net.

If interested you can subscribe to a full text feed of my posts at Musings on Science and Theology.

  • phil_style

    This discussion is probably beyond my pay-grade. But I find it hard to grasp what Lehrer is arguing here. It seems like a variation of the “god of the gaps” argument… but perhaps a “non-causal-chain of the gaps” argument.

    Although are there not some theories that suggest causality, in some cases, can be systemically top down? For example with respect to consciousness? In this case, the constituent parts of the system are, in fact, manipulated by the system itself – so that a reductionist “tinkering” with any of the individual constituent parts in the causal chain (to, for example, break any particular individual cause-effect relationship) would be thwarted as soon as the system goes back online, because the system is downwardly correcting.

    Some might push this to the cosmological extreme and use it to suggest some kind of teleology – that the future influences the present in some way. Although that kind of thinking is a bit excessive.

  • TSG

    I’m with phil_style in that this causal-chain argument is cosmological. It reminds me of the synchronic, diachronic, and deterministic contingency classifying in the social sciences. Contingency is used is so many ways that it loses it usefulness. That is, there are many side tracks. However, used correctly, one comes to a “no hidden factors” debate on indeterminism/determinism. If this sounds like an offshoot of a familiar theological theme that has been danced around endlessly-it is.

  • scotmcknight

    RJS,

    I agree with you: it seems he’s made a logical error in arguing complexity leading to our inability to explain something means a lack of cause. The more we learn means often enough the more complex things become, sometimes leading us to see that what we thought was a ’cause’ was only a small contribution…

    … but that back pain one is important. Bulging disks, not uncommon in my family (both of our children have had significant issues with herniated disks), may be typical in adults but the issue is impingement on nerve … that’s the cause… anyway good article.

  • http://mikeblyth.blogspot.com Mike Blyth

    It’s interesting that the problems cited are examples of ones that are more complex or still under active research. Would there not be a different conclusion if he wrote about tetanus, sickle cell anemia, and lead poisoning, just to take some random examples? Just because a system is complex does not imply that it’s governed by a whole different set of rules other than causality.

  • EricG

    I have a very strong negative reaction to this one. His primary target seems to be pharma. research, but the truth is that research has led to all sorts of health benefits. Take my mom, who has non-hodgkins lymphoma. There is a new drug, Rituxan, that significantly extends life and is much better tolerated than chemo. Same can be said for many other cancers.

    The article can point to hypotheses that turn out to be wrong, and yes there are many. But that is the way science is supposed to work. And it does work, as noted above.

    And, contrary to the article, medical research is dramatically underfunded in many areas, including areas like the stomach cancer I have. If we had spent just a small fraction of the Bank Bailout or War in Iraq on additional cancer research funding we would be much farther along.

  • EricG

    Sorry – was typing too fast – should have said Rituxan is a “relatively” new drug (has been used for around a dozen years). Many other examples could be given of new effective cancer treatments, like the astonishing results on CLL reported at UPenn last summer. And if you read the articles on it, yes, the causes and effects were very complicated. But it worked.

  • phil_style

    @EricG “If we had spent just a small fraction of the Bank Bailout or War in Iraq on additional cancer research funding we would be much farther along.”

    you have my 100% agreement there.

  • RJS

    I posted on this article because it highlights a philosophical argument that I have heard in a number of places in various forms. This kind of argument is in play in some (but not all) of the Intelligent Design argument. It is in play in Philosophy of Science discussions – including this book: Beyond Reduction: Philosophy of Mind and Post-Reductionist Philosophy of Science … soon to come out in paperback I believe.

    Most of these discussions though misunderstand the role reductionism plays in complex systems. Reductionism requires both understanding the pieces and how they relate to each other. As a simple example, we don’t understand water by understanding oxygen atoms and hydrogen atoms. We need to understand how the atoms interact to form a molecule and then how the molecules interact to form a liquid. Then we begin to understand water. The biological questions are the same idea, only much more complex.

  • Susan N.

    “Is science failing us?” When I think of the tremendous progress in scientific and technological knowledge that has benefited modern civilization, I am so thankful for the brilliant minds who have worked, researched, tested, discovered, and published their findings.

    In biblical times, how many people with epilepsy or mental illness were thought to be demon-possessed?

    More recently, how many died before immunizations for common diseases were widely administered?

    OTOH, is it possible, through scientific methods, to learn and know everything? I tend to believe that we should always be learning and looking to develop new or better ways of living.

    “Are causes real, or is ‘it’ mystery all the way down?”

    There are some things that defy logical explanation, imho. God? Belief in God? The way we understand Him to have interacted and continue to interact with the world (often in the most unlikely ways)? After a certain point, cause is difficult to state logically and scientifically. Beyond that point, we are left with our faith — believing in and hoping for what we can’t see. Mystery.

    Is science failing us? Insofar as science is *us*, we can only persevere based on what we know at any given time, and have courage to keep working for the betterment of our world. I think that’s our responsibility as good stewards of God’s creation.

    One could also ask, “Is religion failing us?” Insofar as religion is *us*, what have we made (or failed to make) of religion?

    We need a little faith, either way. God be with us and guide us.

  • Susan N.

    RJS (#8) – thanks for this more explicit explanation. I have felt this, intuitively and based on common sense, in some arguments I have read recently by anti-global warming proponents. The arguments were so over-simplified that even I was wary of believing that was all there was to it. Sometimes, too, the cause and effect–while not directly and immediately observable–can be witnessed over a longer time period, and in indirect ways. Scientists have to be looking at the big picture, even as they focus in on the small and specific in their research. Getting the government to fund research, and getting the public’s support, are going to be tougher and tougher to do at the present moment, I am afraid.

  • Joe Canner

    From the article: “Though scientists constantly remind themselves that mere correlation is not causation, if a correlation is clear and consistent, then they typically assume a cause has been found—that there really is some invisible association between the measurements.”

    This is actually not strictly speaking true. Perhaps in the early stages of drug testing scientist use correlation as a surrogate for causation, but a Phase III clinical trial does not rely on correlation. When there is a treatment group and a control group which are very similar to one another, any differences between the two groups are assumed to be the result of (“caused by”) the treatment, not some random association. Moreover, regression analysis can be used to easily adjust for any random differences between the groups in order to isolate the effect of the treatment

    There are also high standards for determining the cause of disease that go beyond simple statistical correlation. Smoking has been determined to cause cancer not just because smokers get cancer but because there is a lot of other physiological evidence linking tobacco smoke to the causal mechanisms for cancer.

    All that said, it is typically unethical to prove causation definitively because it would require a clinical trial where one group was assigned to the “cause” (e.g., smoking or cholesterol) and the other not. Accordingly, science goes as far as it can with disease causation and then switches to disease prevention and treatment, which can usually be ethically tested. Research done this way has been on balance quite successful, the author’s counterexamples notwithstanding.

    The author is correct about one thing: the diminishing returns of medical research. Most of the low hanging fruit have been harvested and current efforts are only making marginal progress. Absent a home run cure for some of life’s most perplexing maladies and given the current desire to reign in health care costs, there may soon come a time when the marginal cost-benefit of most medical research is considered prohibitively high.

    BTW, like phil_style in #1, this article reminded me of creationist reasoning: that if we can’t easily figure out the root cause of something we should give up looking.

  • Susan N.

    Joe, re: bioethics — did you see the recent news report that scientists had concluded some research on the H1N1 virus, namely how it is spread, but that the gov’t. had prohibited the publication of their findings on the basis of national security risk? The gov’t. fears that the information would be used by bio-terrorists.

    I wonder who will be privy to those findings? The CDC? The average doctor won’t know, presumably, and how, I wonder, will that affect ongoing prevention and treatment for H1N1 flu strain?

  • E.G.

    “Does the complexity inherent in biology undermine the search for cause and effect on which science is based?”

    No.

    1. Anecdotal evidence: We figure stuff out reasonably well.

    2. The charge of reductionism is , itself, reductionist. Most biologists, myself included, actually recognize the inherent complexity and query across multiple scales to more fully understand interacting systems.

  • http://rhymeswithplague.blogspot.com Bob Brague

    Before reading your post, I will answer your question.

    Yes, science is failing us. Science will always fail us. Science is not God. Only God will never fail us. This is not to say that science is not a worthy pursuit, as the gathering of knowledge should (and often does, but not always) lead to truth, and the truth, as Jesus told us, will set us free.

    Now I will read your post.

  • Susan N.

    Addendum to my #12 post…I looked up the article at CNN that I referenced. I misstated the flu strain: It is H5N1.

    Here’s the link:

    http://thechart.blogs.cnn.com/2011/12/21/details-of-new-lab-created-bird-flu-strain-may-be-too-dangerous-to-publish/

  • http://virtualclaritymagazine.org Phil Schomber

    RJS (#8),

    I agree that Lehrer is too quick to jetison causality. Our inability to fully understand how various pieces in certain complex systems work together to lead to a particular result doesn’t mean causation isn’t real. Neither does the fact that we might have been too quick at times to attribute causal significance to something that turns out not to be causal. But, it seems like many are questioning causation today in part because of indeterminancy at the quantum level. In your opinion, does quantum physics actually call causation into question or is there a way to understand quantum events that is consistent with our understanding of cause and effect?

  • http://krusekroncile.com Michael W. Kruse

    I take a more positive view of this article. I agree with him that causes are human not facts. “Causes” are mental abstractions that to one degree or another correspond with the interacting variables as they really behave. While we can have much higher degrees of confidence about some abstractions than others, they are all still abstractions.

    I had the privilege of hearing two MDs last summer speak about complexity in medical care. One was a researcher at John Hopkins and the other an innovative private practice physician in Montana. The researcher pointed to a number of examples of the unfolding complexity they are finding as they try to understand various medical issues. Dwarfism is one I recall him discussing. It turns out there are more than thirty scenarios (if I’m recalling correctly) with varying degrees of overlap that can result in dwarf features. Controlling and analyzing the variables is mind-boggling. What he was pushing against was the mechanistic idea that we can simply A) identify problem with alternative desired outcome, B) identify cause (as though there is an isolated “thing” called a “cause”), and C) manipulate the cause to achieve the desired outcome.

    There was a fascinating article in Stanford Social Innovation Review last year called /*Collective Impact
    */ dealing with how we address social problems. Say we want to address the high school dropout rate. Typically, various groups, based on their proclivities and abilities, develop programs, champion them as THE solution, and then compete for fundraising for their solution. But the dropout problem is not the result of one cause that can be neatly addressed by a program or two, or by a single organization. Collective Impact sees a variety of public and private initiatives functioning independently but through a coordinated process.

    I think Anthony Giddens (sociologist) is right that as the issues we confront become more complex that is less about find cause/effect/solution and more about risk management. And since risk management is directly related to values, we need not only science but ways of publicly coming to consensus about values.

    Maybe I ‘m reading too much into this article but I hear him saying something similar.

  • http://krusekroncile.com Michael W. Kruse

    I messed up the link in #17. Here it is

    Collective Impact http://www.ssireview.org/articles/entry/collective_impact/

  • http://LostCodex.com DRT

    I have several things to say about this. Thanks for this discussion RJS.

    First, I think the most serious defect in the argument is that he proposes a strawman scientist that is, essentially, an incompetent one and then argues that the incompetence leads to bad conclusions. But, competent scientists are trained to understand the interactions and eliminate bias through experimentation. The story of the Pfizer drug is not a failure of science, but a triumph. Science includes the experimentation that led to the rejection of the drug. The failure was a business failure, not a science failure. Science did what it is supposed to do, figure out what is actually happening.

    Second, many of the hypotheses in Hume’s day were much more descriptive rather than analytical. Newton’s laws for example are only a model of observed behaviors, and not causal elements. But the interaction of a Higgs Boson with a non-zero Higgs field expectation is much closer to causal.

    Third, his conclusion is right, science does indeed fail us. If you were to ask me what I want science to do, I would say that I want it to cure all disease (like the ones our friends here have, thanks for sharing), world hunger, depression, war, make everyone happy….. Science cannot do what I want it to do, but it is improving and the improvement is what I take solace in. My Mother recently had back surgery to put rods in and fuse several L vertebrae. I am nearly guaranteed to have the same issue in 20 years, so I hope science can do better than they are doing for her.

    Last, it’s a shame that this article will be interpreted by some as justification for bashing global warming or people refusing to give their children medicine, because I think it will *cause* that.

  • AHH

    Echoing what RJS and others have said, it seems like this guy goes too far. Yes, people (including scientists) can be too quick to jump to simplistic cause/effect relationships. But that doesn’t negate causation — it just means that we don’t have a good enough understanding of the complex interplay of different factors in the systems.

    It seems to be an attack on reductionism, but it attacks a caricature of reductionistic science. RJS mentions the example of water — it is right that one does not get “water” properties in a simple way from hydrogen and oxygen atoms, but the properties of water do come physically from reductionist knowledge of the underlying nuclei and electrons and so forth, without any extra physics at the higher level. Reductionism works at the physical level (at least in physics, chemistry, and biology; levels like sociology that involve free-will choices may be another matter) — the problem is at the metaphysical level, if reductionist science is presumed to be an exhaustive explanation of reality (the “nothing buttery” fallacy).

    Phil S. @16 mentions quantum indeterminacy, which indeed (under the most widely held interpretation of quantum mechanics) means that the cause/effect relationship is not 100%. [This for some reason bothers some hardline Calvinists like RC Sproul whose metaphysics has no room for chance.] But this does not really undermine causation generally for most of the things we deal with in life and in science.

  • http://krusekroncile.com Michael W. Kruse

    I see I left out a key word in #17. It should read “causes are human CONSTRUCTS not facts.”

  • Joe Canner

    Susan #12 & #15: I only heard a few minutes of discussion on this story earlier today on NPR, so I don’t don’t enough details to comment intelligently. The question I have at the moment is what the researchers were trying to accomplish by engineering a more virulent strain of H5N1. That would help determine whether these studies have any practical value to the average practitioner. According to the last paragraph of the article you linked to, the government is working on a mechanism for getting the useful information from this research to those who need to see it.

  • RJS

    Phil,

    Quantum theory refines our understanding of causality to an extent. There is intrinsic uncertainty in the timing and probability of events for quantum objects. I take the position that this uncertainty is ontological. That is, it is an intrinsic part of the nature of things. But the macroscopic probabilistic consequences don’t have the same uncertainty – and they do follow from the causes. The decay of a radioactive atom has an intrinsic uncertainty in timing – but the damage it causes can be traced back to the cause (often absorption of a high energy photon produced in the radioactive decay).

  • Susan N.

    I appreciate your response, Joe. Yes, I saw in the concluding paragraph that the gov’t is looking at a middle way to disseminate the data to those with a direct need to know, while protecting sensitive info from falling into the wrong hands.

    On another assertion that I made previously, in favor of the idea that there are always limits to our knowledge, and thus some mystery in how the world works, another related news report:

    http://gma.yahoo.com/readied-donate-organs-21-old-emerges-coma-204904805.html

    Is this a miracle? Until medical science understands the cause of this man’s remarkable recovery, it is a mystery. At the moment, the appropriate reaction is wonder and gratitude. Then, let medical scientists keep pursuing answers as to “why”. One thing is for sure, we’ll never know the answer(s) unless we ask and try to learn. We don’t just throw up our hands and say, “Whatever will be, will be. God’s will be done. We have some part in creating the world we want to live in.

  • Tim Atwater

    Thanks for this interesting discussion. As a non-scientist I have the same ((above my pay grade) caveats as first commentators, and share the general leaning that the author himself oversimplifies in arguing against reductionism. And at the risk of overstating — a hairy monster factor not much mentioned is money and power.

    Money (and power, a causalities discussion that goes back to the fall i suppose) distorts science, like it distorts theology, politics, economics, virtually everything else. No it’s not sciences’s problem per se, it’s everyone’s — but in a discussion of why science works and doesn’t work better than it does — its absence in the discussion seems strange.

    Because what drives a whole lot (probably most) scientific research is money or power. NOT primarily perhaps at the personal individual level (where idealism is often visible) — but certainly at the powers and rulers and authorities level (Wink, Stringfellow etc) — the public and corporate funding of science is driven mostly by money and power.

    for example — (I’m a pastor)– parishioners often remark how no doctor ever recommends going without drugs for cholesterol or back pain — even though some of them have suffered major complications from drugs and have self-medicated pretty well with out drugs. Friends and friends of friends within the medical establishment confirm — doctors prescribe because they get lots of perks from drug companies. Research gets done because a corporation or a university (often fronting for a corporation) wants a money product and/or prestige that (adds money value).

    I know this is an oversimplification.
    But as a Hypothesis — we choose war over peace and drugs over a simpler lifestyle in large part because war and drugs bring a higher return on investment measured in dollars.
    Hardly an original thought — CS Lewis has said it all somewhere before.

    Christmas blessings,
    Tim

  • http://scienceandtheology.wordpress.com Justin Topp

    I’m actually quite shocked that Jonah wrote this. I have seen him push against reductionism, but not science and causal reasoning. This is odd.

  • RJS

    Justin,

    I haven’t read much that he has written – but a caution against simple reductionism is certainly appropriate. I was surprised by the argument against cause as something fundamental, which is why I thought it worth posting today. I unapologetically start with the assumption that we should be looking for causes … and feel that the burden of proof is on anyone who asserts that “it is mystery all the way down.”

  • Brian Considine

    Does science fail us? No – human nature does.

  • http://practicingresurrection.wordpress.com Bill

    My objection to pharmaceutical research is that it is focused (obviously) on discovering a pharmaceutical solutions to problems that need to be addressed otherwise. Then the consuming public is persuaded by advertising and the medical industry that the pharmaceutical solution, which often does more harm than good to the human being (while enriching the corporate pusher), will solve their problem. So we get sicker and more dependent upon pills, when what we need most is a responsible diet and lifestyle.
    This is research whose goal is profit, not wellness. A sudden resurgence in healthy people in our country would be devastating to the pharmaceutical industry–at least until it was unable to persuade us all that we’re suffering from some new malady which their pills will cure.


CLOSE | X

HIDE | X