Laboratory Life: Does Science Discover Facts or Construct Them?

Laboratory Life: Does Science Discover Facts or Construct Them? May 2, 2018

Shem, you don’t understand science! (Helpful hint: just copy and paste that sentence and post it in the comments below.)

Laboratory Life

I’ve been reading Laboratory Life: The Construction of Scientific Facts by Bruno Latour and Steve Woolgar. This is a classic of science literature from a sociological perspective. In the 70s, the authors spent two years at the Jonas Salk Institute and observed the everyday workings of the laboratory. From their experiences, they created an anthropological overview of scientific endeavor that makes us see empirical inquiry with new eyes.

For anyone who idealizes science as an ultra-critical, “self-correcting” process which brings us fact-by-fact incrementally closer to Truth, this approach will produce reactions ranging from discomfort to outrage. This flatly contradicts the commonly-held definition of science as a process that discovers facts about reality and asserts that science constructs them. Hindsight distorts this process and gives the false impression that our current mode of understanding phenomena is the right one, and that the entire history of inquiry has led inexorably to a circumstance wherein we’re closer to Truth than humans have ever been. This is mythology that panders to our biases about progress and superiority, nothing more. Furthermore, it sells short the hard work done by researchers to establish facts that we consider self-evident today.

Constructing a Scientific Fact

The centerpiece of the work is the chapter titled “The Construction of a Fact: The Case of TRF(H).” Here the authors make an exhaustive (and some would say exhausting) investigation into the development of our understanding of the hormone TRF(H), produced by the hypothalamus. Using the contemporaneous testimony as well as reminiscences of the experts involved, analysis of the relevant published papers and statistics concerning citation patterns from that literature, and the pronouncements of professional institutions, Latour and Woolgar painstakingly piece together a “literary history” of the fact of TRF(H).

The authors demonstrate the amount of contingent variables involved in the “discovery” of TRF(H), from industry rivalries to international disputes. Up until the 1960s, the existence of this hormone was in the hypothetical stage. This was an era when endocrinology had only recently become established as a specialty, and neuroendocrinology was in its infancy; the way the brain regulated the endocrine system was not fully understood. Even ideology played a part in the saga: research in Prague was suspended due to official opposition to the theory that the hypothalamus regulated the endocrine system via the pituitary.

Researchers competing to clarify the hormone’s regulatory activity weren’t sure whether the substance was a peptide or some other unconfirmed type of factor, or even if the observed effects of the substance were due to another factor such as oxytocin. The trouble with procuring enough hypothalamus matter and sufficiently pure samples to test was a major roadblock in research. Papers published on the subject were only unanimous in the vagueness and equivocation of their conclusions, and researchers all cited one another’s work to try to support a more emphatic stance that never seemed justified by the data. By the end of the 60s, the exorbitant costs of research —as well as suspicion in the industry about its feasibility that motivated a review committee at the NIH to investigate the competence of the practitioners— left only two research groups in the running. The objective itself had also narrowed down, and the goal became determining the molecular structure of the purported peptide; the tension between the physiologists and the chemists had resolved in favor of the chemists. Finally a sample was made volatile enough to be analyzed in a mass spectrometer, and the result was a conclusive description of the amino acid sequence that defined the structure of the hormone. For their efforts, the leaders of the two labs shared the 1977 Nobel Prize in Medicine.

Persuasion and Magical Thinking

The problem, then, isn’t with the facts themselves. It’s with the belief that they exist independent of human endeavor, that they don’t depend on how we define phenomena and act in the world. The authors deliberately approach the lab researchers as members of a foreign tribe whose values and perspective aren’t necessarily our own:

[O]ur observer has begun to make sense of the laboratory in terms of a tribe of readers and writers who spend two-thirds of their time working with large inscription devices. They appear to have developed considerable skills in setting up devices which can pin down elusive figures, traces, or inscriptions in their craftwork, and in the art of persuasion. The latter skill enables them to convince others that what they do is important, what they say is true, and that their proposals are worth funding. They are so skillful, indeed, that they manage to convince others not that they are being convinced but that they are simply following a consistent line of interpretation of available evidence. Others are persuaded that they are not persuaded, that no mediations intercede between what is said and the truth. They are so persuasive, in fact, that within the confines of their laboratory it is possible to forget the material dimensions of the laboratory, the bench work, and the influence of the past, and to focus only on the “facts” which are being pointed out. Not surprisingly, our anthropological observer experienced some dis-ease in handling such a tribe. Whereas other tribes believe in gods or complicated mythologies, the members of this tribe insist that their activity is in no way to be associated with beliefs, a culture, or a mythology. Instead, they claim to be concerned only with “hard facts.”

I think Latour and Woolgar have done something extraordinary here. They’ve humanized the process of scientific inquiry, and demythologized the notion of scientific “facts.” We can no longer look at what we know about natural phenomena except in terms of the work that was involved —and the tools that people used— in creating it.

Sam Harris closed his atheist manifesto with the words, Nothing is as sacred as the facts. But we don’t need to fetishize facts. We need to acknowledge that facts are no more sacred than any other human creation.

What do you think? Are scientific facts sitting out there waiting for us to discover them? Or is our knowledge of the world contingent upon the means we use to create it?

"Per an update on my blog post which includes the linked Venn diagram, while Kuhn, ..."

Redefining Science With Thomas Kuhn
"Kuhn gets hung up on "established paradigms".The Big Bang didn't really re-write a paradigm....... unless ..."

Redefining Science With Thomas Kuhn
"His description of the social-institutional context of scientific inquiry makes the idea of "self-correcting" science ..."

Redefining Science With Thomas Kuhn
"I'm not sure "constantly testing their theories" and "validating their theories" are mutually exclusive. There ..."

Redefining Science With Thomas Kuhn

Browse Our Archives

Follow Us!

What Are Your Thoughts?leave a comment
  • Anri

    I could, I suppose, open the top-floor window in a very tall building, step out of it, and write a post-modern paper describing how the process of quantifying gravity was so steeped in cultural bias and deference to arbitrary authority that it calls the results into very severe question for those with the right mindset.

    But I should probably write quickly, and on something durable.

  • I’m not sure whether this article deserves the knee-jerk pomo-bashing. It’s not saying there’s anything wrong with science. It’s simply pointing out that we tend to think “facts” are something separate from human activity, when in fact they’re the product of it.

    If anything, Latour and Woolgar are reminding us of the labwork, argumentation and brainpower that went into constructing what’s now accepted as a “hard fact.”

  • Illithid

    If I believe in an objective universe rather than being a solipsist, I have to think that facts exist independently of our conceptions of the world. We may make mistakes in describing facts, our process may be flawed, and the investigators may be biased. But the model of reality that we have will either reliably predict results (to whatever level of accuracy we find acceptable) or it will not. If not, it must be altered. The “fact” that we never can be absolutely sure we have correctly described reality doesn’t mean that it isn’t real.

  • Where in this article does it dispute that there’s an objective universe, pray tell? Where did I claim that “reality isn’t real”?

    What the authors are trying to demonstrate is that “facts” aren’t just sitting there waiting for us to discover them, like snow-capped mountains in equatorial Africa waiting for European mountaineers to climb them. The chaos of information available to us in the world doesn’t just produce facts. Data points don’t magically interpret and arrange themselves into meaningful narratives. People have to do hard work to create some semblance of order from disorder.

  • epeeist

    This is the Bruno Latour who said that Ramases II could not have died from tuberculosis because the bacillus wasn’t discovered until 1882 by Robert Koch. The person who said that “Before Koch the bacillus had no real existence”.

  • Anthrotheist

    I would tend to agree, that “facts” are human constructs. Even the notion of what constitutes a “fact” is necessarily a human invention (as is all language; emergent though it may be, it is still purely a human creation).

    The “fact” is, there are phenomena that occur in the universe in which we live. The “fact” is, instances of these phenomena are often very similar to one another. This makes the phenomena appear to be following something we would like to think of as “rules” that govern their behavior. While this process of observing phenomena and formulating predictive theories about their behavior (many of which are actually a statistical average and not a consistently exact measurement) is a genuinely valuable and generally beneficial endeavor, it still relies on fundamental expectations (such as the universe essentially being a giant clockwork model, which is handy but not necessary).

  • epeeist

    It’s simply pointing out that we tend to think “facts” are something separate from human activity, when in fact they’re the product of it.

    You are saying that everything is a social construct?

  • Illithid

    Well, then, I don’t think they’ve demonstrated that. Perhaps we’re using the word “fact” in different ways.

  • No. Latour and Woolgar describe the way the fact of TRF(H) was constructed. There were data points and observations at one level, but for this fact to become known, there had to be a lot of argumentation, lobbying, papers published and cited, and even an investigation from the NIH.

    Just because something is culturally constructed doesn’t mean it’s made up out of thin air.

  • I’m not an expert on Latour’s thought or career. But I assume he was talking (with typical pomo insouciance) about the difference between a disease itself and our cultural understanding of it. In Laboratory Life, he makes the point that the Salk Institute researchers would routinely refer to the existence of a substance or factor only after it had been detected and confirmed; prior to that, they’d refer to a phenomenon or effect. In essence, the substance itself didn’t exist for these exacting scientists until the point of confirmation.

  • Really? You think data points just magically arrange and interpret themselves into constructs we can comprehend?

  • epeeist

    But I assume he was talking (with typical pomo insouciance) about the difference between a disease itself and our cultural understanding of it. In Laboratory Life

    And in doing so makes a standard map/territory error. Our knowledge of the existence of the bacillus may not have come into being until the time of Robert Koch but this is different to the bacillus existing.

  • epeeist

    You think data points just magically arrange and interpret themselves into constructs we can comprehend?

    No, of course not. But there are numerous heuristics that scientists will follow when trying to construct an explanation for a particular set of “data points”. You might want to try something like Fr. Ernan McMullan’s The Virtues of a Good Theory as a counter to some of the ideas that the likes of Latour put forward. Other authors you might want to consider are Huth’s A House Built on Sand: Exposing Postmodernist Myths About Science. This is particularly good in exposing Latour’s complete misunderstanding of the theory of relativity. Paul Boghossian’s Fear of Knowledge and Alan Sokal’s Beyond the Hoad are also worth reading.

  • Illithid

    *sigh* No, I do not. Per my previous comment, is a “fact” an aspect of reality, or our concept of that aspect? If the former, it exists independently of our knowledge. If the latter, then yes, we must work to construct it, and our construct is perhaps inevitably imperfect.

    I generally use the word “fact” in the first way. If you want to use it to designate an idea we have about reality, what word should we use to refer to the reality itself? Or should we even have such a word, since we cannot know such things with absolute certainty?

  • chemical

    Re: Constructing facts vs. discovering facts. I think we can all agree that there is an objective reality that has properties that are describable. The work that scientists do is describe those properties. They translate and communicate these properties in a way other humans can understand. After all, reality is not obliged to conform to the languages we create; rather we modify our languages to better describe reality.

  • epeeist

    So are there facts on display at this conference?

  • Swing and a miss, amigo. I’ve never posted anything that supports conspiracism, anti-vaxx nonsense, or flat-Earth numbnuttery.

    I apologize if even a little healthy skepticism about science is too much for you to bear.

  • The idea that we impose order on a chaotic universe is one that still frightens people. They’d rather hear that science is “unlocking the mysteries of the universe” or something. They want science to tell them that the way we understand things is fundamentally correct, unlike our benighted ancestors who thought the way they understood things was fundamentally correct.

    I’m always the one being told I don’t have enough respect for science, but at least I think science isn’t here to pander to my prejudices.

  • Thanks for the reading assignments. I still don’t see how this article is somehow emblematic of a fear of knowledge, or an outrageous affront to people who conduct scientific inquiry. It’s just saying that empirical inquiry is a communal endeavor with its own beliefs, norms and codes. The activity of science can be studied just as sociologists study any other type of group behavior.

  • epeeist

    The question is why are the “social constructs” of organisations like the Flat Earth society different to the “social constructs” of, say, general relativity?

  • epeeist

    It’s just saying that empirical inquiry is a communal endeavor with its own beliefs, norms and codes.

    The fact that science is conducted in a social milieu isn’t under dispute. What Latour and other sociologists of science (try the “Strong Programme” for example) were about as much as anything was epistemic penis envy. The aim was to either dethrone the epistemic privilege that they saw science had or to try and gain the same privilege for sociology.

  • Once again, I don’t know why it’s incumbent on me to defend flat-Earthism. I don’t dispute any mainstream scientific theory: Big Bang, unguided species evolution, the efficacy and safety of vaccines, anthropogenic global warming, etc.

    As far as I can tell, general relativity theory was produced through responsible collaborative effort, has a high degree of experimental verification, and is supported by the majority of relevant experts. The flat Earth idea is an incoherent pile of factoids, and an online hoax touted by message-board trolls.

  • Like anything else, I suppose it depends on your perspective. I see a lot of people who use the word science like it’s synonymous with reality, and can’t abide even a modicum of skepticism about the subject.

    If you’re not one of those people, what’s the problem?

  • epeeist

    I would call myself a “scientific realist” based as much as anything on Hilary Putnam’s “No miracles” argument.

  • Fine by me. But present this Putnam quote to the folks at the Where’s-Your-Evidence blog where you post so much and tell me how well it goes over:

    If objects are, at least when you get small enough, or large enough, or theoretical enough, theory-dependent, then the whole idea of truth’s being defined or explained in terms of a “correspondence” between items in a language and items in a fixed theory-independent reality has to be given up. The picture I propose instead is not the picture of Kant’s transcendental idealism, but it is certainly related to it. It is the picture that truth comes to no more than idealized rational acceptability.

    We’re great at stress-testing other people’s closely-held beliefs, but when’s the last time the correspondence theory of truth got a workout in the atheist blogosphere?

  • I’d never say something as silly as there’s no reality. only that reality is so staggeringly complex that we struggle to impose order on it in a lot of different ways. Scientific inquiry, by limiting itself to empirical factors, has been extremely successful in generating useful information that can be arranged into comprehensible narratives. But we forget how much cultural and personal input we have even in a process we consider so algorithmic and objective.

  • Anri

    I was answering the question posed at the end of the article.

    People wondering if gravity is an independent fact or a product of human understanding could unknowingly fall from a height and discover if their ignorance of gravitational effect prevents it from establishing itself on them.
    More peacefully, they could determine if gravity acted on anything prior to human understanding of it. Or awareness of it. Or, indeed, prior to humanity.

  • It may be great fun to pretend that I’m claiming that the gravitational pull of planet Earth itself is merely something that humans created through language, but it’s way off the mark. What I’m talking about is the way our understanding of natural phenomena derives not only from the data points that result from the study of these phenomena themselves but also the means we use to study them, the process of interpreting and arranging the data points, and the details of the social and institutional contexts in which the study of the phenomenon takes place.

    We don’t have unmediated access to reality, unfortunately.

  • Anri

    We don’t have unmediated access to reality, unfortunately.

    Are you aware of any scientist who claims we do?

    This seems to be an entirely semantic “puzzle”: if we define facts as things only as we understand them, than thing we don’t understand aren’t facts.
    If we define facts to be things we discover about the universe, then there are facts we don’t yet understand.

  • Latour’s point was more that a person can’t have died of tuberculosis before the bacillus had been characterized, because that medicalized name is derived from a description of the bacillus. On the other hand, a person could easily have died of consumption. Talking about Ramses dying of tuberculosis is like referring to a map which has places labeled by only the names they’ll have in the future, like out of a Laurie Anderson song.

  • epeeist

    Latour’s point was more that a person can’t have died of tuberculosis before the bacillus had been characterized,

    Latour goes on to say, “Before Koch, the bacillus had no real existence”. This would seem to undermine the point you are making.

  • Because he is contrasting this understanding with prior etiological speculation; the concept of microscopic cells eating the lungs would have not entered the mind of a person unaware of bacteria, probably hewing instead to some hypothesis involving unbalanced humors or miasmatic exhalations. As far as medicine was concerned, bacteria were still part of the unknown, undifferentiated, unnamed piece of the universal background for all phenomena called “nature”, and had no distinct existence relevant to an observer.

  • Latour seems to push a lot of your buttons, doesn’t he? Considering how dedicated science fans are to getting to the truth through ceaseless inquiry, you seem affronted whenever someone subjects science itself (as well as how we as a society conceptualize it) to rational analysis or approaches it from a sociological or philosophical perspective. It seems you’d rather we just accept what we’ve been told about science’s total objectivity and magically self-correcting nature and not think too hard about it.

  • epeeist

    Because he is contrasting this understanding with prior etiological speculation

    But this is an argument about the map, which I have no argument with. I have no difficulties with admitting that our understanding is wrong or incomplete. But what Latour seems to be saying in my second quotation is that at the time of Ramases II and until the time of Robert Koch the bacillus had no existence. In other words he is making a claim about the territory rather than the map.

  • The one here confusing the map with the territory seems to be you, because you’re making it sound like the provisionally useful way we have of conceptualizing illness isn’t just another map, it’s the territory itself. It may irk people to be reminded that we don’t have unmediated access to reality, but that’s the way it is.

  • epeeist

    Considering how dedicated science fans are to getting to the truth through ceaseless inquiry

    I’m sorry, where have I said that science is in pursuit of truth?

    you seem affronted whenever someone subjects science itself (as well as how we as a society conceptualize it) to rational analysis or approaches it from a sociological or philosophical perspective.

    If you read my profile you will see that one of my interests is the philosophy of science, so I have no problems on this front.

    As for the sociology of science, it really depends whether the endeavour is undertaken dispassionately or not. As I have said above, I have no problem acknowledging the fact that science takes place in a social milieu by people who have all the usual biases and fallibilities (I speak from experience, having worked in academia).

  • I wasn’t talking just about you. Tell you what, you count how many times you hear science defined as an objective, self-correcting device that eliminates bias and brings us incrementally closer to truth, and I’ll count how many times I hear science defined as a human endeavor and a historical construct that deserves our skepticism as much as any other human creation. Wanna bet whose bucket fills up first?

    I was amused that you didn’t even engage with the OP, you just reacted to the mere mention of Latour’s name with a glossary of triumphalist literature for me to read. As I discussed in my most recent post here, it seems like science fans just kept their fingers in their ears while postmodernism was in vogue, and now they just assume anyone who talks about the relationship between knowledge and power is some sort of crackpot.

  • I contend that Neptune existed long before we found it, and so its existence was a fact long before we knew it to be so.

    The problem is that’s like saying, I contend that Istanbul existed long before people named it Istanbul. The ways we conceptualize the solar system, arrange and interpret celestial observations, and define things like planets, construct the facts about Neptune. Reality only reveals itself to us through the means we use to study it.

  • William Meyer (JrSage)

    Ah! But Istanbul (as any fan of They Might Be Giants will tell you) DID exist long before the Turks named it Istanbul, and even then Europeans insisted on calling it Constantinople for a long time. If you asked someone in the Ottoman Empire if Constantinople existed they would probably say no, even if they were living there according to European maps.

    Bottom line, something can exist if it has no names or many, and Neptune existed before it was theorized, found, and named. If the fact of Neptune’s existence was a human construct, we would have found Vulcan as well.

  • The point is that Istanbul is more than just a label for a set of co-ordinates. It’s a whole bunch of socially constructed networks of laws, boundaries, and institutions, populated by various evolving networks of people. It’s problematic enough to try to provide a true-and-complete description of Istanbul, let alone just blithely say it always existed but people just didn’t call it the right name yet.

  • William Meyer (JrSage)

    I agree! Istanbul is also, quite literally, a human construct; humans built it, they didn’t discover it. Neptune, on the other hand, is not. All the evidence we have so far suggests it formed long before the first human being was born, and it was out there well before any astronomer went looking for it.

  • I hope you don’t think I dispute that Neptune exists.

    What I’m trying to point out, to no apparent avail whatsoever, is that what we know about Neptune (the “facts”) aren’t theory-independent and self-evident to all; they’re conclusions based on the most coherent and well-accepted interpretation of the body of data amassed to date concerning our solar system.

  • William Meyer (JrSage)

    It could be a matter of semantics, then; what scientists commonly call facts are what you might call Truth. To use the Plato’s Allegory of the Cave, the facts are the objects casting shadows, the light behind the objects is the data, and the prisoners theorize what the facts might be using what the data shows them. Brighter light (more data) paints a sharper picture on the wall for more accurate theories, but the objects casting the shadows (facts) are the same as they ever were. If the facts didn’t “exist independent of human endeavor” or depended “on how we define phenomena and act in the world” there wouldn’t be any shadows to theorize about in the first place.

    What are “facts” in practical terms are theories by scientific standards, which ARE constructed and falsifiable, so they should definitely not be fetishized. However, the pursuit of “hard facts” (what some call Truth) is what gives scientific endeavor direction and keeps it grounded in a shared reality. Remove the concept of “theory-independent and self-evident” facts that can potentially be discovered by anyone and there’s no longer any reason to put any theories to the test; you can just proclaim “facts” from on high and call it a day, like ancient priests in clean white coats.

  • Antoon Pardon

    I disagree. IMO this seems a confusion between facts and our understanding of the facts. The labwork argumentation and brainpower constructed our understanding of TRF. They didn’t construct a (hard) fact itself.