Archive for May 23rd, 2011

Courtesy of Brandon Doran CC-BY-NC-ND 2.0

I really kicked the hornets’ nest with my last post, and library constructivists are going out of their way to defend what I believe to be an indefensible position. I should point out that on the constructivist view my arguments for realism are an equally valid viewpoint, so why the fuss? I should also point out that I find it odd that, when pressed, the most common constructivist retorts I’m seeing are: “But, look at all the oppression in the name of science!” Should I even dignify such naïveté with a response? Of course there is oppression in the name of science, but that has nothing to do with science. Really, if you can name it, someone has probably been oppressed by it, and that includes social constructionism. Completely abandoning rationality and objectivity because bad things have happened in the name of rationality and objectivity is throwing out the baby with the bathwater. Tell you what, here’s a bit of Latin you can get as your next bitchin’ tattoo: cum hoc, ergo propter hoc. Better yet, tattoo it on the baby, so you don’t forget. (Bonus if you add “4 Lyfe” after it.) 

“Pokemon 4 Lyfe”
Bob Jagendorf. CC BY 2.0 

Anyway, two things have come out in the ensuing debate that are commonly confused. On the one hand there is the question, “What is the world really like?” Is it a social construct or is it a mind-independent reality? On the other hand, there is the question, “What can we know about the world?” Are claims to knowledge socially constructed, or are they objective? Though these are separate philosophical issues, and I tried to distinguish between the two when I covered constructionist theories, I’ve been asked to explain my views on each. First, I’ll address the issue of what the world is really like; the metaphysical realist vs. anti-realist debate.

NKESR stands for “Use Common Sense, Dummy”
Why am I a metaphysical realist to begin with? I suppose I could give a personal history of the courses I took and the lectures I’ve given, but you kind of had to be there. Instead I’ll take an old lecture I gave in a philosophical problems class and blog-o-tize it for mass consumption. (Skip it if you really don’t want to read about Kant.)

Start with Kant, who held that things that exist independently of our minds cannot be known “in themselves”. These objects (noumena he called them) are only experienced through representations in the mind. In fact, according to Kant, all we have direct access to are our mental representations of external objects (he called these phenomena). Some philosophers ran with this idealism and created theories to the effect that we can have no knowledge of the world outside of ourselves and our own minds. But, Kant had a trick up his sleeve. Our mental representations are remarkably well structured. Things like causality, extension in space or the fact that things persist through time are inescapable constraints on how we see the world (he called these schema). Where do these constraints come from? His answer was that they come from the external objects themselves. So, we may not have access to the physical objects in themselves but we do have access to the causal and relational properties between them.

Filtered through a few centuries, Kant’s theory leads to Kantian Epistemic Structural Realism. This is the view that we cannot know anything about physical objects in themselves, but we can know their causal and relational properties. For example, I may not be able to see atoms, but I can describe the way they interact, make predictions, and use atoms to investigate other aspects of nature. So, scientific theories are just accounts of the causal and relational properties of objects, and not of the objects themselves. Yet, this is still a realism because it acknowledges that certain relationships do in fact obtain in some external reality, even if the objects themselves cannot be directly experienced.

Now, for one last twist. Scientific experimentation is remarkably consistent in its results. Actually, the successes of science are the best evidence that reality is not a social construct. In fact, as Putnam put it, realism is the only attitude that doesn’t make the successes of science miraculous. (See this article for a brief overview). In reference to Kantian structural realism, I think that the remarkable consistency and predictive power of our scientific theories is the best evidence that the objects they describe really do exist. The causal and relational properties described by our theories are so powerful, that the objects in those relations must be real. So, call this neo-Kantian Epistemic Structural Realism (NKESR). That’s what I believe. There are some good criticisms of this position, especially Arthur Fine’s Natural Ontological Attitutde and Bas van Fraassen’s constructive empiricism. These are criticisms that I take seriously, though answering them exceeds the scope of this post.

(Update 5/25/11: No sooner did I post this than PNAS published an article somewhat corroborating Kantian psychology: apparently intuitions about Euclidean geometry are innate and independent of social constructs…including language. Neat!)

So, the simple version is just that whatever reality is actually like, socially constructed or not, if you jump out the window, you’re going to fall. It just so happens that the best account for why this is the case is that there really is an external, physical world. To quote a contemporary realist philosopher, “What they eat don’t make us shit…Real talk.”

“Girl, concepts are just freely repeatable elements
in propositional representational contents. Real talk.”

So, what is knowledge?
Everything I just said was about metaphysical realism (ontological realism). How we should describe reality is a separate, epistemological issue. The question becomes, “What can we know about the world?” I should take a second to say what knowledge is, in order to set up epistemological realism. Knowledge is non-accidentally true belief. To say that you know that the head office of the American Library Association is in Chicago means that

  1. You believe the ALA head office is in Chicago,
  2. It is true that the ALA head office is in Chicago, and
  3. You are justified in believing that the head office of the ALA is in Chicago.
What are truth and justification? Well, in my last post I explained constructivist theories of truth (fact), and those don’t work, so let’s go with what works and adopt a correspondence theory: a proposition is true if it corresponds with reality. Constructivists should have no problem accepting this theory, for them reality is a social construct so truth would just be what corresponds to social agreements. But, as I showed earlier, reality is objective and external, so truth corresponds with that. Now, what about justification? That’s actually a HUGE thing to answer. 

You see, we can accept that there is a mind-independent physical reality, but that doesn’t tell us what we are justified in believing about that reality (inversely, we can be realists about our theories, but not about physical objects. cf. instrumentalism). Epistemic social constructionism tells us that all beliefs about reality are equally valid social constructs and that though there may be genuine facts out there, there is no privileged way of representing the world. Boghossian (2006) describes it this way:

Of course, the world doesn’t just inscribe itself onto our minds. In trying to get at the truth, what we do is try to figure out what’s true from the evidence available to us: we try to form the belief that it would be most rational to have, given the evidence. But is there just one way of forming rational beliefs in response to the evidence? Are facts about justification universal or might they vary from community to community? (p. 58)

Most of the more persuasive constructionists follow this type of epistemic constructionism; Richard Rorty is a prime example of someone who follows this line. According to Rorty, facts aren’t socially constructed and relative at all. The same facts hold for everyone, everywhere. This is a metaphysical realism. BUT, Rorty hastens to add that there is not a privileged way of representing the world. We can create wildly different epistemic systems based on the same evidence, and all of these systems are equally valid. So, it isn’t that facts are relative, it’s that facts about rational belief are relative. For the epistemic constructivist, knowledge is still justified, true belief, but justification is relative to a particular community. As Boghossian puts it, “different people may rationally arrive at opposed conclusions, even as they acknowledge all the same data” (p. 59)

On the flip-side, epistemological realism tells us that there is, in fact, a right way and a wrong way to represent the world and we can accurately represent the world as it really is. So, what is the right path to knowledge? I for one follow a reliablist account of justification: a belief is justified if it is the outcome of a reliable, truth-conducive process. Since the scientific methodology is the most reliable, truth-conducive process, the scientific methodology gives us a true account of reality. And just to be clear, a scientific methodology is not the same as the hard sciences. The scientific methodology is just the application of logic, reason, empirical evidence, experimentation, etc. to questions of fact. So, for example, on the question of which is better, the Dewey Decimal System or the Library of Congress Classification system, we use logic, reason, and analysis to examine each carefully. The context, the social agreements, the ingrained biases…these are still relevant and important factors that we must consider as evidence or “experimental controls”. When I want to figure out which to adopt in my library, there is a fact of the matter that one is better than the other in my situation, and I can figure it out through reason. If I were a constructivist about knowledge, I may appeal to tradition, to holy scripture, to what feels right, or some other method of justification, since none are superior. As a realist, I appeal to reason and evidence. Since I want to be done with the philosophy part of this post, I’ll just direct you to the Stanford Encyclopedia of Philosophy if you need more, because I’m supposed to be talking about libraries.

Courtesy of nualabugeye. CC BY-NC-SA 2.0

Libraries and Objectivity

Okay, back to libraries. What am I advocating for the foundations of library and information science? Well, I think LIS should be committed to the following things:

  1. Metaphysical realism: There are mind-independent facts about the way the world really is.
  2. Epistemological realism: We can, in fact, accurately represent the way the world really is.
  3. Reliablism: We are justified in believing the truth of our representations if those representations are the product of a reliable, truth-conducive process.
  4. Scientism: Scientific methodology (reason and evidence) is the most reliable, truth-conducive justificatory process.
The best approach to a philosophy of librarianship begins with embracing reason and truth, understanding that both are grounded in an objective reality. Our metaphysical commitments and epistemic commitments are united insofar as our knowledge claims are about real things and can be evaluated in terms of an objective reality. Let me quickly cut a misconception off at the pass: social interaction, culture, bias, and all the other weapons of constructivism are still relevant in an important way. There are biases in how we evaluate information. There are cultural influences on what we believe. There are social processes that inform us. But all of this is at a different level from actual knowledge. I have to keep coming back to the very important and often overlooked distinction between what we think about the way the world is, and how the world really is. A great deal of what we think about the world is, in fact, socially constructed. But, the way the world really is is independent of what we think. Since the demythologizing of the ancient Milesian philosophers, the concepts of objectivity, reason, rationality, and truth…in a word knowledge…have been developed to cut through our socially constructed biases and lift the veil on reality.
Demolition science
Librarians who want to treat all knowledge and truth as socially constructed are adamant that we need to get in on the construction business and assist our patrons in creating their realities. I’ve already said what I think about that. My proposal is that libraries enter the demolition business instead. We need to use the tools of reason and objectivity to tear down cultural biases, falsehoods, and misconceptions. We need to provide society with the tools to stand up to misinformation, disinformation, and deception. We need to blast a big-ass hole in the wall and let our patrons become educated and enlightened so they can stand up to whatever society throws at them. It’s that whole speaking truth to power, truth-shall-set-you-free thing that guided us through the liberalism of the 1960s. Let me give some brief examples:

Collection Development
Don’t listen to postmodernists when they tell you that realism will force libraries to only collect what the librarians have deemed to be acceptable knowledge. Far from it. Libraries should provide access to as many varied viewpoints and perspectives as possible, because these perspectives are the evidence we use to determine what is true and what is false. My library makes available books and articles in favor of intelligent design and in favor of evolution. A constructivist would say that the library collects these contradictory positions because both are equally valid accounts of the world. A realist would say that only one theory is correct, but in order to determine the truth, you have to consult all of the evidence. So, as realist librarians, we are committed to making available the widest possible selection of viewpoints because each is another bit of evidence for or against the way the world really is.

Don’t listen to postmodernists when they tell you that realism will turn reference librarians into authoritarians who pass judgment on what counts as acceptable knowledge. First, our epistemic commitments are distinct from our moral commitments, and reference librarians have a moral obligation not to assume how or why patrons will use the information they seek. All we can assume is that information seekers often do have the goal of acquiring true beliefs (Fallis, 2004). By directing patrons to the viewpoints most relevant to their information need, we can be assured that we are providing them with the relevant evidence. Second, we have to avoid focusing on information wants and instead focus on information needs. When a patron asks for information, we should avoid simply handing over all and only the materials that meet that specific request. Realism entails that information exists independently of what we believe, so the realist librarian focuses on identifying a patron need and satisfying that need with the best resources available. Obviously, what the patron wants will usually be a part of the information need, but other viewpoints may be relevant, and it is the reference librarian’s job to acknowledge that they exist.

For example, when a student asks for books and articles that prove that homosexuality is immoral, I won’t take the constructivist approach and hand over only what will confirm the student’s preconceptions. I’ll take the realist approach and direct the student to books and articles about homosexuality and morality, which include competing theories, and I’ll let the student use that evidence as he sees fit. Hopefully, the student will critically examine all of the relevant arguments and come to understand the way the world really is (Which is that homosexuality is not immoral. I am inflexible on this, so constructivists, don’t even try to argue the point.) I should add that this applies to all information needs, regardless of what the reference librarian believes. If a patron wants to show that homosexuality is immoral, he needs to look at all the evidence; if a patron wants to show that homosexuality is not immoral, she needs to look at all the evidence. Put another way, the scientific approach is to critically examine all the relevant evidence in order to form true beliefs.

Don’t listen to postmodernists when they tall you that realism will lead to artificial hierarchies and divisions in how we organize information. They routinely criticize the supposed rigidity of the “ontologies”  librarians impose on information and offer the “wisdom of the crowds” as non-discriminatory and emancipatory alternatives. The Semantic Web, folksonomies, even Wikipedia as an authority control…these are purported to be the way of the future. In contrast, the realist cataloger knows that these crowd-sourced options are incredibly powerful tools, but she recognizes that information has a life outside of what we believe.

Library Instruction
Don’t listen to postmodernists when they tell you that realism will force library instructors to adopt an authoritarian stance and force students to adopt one and only one predetermined method of research. Realism in instruction begins with the acknowledgement that research skills are not monolithic; there are several methods of access and evaluation. However, there are better and worse ways of accessing and evaluating information. These evaluative tactics have to be themselves evaluated, and this is only possible if research is directed towards something fixed, such as information that exists independently of what we believe. One of the most common forms of research you’ll see among college students is research ex post facto. Students decide on a thesis, write their arguments, and then look for articles that support their position. This method of research becomes a means of corroboration, rather than a means of discovery. As I showed in my last post, this method of “research” is consistent with constructivism…perhaps even preferred by constructivism. If social agreements are the path to knowledge creation, then it only makes sense to seek out agreement and adopt it as justification. But, from a realist stance, this simply will not do. The scientific approach begins by abandoning preconceptions and letting the available evidence guide us. Students should collect information first, synthesize it, and let it determine whether the thesis is or is not supported. The scientific method is to adopt a hypothesis, test it, and either confirm it or change it in light of the evidence. This is the realist approach to research and information literacy.

I should add that, over the past week, I have repeatedly seen critics and supporters of realism mistakenly confuse constructivist epistemology and social constructionism with constructivist and constructionist learning theories. Mixing and matching these is a flat-out category mistake, but it occurs with alarming regularity. For the record, I am only critical of the epistemological theories. Philosophical realism is entirely consistent with constructivist teaching techniques (as Socrates no doubt would have lead you to conclude.)

“Damn, Crito, bust off my sizzurp!”

This ain’t a scholarly paper
This is just a cursory explanation of a realist foundation for librarianship; there’s a lot more to be said, for sure.
I’ll probably wind up posting more on realism in the future, but for now I’d like to end this long post with one final observation that was just pointed out an hour ago in the comments to the last post. As Paul H. points out,

Even a thoroughgoing realist can admit that language is a social construct. A realist might say that we construct a language to enable us to talk about objective reality; the fact that the language is socially constructed doesn’t affect objective reality. For example, French people have a different set of agreements for their language than English speakers—we can all admit that—but the French and English speakers are still talking about the same reality, we can say they agree with each other or disagree, etc.

Paul is absolutely right. Realism does not deny that there are social constructs. Languages, libraries, the internet, governments…these and more are excellent examples of social constructs. All realism posits is that these socially constructs are aimed at something real and objective, and they can be evaluated accordingly.  So, I’m not denying the weak sense of social construction. I’m not denying that social factors influence us all the time. This weaker sense of constructivism is uncontroversial, uninteresting, and it doesn’t add anything valuable to the conversation about the theoretical foundations of librarianship. If anything, I’m discussing the stronger form of constructionism in order to avoid creating a straw-man argument.

So, If you disagree with me, I say “Awesome!” Let me know how and why. I’m not a terribly good philosopher, so I may have jumbled things up a bit. I’m also pretty green as a librarian, so let me know what aspects of library science I need to work on (I know cataloging is one area). This blog is my source for peer-review, where I can bang out an idea and see if it sticks well enough to write a more formal treatment, so feel free to chime in.

Reading list

  • Boghossian, P. A. (2006). Fear of knowledge: Against relativism and constructivism. Oxford: Clarendon Press
  • Burge, T. (2010). Origins of objectivity. Oxford: Oxford University Press
  • Fallis, D. (2004). On verifying the accuracy of information: Philosophical perspectives. Library Trends, 52(3), 463-487. [link]
  • Fine, A. (1984). The Natural Ontological Attitude. In Leplin, J. (ed.). Scientific Realism (pp. 83-107). Berkeley: University of California Press.
  • Goldman, A. I. (1999). Knowledge in a social world. New York: Oxford.
  • van Fraassen, B. (1976). To save the phenomena. Journal of Philosophy, 73(18), 623-32. [link]

Read Full Post »