Tuesday, July 12, 2011

Objectivity

A colleague recently sent out a link to the Postmodernism Dictionary, which has an entry for the term Objective:
Being objective means to have no bias or distortions; to see things [as] they actually are. It assumes the individual is able to bracket their subjective perspective, biases, and prejudices. Postmodernism, in general, questions the degree to which we can obtain objectivity.
 This is not a good scientific definition of objective; it is, rather, a straw argument, set up as a convenient wall against which to play intellectual ping pong.  Lest I be accused of setting up my own straw postmodernism, let me call your attention to Schultz and Lavenda's textbook, Cultural Anthropology: A Perspective on the Human Condition (Oxford 2012).  This book is written largely from the perspective of non-scientific, postmodern, and interpretivist anthropology.  On page 44, they define objective knowledge as:
Knowledge about reality that is absolute and true.
This is no better.  It reduces the notion of objectivity to a cartoon of itself.  But before I offer something more, er, realistic, let me explain why I am so incensed by these kinds of definitions.

I take anthropology to be a social science.  This is relatively uncontroversial; here at UNF, we're even located in the Social Sciences building (but then, so is the Dean's office!). It is true that we often say that anthropology as a discipline overlaps with the "sciences" and the "humanities" (history, philosophy, world languages, literature, etc.), as if these were normally non-overlapping magisteria, to borrow from Stephen Jay Gould. Scientific method applies here, but not over there.  (I disagree with this divide, I think it's an artifact of a particular cultural history, but that might be another post.)
In any case, the business of the sciences is to develop what I am going to call, after Lett, objective synthetic propositional knowledge. A synthetic proposition is one that's not simply an identity. For example, the proposition "bachelors are unmarried men" is not synthetic, it's analytic, because it's simply a definition.  On the other hand, the proposition "all bachelors are unhappy" is synthetic, it's not a definition but rather a proposition that can be tested and shown to true or false.  I don't want to go any further with these terms; the object of this post is to talk about the notion of objective.
Now, organisms need to be able to acquire knowledge of the world around them to survive, multiply, and prosper.  The knowledge of the world that any organism can acquire and make use of is contingent upon the sort of organism that it is. The contingency is defined by the complexity of the organism's nervous system, and also by the needs of the organism- what it has to "know" to make it through its world.  No organism takes in, processes, and acts on raw data; all organisms "filter" incoming data through their senses, which have been shaped by natural selection.  Frogs, for example, have a visual system that is tuned, by evolution, to make them aware of those things around them that they need to "know" about in order to prosper.  Specifically, frogs' visual system consists of the following sorts of "detectors" (Lieberman 1984: 54-55):
  • Edge detectors identify boundaries of objects.
  • Bug detectors identify small convex moving objects.
  • Event detectors identify sudden movements.
  • Dimming detectors identify falling light intensity.
  • Blue detectors identify bodies of water.
Having knowledge about these aspects of the world allows frogs to eat, sit by the waterside, and leap into the water when a potential danger appears.  Frogs need to "know" these things (and some others) about the world if they're going to live long enough to reproduce.   This is as true for humans as it is for frogs, although humans, via culture, can manipulate to some degree the contingencies that apply to them.  So, although we have evolved to be able to perceive and respond to narrow (compared to what the Universe makes available) ranges of light and sound, we can create technology that allows us to see and hear beyond the limits of our native visual and auditory systems.  We can do a lot "better" than frogs, in the sense that our visual system allows us to develop more fine-grained visual knowledge of the world around us.  But we, and frogs, are both constrained by our natures; neither of us can develop knowledge about the world that is "absolute and true."
So, back to objectivity. A scientific definition of objectivity as it relates to the construction of propositional knowledge might go something like this (Lett 1997: 46):
[A proposition] is objective in the scientific sense of the term if it is both publicly verifiable and testable.
Example:  I tell students that the Aymara word for 'your house' is utama.  This bit of knowledge is objective not because it's "absolute and true," but because my students can go to Bolivia or Perú, or nowadays even email an Aymara speaker, and ask them how to say 'your house', and the answer should come back utama.  It's publicly verifiable and testable.
Subjective knowledge is about me: The Aymara language sounds beautiful. Not publicly verifiable, not testable.  Objective knowledge is about us, working together, to develop an understanding of the world: The Aymara language is Head-final (heads of phrases follow their complements).  That proposition can be publicly verified and tested.  And that's what science is about.

References
Lett, J. 1997. Science, Reason, and Anthropology: The Principles of Rational Inquiry. Rowman & Littlefield Publishers.
Lieberman, P. 1984. The Biology and Evolution of Language. Cambridge, MA: Harvard University Press.
Schultz, E. and R. Lavenda. 2012.  Cultural Anthropology: A Perspective on the Human Condition. Oxford University Press.

9 comments:

  1. While I can agree with everything you've stated above, I have to say that "pomo" often gets a bad rap. Thus, as an anthropologist that views it as a necessary critique of the ethnographic theory that precedes it, the boundaries for legitimacy (in terms of objective science-social or otherwise) have been relaxed and restricted in direct relation to the theoretical assertions of women and others since they entered the discipline of anthropology. For example, although Zora clearly developed theories about those she was attempting to understand-and she tested those theories employing varied and rigorous methods in the field- her work was always positioned as a subjective reality because she shared phenotypic features with those she was studying... so much so that we lost her to Literature (due in large part to Boas' position that she could not be fully objective). Would you agree or disagree with that?

    While I bemoan the wild card techniques now considered "postmodern" (as many are simply not verifiable and testable as either objective or subjective because they lack any firm grounding in social theory and elevate experiential data as on equal footing) I can't imagine where my work would be without those considered pomo- not to mention that the postmodern critique ushered in the poststructuralist models I draw from regularly. I agree that the entry is ridiculous and misleading, but i wanted to get your input on the points I am attempting to make. :)

    Sincerely,
    The most junior of the junior resident assistant professors,
    Melissa Hargrove

    ReplyDelete
  2. Melissa, I confess that I didn't know that Boas thought Zora Neal couldn't be "fully objective," but if he did I think he was wrong (I think I just saw a lightning bolt!). Of course, we have to consider the times, etc, yadda, yadda. My central point in the above is not to pound on all of Pomo, which I agree has given us some useful stuff. I wanted to critique the mischaracterization of "objectivity," as well as of science in general, that seems to have intensified under some proponents of Pomo.

    I'm not sure I know what you mean by "the boundaries for legitimacy [...] have been relaxed and restricted in direct relation to the theoretical assertions of women and others." I will say that my own mentor, MJ Hardman, is a female professor of linguistics and anthropology; no boundaries were relaxed during the courses I had with her.

    Let's keep the convo going!

    ReplyDelete
  3. Ron, I have lifted the definitions (with their references) for a post scheduled for July 22nd. I have asked my readers to decide which of the three definitions you give is the best, or is there a better one, and referred them to your post to find your own conclusion. I have then suggested that teachers ask their students to look up a similarly ambiguous word, like "truth," "beauty," or "family" and see what discussion ensues.

    Thanks for giving me the foundation for a great exercise. I hope you don't mind my cribbing it.

    ReplyDelete
  4. Ann, no problem. Please let me know how it works out!

    ReplyDelete
  5. Ron:

    I just published an essay in the Society for Archaeological Sciences Bulletin in which I discuss bias and science and subscribe to your view of objective. In part I state:

    "..one important hallmark of science is the collection of data that can be reproduced by other researchers. This enables others to verify the original work of scientists. It is this, and not some supposed unbiased attitude that makes science a powerful way to make sense of the world. Gould was right in noting that scientists are products of, and indeed constrained by, their contexts and their biases. Indeed, as Martinson et al. (2005) have demonstrated, scientists frequently engage in all manner of misbehavior and succumb to their predispositions and predilections. But when the work of a scientist can be examined and their data recollected (or checked in some fashion), then we can uncover poor and erroneous conclusions."

    I'll send you a copy...but interested readers can find it at: http://www.socarchsci.org/sasb.html in volume 34 (3), pages 21-23. My essay is about the recent Gould-Morton-Lewis et al kerfuffle.

    Gordon

    ReplyDelete
  6. I agree, and there are also degrees of testability. I've been having a discussion on #nlp on freenode which ended up in personal attacks.
    The basis of the descussion was,in short, as follows.
    The statement.
    What's wrong with tap water
    Which i said was more objective than
    What's wrong with water
    Since water could mean lots of things, esp in comparison to tap water.
    It was theirfore more open to subjectie analysis, that is an exhaustive, so objective, test would be much harder since water could mean many things where as tap water limits the scope. For instane it's not the de-ionized distilled water i used in chemistry, nor bottled water.
    The other party said it was more subjective as it added personal bias and went on about morality.
    Now whilst in context that is true, since the objective of the context was to find out how much beer I could drink. Alone the statement could be seen as more objective with tap water than just water
    similaly know is more objective than think, since know is testable where as think is only partly testable. But in context know could be more subjective than think since it limits the will of the other person. I don't think is often used too.If I know is used and the person doesn't know then in context it's more subjective but in isolation its more objective since it can be tested. The person who says they know, however, may not accept being corected. But that's a whole different type of analysis.

    ReplyDelete
  7. to put it another way,
    I walked a mile to work is more objective than I walked to work because in the null context is allows for more objective contexts to be formed. So for example the objective contexts of what type of exercise do people get and how much exercise to people get as opposed to just what type of exercise people get, and indded the type of exercise context in itself isn't very self comparable, that is in the domain of exercise it may not be clear how different types of exercise relate to each other, so the domain is not as testable, self objective.

    ReplyDelete
  8. one further thought, subjective, at or pertaining to the subject could be either the topic or the actor, though for some reason it seems to also relate to the speaker, but that is a gross function of the actor and topic, and attempt by the speaker to change the actor or the topic. which can be more fundementally broken down into testability, objectivity of the statements and the objective/subjective testability of the speakers statements in relation to the objective or subjet of the previous context/objective. subjectivity of the speaker, as in personal bias, is not measurable in any meaningfull way. for instance they may change both the topic and the actor but still be correct.

    ReplyDelete
  9. also how does that postmodernist view relate to any form of computability, does it include folk law, does it include a valid personalbias. For instance the topic is what colour cars should we make, oe speaker says we should make fast cars. so a potentially strong personal bias in the lesser context of colour but in the greater context of making cars quite possibly not biased at all, if the gross objective is to increase sales. and what does that mean for all the people that where talking about colour. some of the most objective things have come from introverts.

    ReplyDelete

Comments and feedback are welcome, as long as they conform to normal standards of civility and decency. I will delete comments that do not meet these standards.