|
Author
|
Topic: The "God's Eye" View
|
Janitor@MIT
Member
Member # 125
|
posted 20. January 2003 13:15
“Here that fact that there is a “target” is irrelevant, the question is whether physical nature would perform similarly. If so, then physical nature creates “new information” if this circumstance is set up in nature (unless we are using a definition in class “A” for “information”) -- point demonstrated!”
I’ve pulled this comment out of its context (http://www.iscid.org/boards/ubb-get_topic-f-6-t-000279-p-2),
with all due apologies to gedanken, because I thought that there is a broader issue and point of contention than is narrowly topical, but which might be fruitfully explored. So rather than post a response in that topic, I thought I’d start a new topic.
I’m not sure what exactly is the “point demonstrated” here. I suspect the argument is something like, “[The experimenter] is an isolable, negligible, and/or eliminable factor in the experiment, and therefore the experiment proceeds as if the designer of the experiment didn’t exist.” I don’t know how much sense it makes in science to “pretend” as if experiments are not designed, and that experiments are not at least as informative about “nature” as they are that element of nature that designs and conducts the experiment. We might all agree that this is a scientifically “useful fiction,” but Dr. Dembski might disagree- that this is more “fiction” than “useful.” LOL
The “point demonstrated” is little more than “question begged.” (Not to sound rude.) Why in the world should I arbitrarily discount the role of the designer of the experiment, especially as my intuitions and my experiences tell me that this is never the “path to wisdom” in science? I have come to the conclusion, utterly unassailable in my mind, that it would be pure folly for me to not take into account all aspects of the experiment, the methods, implementations, theory, the results and their interpretation, and the experimenter—and I do not set artificial and arbitrary bounds around any of these aspects of “physical nature.”
Does this make scientific knowledge of “physical nature” impossible, as someone has seriously suggested to me? No. Quite the opposite—it accounts for everything in “physical nature” and makes no arbitrary “philosophical” demarcations.
As I have emphasized in many posts this “methodological” prescription (and its not even that actually) of “naturalism” (which I think I detect here) does nothing but divinize experimenters and turns all of our experiemtns into nothing less than exercises in .html"special creation," in a supremely ironic self-contradiction. As if experimenters had a “God’s Eye” view of nature! As if experimenters stood in relation to their experiments in the same way that some religions imagine God stands in relation to the world.
To my mind, a bizarre way of thinking about science, quite frankly.
(Per my usual policy I will be as inexcusably neglectful of this topic as all the others I started. LOL But I couldn’t recall if this was the subject of any topic here yet. Might generate some interesting discussion though.)
IP: Logged
|
|
andyg
Member
Member # 415
|
posted 20. January 2003 14:06
Janitor,
I don't know what the moderator wil make of this reply, but I'll dip my toe in the water and see.
Your post starting this thread was extremely poorly written. It is very hard to understand what you are trying to say in it. It *sounds* vaguely impressive, but I find myself scratching my head when trying to figure out exactly what you wanted to say.
I think Brainstorming is an excellent pursuit, but I still feel we have a responsiblility to express our ideas as clearly and succinctly as possible.
AndyG
IP: Logged
|
|
Moderator
Administrator
Member # 1
|
posted 20. January 2003 14:20
I'd have to agree with AndyG to a certain degree.
Since I'm a moderator and as a moderator at Brainstorms I need to moderate content as well as behavior, it would be great, Janitor, if you could do a better job of articulating your point in this thread.
As I read it now, you are trying to say that methodological naturalism actually divinizes the experimenter, and expects or requires a God's Eye View.
If this is your point, then I think it borders on not being an appropriate Brainstorm because it is not offering a positive hypothesis but rather critiquing a methodological stance.
Can you help me out?
IP: Logged
|
|
Janitor@MIT
Member
Member # 125
|
posted 20. January 2003 14:34
Dispose of the topic as you see fit. I will not pursue the matter.
IP: Logged
|
|
RBH
Member
Member # 380
|
posted 20. January 2003 15:14
I'm not sure that Janitor's point is negligible, though I don't think it necessarily has considerable force, either. If I understand it correctly, it is that in considering the results of an experiment it is necessary to take into account the role of the experimenter (who set up the experiment, translated theory language into operational language, defined the independent and dependent variables, determined the measuring apparatus to be used, and so on) in determining its outcome.
In some disciplines that is appropriate. For example, I seem to recall research decades ago in psychology (Rosenthal?) that showed that in social psychology experiments the expectations of the experimenter about the outcome could be transmitted in various out-of-awareness ways to the experimental subjects, who in turn altered their behavior so as to produce the results expected.
In other disciplines, though, it seems less pressing. For example, an experiment examining the solubility of some material in water appears to me to depend less on the experimenter's presence and behavior than, say, an experiment in social psychology. In the latter, the experiment is itself an instance of the domain being experimented in.
However, there are ways of overcoming that kind of bias - double blind designs are one, and replication by independent experimenters is another. I also seem to recall (from philosophy of science, maybe?) a distinction among varieties of replication ranging from absolutely literal reproduction of the original experiment in all its technical details to a new experimenter using the concepts and variables identified as important by the original research but independently defining experimental operations and manipulations, measurements, and so on, so that one is testing the more general conceptual structure of the original research rather than the literal details.
So while an "experiment" clearly includes an experimenter, that's not a disabling or invalidating consideration given appropriate methodological safeguards. Those can in effect take the experimenter out of the picture. As I've said elsewhere, researcher-independent methodologies are necessary, not methods that depend on idiosyncratic properties of particular investigators.
RBH
[ 20. January 2003, 15:20: Message edited by: RBH ]
IP: Logged
|
|
andyg
Member
Member # 415
|
posted 20. January 2003 20:07
RBH:
quote:
So while an "experiment" clearly includes an experimenter, that's not a disabling or invalidating consideration given appropriate methodological safeguards. Those can in effect take the experimenter out of the picture. As I've said elsewhere, researcher-independent methodologies are necessary, not methods that depend on idiosyncratic properties of particular investigators.
Any experiment is only as good as the thought that goes into it. The critical feature of an experiment is the controls that one performs alongside the experiment itself. An experiment without an appropriate control is meaningless, as you have nothing to compare the data with. This effectively dispenses with the "experimenter effect" (whatever that is), because it is present in both conditions
AndyG
IP: Logged
|
|
|
Printer-friendly view of this topic