Philosophy 167: Class 7 - Part 12 - Getting Beyond Mere Conjecture: What Galileo Achieved, and the Problem of Scientific Truth.

Smith, George E. (George Edwin), 1938-


  • Synopsis: Discusses Galileo's views on the relationship between theory, experimental results, and scientific truth, and the nature of science.

    Opening line: "Okay. This is from a note that Galileo published concerning the supernova of 1604."

    Duration: 10:45 minutes.

    Segment: Class 7, Part 12.
This object is in collection Subject Genre Permanent URL
Component ID:
To Cite:
TARC Citation Guide    EndNote
Detailed Rights
view transcript only

Okay. This is from a note that Galileo published concerning the supernova of 1604. I singled it out, I first discovered it in Fantoli's book on the trial, but what I wanna focus on, I'll read it quickly, or start reading it. Because I too, among so others, have had the idea to submit to the judgment of the public what I think not only about the location and the motion of this light.
Everybody knows what a supernova is? Something that suddenly appears in the sky as a star that wasn't there and never been seen before. So there's a supernova in 1604, not the first one in this era. What I think not only about the location and motion of this light but also about it's substance and it's origin because I have believe I have found an opinion which contains no obvious contradictions in which therefore might be true.
It has been necessary for me so that I might be sure of myself to go ahead slowly and await the return of this star in the cast after its separation from the sun. And to observe again very diligently what changes it might have undergone, both in its location as well as in its visible brightness, etc.
I have finally, and continuing my speculations about this marvel, I have finally come to believe that I could know something more than what ends in mere conjecture. It's that phrase is the sole reason I put this up there, okay. I could know something more than what ends in mere conjecture.
Okay. That's what is reasonable to attribute to Galileo in all of his work. He wants to get beyond mere conjecture. And getting beyond mere conjecture, particularly near the surface of the Earth, was very, very difficult. It was a lot better belestially because regularities were so well behaved. Near the surface of the earth it was very difficult.
So what could he mean or what could anyone mean at the time by getting beyond mere conjecture? Well, we've seen three possibilities already in Galileo. And it's not clear to me, going back to Isaac's question much earlier, which one he really means to hold to, or perhaps he means to hold to all three, and perhaps he doesn't even see they're different.
Okay first one, a science seeking detailed agreement within observational accuracy between all of its predictions and the results of experiments in which external confounding effects have been suitably controlled. Okay, and now the way to judge that is what do discrepancies tell you? They tell you their source of continuing evidence.
Okay? Now this fits Kepler. It actually fits Ptolemy up to a point, because he revised his theory of the moon because of the discrepancies, okay? But the key thing here is exact agreement. If we don't get exact agreement and we can't attribute the discrepancies to measurement error or some other confounding effect, we have to take the discrepancy seriously, they're telling us something.
Second possibility, a science, and do notice the scare quotes there, seeking agreement between a more striking mathematical consequences and the results of select experiments centered on unusual distinctive phenomena to a degree sufficient for it not to be clearly falsified. And in this case, we wanna just explain away any discrepancies.
The whole point of getting an experiment in which we can do quasi-quantitative agreement is sufficient, is we can dismiss discrepancies as arising from things we can't control. And that probably describes Ptolemy's astronomy, because what he was trying to do was predict certain distinctive events and didn't worry about discrepancies in other places.
Probably even thought the motions weren't that perfect. That's a possibility, not probable. Or third, a science seeking, once it is suitably calibrated, sufficiently good agreement with empirical phenomena of interest for it to serve practical purposes in prediction primarily, but also in explanation where discrepancies, to use the phrase I used before, are to be swept under the rug.
Now those are three different pictures of science. They're still with us, all three pictures. Galileo has, at one point or another, put forward all three of them in two new sciences. And I don't know what he thought about the difference between them, the relations between them, etc. Okay.
We're working out what a science can be. The third is what's usually called engineering science. Okay. Good I have three minutes for my last slide. I'm actually going to over run, I wasn't as good as I thought. I forgot this slide, then I have a final slide, so I'll be quick with this one.
Galileo did two extraordinary things from the point of view of the role of experiment within two new sciences. The first is to do much more than just intervene in nature. He recognized the value of constructing totally contrived situations that never occur in nature at all as a source of evidence.
And in conjunction with that, he equally recognized that the way to do that is to take theory very, very seriously, and use theory both to contrive the experiments and to figure out how to measure things by proxies, etc. That's a real birth of serious experimental science and experimental physics.
Experimental physics has had a history of doing experiments in situations that have nothing to do with normal events in nature. Optics is almost entirely a science of getting light in a situation that never occurs naturally. Okay. And just the four things here you can falsify, use experiments to falsify theoretical claims.
You can justify conceptual assumptions with them. You can even confirm a theory by successful selling predictions. And the other use, Galileo does almost none of this, but Mersend and Rochioli start trying to measure constants of a theory, especially constants of proportionality. You can think about that yourself, I wanna turn to this, which is this book.
This book, I'm told, it has sold the second most copies of any book in the history of science in the 20th century except Tom Coon Structure of Scientific Revolutions. And it is surely the most influential book ever in structure of scientific revolutions. It's done by a sociologist of science, Steven Shapen who is now at Harvard.
He was not at Harvard when he wrote this. And Simon Schaffer, the leading historian of science at Cambridge University, and arguably, in all of England, he's the person who set the style. The purpose of this, or the reading that people have put on this book, and I don't know Steve Chapin, but I know Simon Schaffner quite decently.
The whole point of this book, and it's not about what we're studying, it's about the air pump and Boil's Law, is scientific truth as a social construct. And the line throughout is somebody does an experiment, and now it's a complex social construction to figure out what that experiment is showing.
And they simply take the series of experiments done with the air pump in the royal society and other places and show how it was an adjudicative process to arrive at what result was showing. Okay? And I don't deny that. I don't deny at all that the result of any experiment is a social construction.
It doesn't stare you in the face. I hope the one thing that's come out of the last two classes, results of experiments do not stare you in the face the way you're told. Now, very far, deep into the book is the one claim they make that I- By the way, you can think of all my work as trying to repudiate this book.
As I said to Simon, why'd you stop the book in 1680? We continued doing pneumatic experiments down to the present time. Don't they count too?
In other words, it's not just a local thing of one group of people talking to one another. It's a long-time thing, and that's what I mean by the importance of assimilation within a larger community and various things.
But they're right about one thing. Here we endeavor to recover this labor for our historiographical purposes. What's the purpose? Notice the first thing. Establishing matters of fact did require immense amounts of labor. To show the inadequacy of the historiographical method, which regards experimentally produced matters of fact is self evident and self explanatory.
Now what they're attacking is the Harvard studies on the history of science. Openly, they're saying. And if you go look at those, Jim Conan, James Conan. Jim Conan is his grandson. James Conan, president of Harvard, did indeed contrive these wonderful studies in the history of science in place after place, they describe the experiment.
And the result seems absolutely transparent in their description, just the way it does in every textbook you've ever read. And their point is, that's not the way real experiments work. And I hope I've shown in these two classes that it's totally true with Galileo. But it doesn't mean that a truth is a social construct.
What it means is for facts to be established, whole communities have to be involved with all sorts of cross checks, all sorts of complementary experiments, etc. It is not a simple process. And that's the main point I wanted to get across tonight is just how much of a complex, social process it is to establish scientific truth.
That doesn't mean it's a social construct, however.