Sciences vs. Humanities, Primary vs. Secondary

Thoreau offers without qualification some observations about the different approach to books taken by sciences vs. humanities. Specifically, he notes that despite frequent claims that it is the Most Important Book Ever, nobody actually reads Newton's Principia Mathematica

This is totally different from humanities. In humanities, people make a point of reading the original thinkers. They don't just say "Well, philosopher so-and-so influenced lots of other people and got the ideas rolling, so let's read somebody influenced by him and maybe a Cliff's Notes version of the original." They actually sit down and read the original, or a translation of the original.

I guess the difference is that humanities scholars are interested in people and ideas, while we evil fiziks types claim to be interested in the external world.

I always think of this in terms that were used during the discussion of our general education curriculum a few years ago. In talking about the content of the mandatory first and second year classes, people from the humanities and social sciences said many times that it was critical for students to learn the difference between primary and secondary texts. Many of us in the sciences sort of scratched our heads at that, because it's not a distinction that comes up in the physical sciences.

The difference here is in what's being studied. The physical sciences are studying the nature and behavior of the universe, which isn't something that's written down. The analogue of reading a primary source would be doing an experiment, not reading anything.

In most humanities disciplines, though, the text is the point. As a result, the distinction between primary and secondary sources is critical. The whole point of the business is to read and interpret primary sources-- trying to produce humanities scholarship without reading the original texts is like trying to do science without experiment or observation. It's the humanities equivalent of string theory.

Thus, I'm not terribly concerned about the failure to read original sources in physics. While there are sometimes advantages-- the first paper presenting a new idea often explores it in more detail than subsequent articles-- there are just as often disadvantages, as more modern notation and mathematical apparatus are often clearer and more compact than the original approach. Newton's Principia is kind of a special case, too, as it's famously difficult to follow, possibly deliberately.

Categories

More like this

Academics of all sorts are highly protective of their scholarly territory. It's an unavoidable consequence of the process of becoming an academic-- I've often joked that getting a Ph.D. requires you to become the World's Leading Expert in something that nobody else cares about. To make it through…
Marie-Claire Shanahan teaches science education at the University of Alberta, and blogs about her own research and about the state of science education (and science education training: science education education if you will). Her latest post summarizes her findings from reviewing science…
One of the standard elements of most academic hiring and promotion applications, at least at a small liberal arts college, is some sort of statement from the candidate about teaching. This is called different things at different places-- "statement of teaching philosophy" is a common term for it,…
Timothy Burke has some interesting thoughts about the College of the Atlantic, which represents a real effort to build interdisciplinarity on an institutional level. "Interdisciplinary" is the buzzword of the moment in large swathes of academia, and the College of the Atlantic, which doesn't have…

I think the equivalent for science is that we always have to go to the primary literature - not to a review article. We need to see the data to draw our own conclusions.

We use textbooks to cover the "old" stuff that is sufficiently accepted that there aren't debates any longer. All the new sections in textbooks, though, come with references to where you can find the original papers.

The analogue of reading a primary source would be doing an experiment, not reading anything.

Yep. I tell my humanities friends that my primary texts are rocks.

Of course, they think I'm weird...

The historical equivalent of the primary source is the experiment -- in both disciplines, the thing in direct contact with the original reality is how we learn, and while we may not have undergrads read crumbling parchment archives any more than we have them measure the melting point of plutonium, in both cases we strongly urge them to be at least aware of what's going on.

I have frequently found, on both sides of the Two Cultures debate, that when I'm trying to get a general picture of "what on earth is this thing all about", that some seminal work from the foundation of the discipline is often _extremely_ helpful, mostly because the work had to be written with an absence of not-yet-invented jargon for an audience that had no familiarity at all with the subject. Darwin's _Origin of Species_, Smith's _The Wealth of Nations_; other possibilities come to mind as well.

The analogue of reading a primary source would be doing an experiment

I once made the argument, in a Shakespeare class, that reading Macbeth is a thought-experiment on ambition. It occurs in restricted, artificial circumstances (setting and scenes). The elements are reasonable approximations of the objects under study, "distilled" in such a way as to make observed behaviors attributable to particular characteristics of those objects (characters and motivations). The action unfolds as a reasonable analog for real events. If you think about it, it's really not that different from imagining a person in an elevator, and then in a rocketship. Granted, Shakespeare writes a better monologue than Einstein. :-)

Some of my fellow students got it. Some of them looked like I was speaking Martian . . .

To some extent, the original resources are still a great place to go for some stuff in biology. The Origin of Species is still read by numerous people, and even books like Fisher's Genetical Theory of Natural Selection are arguably somewhat worthwhile, if not just for their idiosyncrasies.

Nonetheless, I have always found it weird that physicists don't seem to worship the Principia like we worship the Origin <_<

Well, one big difference between the Origin of Species and Principia Mathematica is that the Principia is demonstrably wrong! ;-) Well, demonstrably a low-velocity, low-gravity approximation, anyway. Yes, there are many areas in which modern biology has superseded the Origin, but the core of Darwin's great work (evolution by natural selection) remains accepted today.

More broadly, I think the fundamental distinction here is not between primary and secondary sources, or between studying the external world versus texts. It's the idea of science as a process which improves our understanding over time. The natural instinct of a physicist is not to go back to the oldest paper on a subject, no matter how important its insight was, it's to go find the most recent articles in the field that contain the most up-to-date scholarship, the best experimental techniques, and the distilled wisdom of everybody that's tried it before.

My impression is that in the humanities, no one really tries to surpass Shakespeare or van Gogh - but we surpass Newton and Darwin and Einstein all the time.

History and math are the two extreme cases. In history, anything in a secondary source that didn't originate from some primary source is at best interpretation or conjecture. There's a conservation law: nobody can generate new historical documents or artifacts (you can only find ones that already exist).

By contrast, in math the origin of ideas is irrelevant to their importance. Our understanding of discoveries tends to increase over time as more researchers elaborate on them and develop more examples and better explanations. The original sources are still worth studying (to see how the ideas developed historically or how to assign credit), but they have no special status. Some are illuminating and beautifully written, while others aren't, and there's little point in reading the latter when a better exposition is available.

I don't entirely agree that the scientific analogue of primary sources is experiments. There's a lot of truth to it, but it's not the full story. For example, consider Einstein's paper on special relativity, where all the experimental facts were already known. You could consider this a secondary source in the sense used in history, since it interprets and explains primary sources (namely the experiments), and indeed that is logically correct. However, I think this underestimates theoretical physics. In history, any interpretation is highly limited and misses much of the truth, and many interpretations are possible. By contrast, in physics there's no serious competition to "Einstein's interpretation", which gives it a very different status. From this perspective, his paper feels much less like a secondary source.

By Anonymous (not verified) on 18 May 2009 #permalink

eofhan: You actually made a good argument. Zola wrote a fair bit about the idea of a novel as an experiment, since he felt that an author started by choosing a setting and some characters and then letting them follow their courses. All he had to do was write down what they did. He would have been right at home with the Sims. He also wrote the original pot boiler, or at least the first novel explicitly called such.

Marshall: Let's pile on Newton. Not only was his physics wrong, but he didn't have a clue as to why his calculus worked. It took another century or two for someone to figure that out.

----

I read original science because I am curious about the history of ideas. If nothing else, it gives me a sense of where we are today. There are always a lot of things half understood, and a sense that clarity is just around the corner, or not. I have an old book on the Bohr theory of the atom, and it is totally bizarre. The fold out sheet of weird ass "orbitals", some circles, some ellipses, some like bunnies, was almost too weird, but Bohr was integrating what had been learned from studying spectral lines. His catalog of orbital characteristics was critical in the development of quantum mechanics. Of course, anyone trying to do physics out of this book today would get curious looks.

(An earlier owner had scrawled in the book that an atomic bomb was probably impossible, but a partial atomic bomb was quite likely. I assume the former implied complete conversion of matter to energy. I had always thought that our atomic bombs were plenty complete.)

Of course primary texts are really important in physics -- they area called experimental results. Galileo called them "the book of Nature".
And it is crucial that students learn to read it, because otherwise they end up as crackpots who think that "theory" means "something I just made up".

This is something that is actually a big focus of my pedagogy. Teaching at a SLAC where essentially 0% of the students are scientists, I find that good way to get the humanities students into the subject is by taking the approach that they are familiar with. Thus we spend the first couple of weeks of the general physics course doing a "close reading" of various sections of Aristotle and of Galileo's "Dialogues." When we get to Newton, we read his "Rules of Reasoning" from the Principia, which is a good starting point for some "how does science work?" discussions. Then when it comes time to do Newton's Laws, I put them up on the screen in the original Latin, and we translate them one cognate/root at a time. ("corpus" = body... "perservare" = remain... "quiescendi" = stillness...). And one can argue that I'm wasting time that could be better used by doing more block-on-an-inclined-plane problems, but I think it serves a particular body of students who DO care about the character of the original texts, and makes them feel like they are doing something familiar rather than alien.

Interesting topic.
As for me the distinction between "sciences" vs "humanities" can be quite if we talk about studies but we often forgetthat it's artificial as long as we talk about people. Every single humanist must have at least some features od scientist (just imagine psychologist as the most conspicuous example...) and vice versa.
However the point about reading primary or secondary book is really noteworthy. But I found it quite natural - as it was said the first version of "humanistic" book is the best one (just try to rewrite Makbeth in your own word without loosing the atmosphere, haha...), however the first version of "scientific" book can be a bit... yes, a bit unclear. But I want to emphasise it's rather a matter of specifity of study area, not of the indyvidualistic human feature.

The distinction is not so much that humanities people use original sources and scientists don't, as that scientists always build on sources, so that anything other than a review paper or textbook is an original source. It can be fun to go back and read an original source. For example, a couple of years ago I tracked down the original reference on the Haar wavelet, a 1910 paper written in German--about 70 years before anybody knew what a wavelet is.

I must disagree with the comment of Anonymous #9: There's a conservation law: nobody can generate new historical documents or artifacts (you can only find ones that already exist). On the contrary, we're all generating new historical documents all of the time. Even a scientific paper is a historical document, a tracer of what we used to think about a given topic.

By Eric Lund (not verified) on 19 May 2009 #permalink

Do they read the Principia in the original Latin at St. Johns, the way a Hums person would, or the Cotes translation, or a more recent translation? Or all three with a careful textual analysis?

That said ...

There is some value in reading the original Newton, particularly if you are a fan of the efforts to confront Aristotelian misconceptions in first-year physics. Newton goes to great length, right down to inventing a way to split forces into the "real" ones (impressed forces) that show up in the 2nd Law and the ones Aristotle used to keep things moving (along the way renaming this vis insita as the vis inertiae), then discarding the latter with the 1st Law. He also goes to great lengths to point out that an impressed force only acts while it is being impressed - so when the string breaks, the object travels in a straight line.

PS -
I've actually read a first edition in the Library of Congress. Quite a remarkable experience.

By CCPhysicist (not verified) on 21 May 2009 #permalink

At St. John's (Santa Fe, at least) we usually read the Densmore/Donahue translation, which includes a great deal of helpful explanation/simplification.

By William Davis (not verified) on 22 May 2009 #permalink

I read the Principia in the Latin at SJC Santa Fé in the 70s, but everyone else used the same English translation. I think the reasons for confronting an original text in the sciences are the same as for confronting one in, say, literature - to confront the original thought without filters.

By Kris Shapar (not verified) on 24 May 2009 #permalink

The trend of physicists ignoring all but the most recent papers that are relevant to their work is fairly recent. If you look back before World War II most of the physicists seem to have had a thorough knowledge of the history of their discipline. A close examination of the history of physics can lead to some interesting insights. One can make a case that Ernst Mach's "The Science of Mechanics", which was largely a historical work, inspired much new work in theoretical physics (including, at least to some extent, Einstein's work on special relativity). Certainly most physicists don't need to read a lot of 150 year old papers, but it is probably a good idea for some of us to do it every now and then.

I agree with some earlier comments that a historical approach is particularly useful for TEACHING non-science majors. And textbook writers could stand to read a few more original papers rather than just passing along the urban legends that have built up over the years.

A good English edition of the Principia is the Cohen translation. It has lots of explanatory comments (several hundred pages) at the beginning. I've read parts (someday I'll read it all, but I say that about Ulysses too). If you're looking for a good read, Galileo is fantastic (and available in good translations by Stillman Drake).

By Todd Timberlake (not verified) on 24 May 2009 #permalink

The trend of physicists ignoring all but the most recent papers that are relevant to their work is fairly recent. If you look back before World War II most of the physicists seem to have had a thorough knowledge of the history of their discipline. A close examination of the history of physics can lead to some interesting insights. One can make a case that Ernst Mach's "The Science of Mechanics", which was largely a historical work, inspired much new work in theoretical physics (including, at least to some extent, Einstein's work on special relativity). Certainly most physicists don't need to read a lot of 150 year old papers, but it is probably a good idea for some of us to do it every now and then.

I agree with some earlier comments that a historical approach is particularly useful for TEACHING non-science majors. And textbook writers could stand to read a few more original papers rather than just passing along the urban legends that have built up over the years.

A good English edition of the Principia is the Cohen translation. It has lots of explanatory comments (several hundred pages) at the beginning. I've read parts (someday I'll read it all, but I say that about Ulysses too). If you're looking for a good read, Galileo is fantastic (and available in good translations by Stillman Drake).

By Todd Timberlake (not verified) on 24 May 2009 #permalink

The trend of physicists ignoring all but the most recent papers that are relevant to their work is fairly recent. If you look back before World War II most of the physicists seem to have had a thorough knowledge of the history of their discipline. A close examination of the history of physics can lead to some interesting insights. One can make a case that Ernst Mach's "The Science of Mechanics", which was largely a historical work, inspired much new work in theoretical physics (including, at least to some extent, Einstein's work on special relativity). Certainly most physicists don't need to read a lot of 150 year old papers, but it is probably a good idea for some of us to do it every now and then.

I agree with some earlier comments that a historical approach is particularly useful for TEACHING non-science majors. And textbook writers could stand to read a few more original papers rather than just passing along the urban legends that have built up over the years.

A good English edition of the Principia is the Cohen translation. It has lots of explanatory comments (several hundred pages) at the beginning. I've read parts (someday I'll read it all, but I say that about Ulysses too). If you're looking for a good read, Galileo is fantastic (and available in good translations by Stillman Drake).

By Todd Timberlake (not verified) on 24 May 2009 #permalink

The trend of physicists ignoring all but the most recent papers that are relevant to their work is fairly recent. If you look back before World War II most of the physicists seem to have had a thorough knowledge of the history of their discipline. A close examination of the history of physics can lead to some interesting insights. One can make a case that Ernst Mach's "The Science of Mechanics", which was largely a historical work, inspired much new work in theoretical physics (including, at least to some extent, Einstein's work on special relativity). Certainly most physicists don't need to read a lot of 150 year old papers, but it is probably a good idea for some of us to do it every now and then.

I agree with some earlier comments that a historical approach is particularly useful for TEACHING non-science majors. And textbook writers could stand to read a few more original papers rather than just passing along the urban legends that have built up over the years.

A good English edition of the Principia is the Cohen translation. It has lots of explanatory comments (several hundred pages) at the beginning. I've read parts (someday I'll read it all, but I say that about Ulysses too). If you're looking for a good read, Galileo is fantastic (and available in good translations by Stillman Drake).

By Todd Timberlake (not verified) on 24 May 2009 #permalink

OK, this comment function is being a pain. Why did it repeatedly tell me that it had timed me out and that if I was trying to post a comment I should try again? Obviously I tried a few too many times. Sorry about that....

By Todd Timberlake (not verified) on 24 May 2009 #permalink