?

Log in

Previous Entry | Next Entry

math/physics genealogy

So, I finally submitted that paper I've been trying to finish for weeks (or months, really). I'll post a link when it becomes available later on tonight. Since I'm done with my research for the moment, I've been having fun surfing mathematician and physicist advisor trees. And that led me to calculating Erdos numbers of various people as well.

One thing I was quite pleased to learn is that my advisor's advisor's advisor's advisor's advisor's advisor's advisor's advisor's advisor's advisor was none other than Carl Friedrich Gauss! In case you don't know what a badass Gauss was, the following Wikipedia page may help:

http://en.wikipedia.org/wiki/List_of_topics_named_after_Carl_Friedrich_Gauss
(a list of links to 48 other Wikipedia articles on subjects named after Gauss)

Tracing my advisor ancestory back as far as the records go eventually leads back to a guy named Otto Mencke, who got his PhD in 1666 from the University of Leipzig, and was a contemporary of (and a corresponder with) Isaac Newton. Wow, that was a long time ago! Here is the full chain of advisors and students that connects me to Mencke:

Mencke -> Wichmannshausen -> Hausen -> Kastner -> Pfaff -> Gauss -> Gudermann -> Weierstrass -> Runge -> Born -> Weisskopf -> Murray Gell-Mann -> Sidney Coleman -> Carl Bender -> Tom Banks -> spoonless

Another person on my dissertation committee is Howie Haber (the man I learned QFT from, aside from one quarter with my advisor). His advisor chain is exactly the same up to Weisskopf, but after that it goes:

Weisskopf -> J.D. Jackson -> Gordon Kane -> Howard Haber

Any physics graduate students is familiar with J.D. Jackson as the author of the canonical graduate textbook on Electromagnetism, the most universally used textbook of any graduate physics textbook.

Although it sounds cool that my lineage goes all the way back to Gauss and Mencke, the truth is that something like a third of working mathematicians (and a sizable fraction of physicists) can say the same. According to the Math Genealogy Project, Mencke has 38,590 descendants, 37,315 of which come from Gauss... whereas the entire database only includes 109067 records as of 1 August 2007. I think this is a result of the fact that there were only a handful of good mathematicians/physicists in the 17th century, whereas it has branched off and exploded since.

The main source I've been using to explore the geneaology from the physics side is Spires. You can view an entire tree of students for anyone (although their database is incomplete in many ways, for instance, my advisor never bothered to list who his advisor was in his record so he doesn't show up). One really interesting tree to look at is Max Born's (you can't view any tree higher or it causes server errors):

http://www.slac.stanford.edu/spires/find/hepnames/www?rawcmd=FIND+NAME+BORN&FORMAT=WWWTREE
Among others, this tree includes Michael Peskin, Kenneth Wilson, James Hartle, Paul Steinhardt, Lee Smolin, David Griffiths, Robert Oppenheimer, George Smoot, Jacques Distler, Greg Moore, David Politzer, Anthony Zee, Roman Jackiw, Joe Lykken, Andrew Strominger, Matt Strassler, Barton Zweibach, Emil Martinec, Bjorken, Drell, as well as the ones I've mentioned above (Gell-Mann, Weisskopf, JD Jackson, Gordy Kane, Sidney Coleman) and lots more! People I know personally in the Born tree: Joel Primack, Howie Haber, Ian Affleck, (and Tom Banks, who is not listed). There are probably even a whole lot more interesting people in the tree, if everyone cared enough to keep their info updated.

It should be noted that the "Born tree" I linked to above is almost the same as the "Wigner tree" since Wigner and Born are both listed as advisors of Weisskopf (don't ask me how one can have more than one advisor, I don't know), and almost the whole tree is under Weisskopf (Oppenheimer being one of the few exceptions). However, Wigner's advisor is not listed in Spires, so I don't know what his chain is before that... could be interesting though.

But the most awesome advisor chain I've found so far while searching through Spires is one starting with Leibniz... I've never seen a longer chain of geniuses:

Leibniz -> Jacob Bernoulli -> Johann Bernoulli -> Euler -> Lagrange -> Fourier, Poisson -> Dirichlet -> Lipschitz -> Klein -> Lindeman -> Hilbert, Sommerfeld, Minkowski

(Fourier and Poisson share a node in the chain above because, amusingly they were both students of Lagrange and both advisors of Dirichlet!)

after that, it splits a bit but still remains interesting:

Hilbert -> Weyl

Sommerfeld -> Debye, Pauli, Heisenberg, Bethe

Bethe -> Michael Naunberg (aka the "Santa Cruz heckler")

Bethe, Kenneth Wilson -> Jackiw

Pauli -> Fierz

This leads me to the advisor chains for some lj friends...

Murray Gell-Mann -> Kenneth Wilson -> Michael Peskin -> Matt Strassler -> pbrane (if he'd stayed in physics)
(before Gell-Mann, pbrane's chain is the same as mine so it also goes back through Born and Gauss)

cocacolaaddict's advisor chain also goes back to Otto Mencke, but takes a different interesting route which splits off at Kastner:

(recall Mencke -> Wichmannshausen -> Hausen -> Kastner)

Kastner -> Erxleben -> Weigel -> Rudolphi -> Muller -> Helmholtz -> Webster -> Wills -> Rabi -> Schwinger -> Sommerfield -> Howard Georgi -> Lisa Randall -> Csaba Csaki -> cocacolaaddict

Oh, and I almost forgot:

Sidney Coleman -> Lee Smolin -> moonaysl
(making hers the same as mine aside from the last link or two)

After all this intellectual geneaology masturbation, it led me back to a question I had wondered before, which is "what exactly is my Erdos number?" And by extension, what is my Erdos-Bacon number?

There are 511 people in history who have Erdos number 1, and 8162 people with Erdos number 2. After a lot of detective work searching various paths, I've discovered that my Erdos number is 5, which is within the margin of error of what I had guessed previously. Since my Bacon number is 2, that means my Erdos-Bacon is a pretty respectable 2+5=7, which happens to be the same as Stephen Hawking's and David Z Albert's. I discovered a lot of interesting Erdos connections which I might make another post about, but basically I found a lot of different ways in which I was 6 links from him, but it took me a long time to find a 5-link connection and I'm certain there are no 4-link connections. It's possible (although unlikely) there are other 5-link connections, but the one I've found goes spoonless - Banks - Aharonov - Bergman - Strauss - Erdos. Peter Bergman was Einstein's assistant and coauthored several papers with him, meaning my Einstein number is 4. Einstein is one of the 8162 people whose Erdos number is 2, since he also published with Strauss. Other Erdos number 2 physicists include Sheldon Glashow, Peter Shor, Freeman Dyson, Hugh Everett, and Peter Freund. I thought for sure I would find a 5-link connection through Freund at first, since he does stringy work... but apparently not. Interestingly, if I had accepted BU's offer for grad school, there's a reasonable chance I'd have ended up publishing with Glashow which would have made my Erdos number 3. For several reasons, he and the other physicists there were unable to convince me to come. Ian Affleck was the most persuasive, though, and the only one who really gave me pause.

Comments

( 11 comments — Leave a comment )
easwaran
Aug. 2nd, 2007 02:03 am (UTC)
A good proportion of philosophers go back to one of those chains too - I think Kant was just two steps from Leibniz, and a lot of important philosophers in logicky areas go back to Tarski or Church or Fitch, who are all in the math chains. Not to mention that just about everyone seems to be descended from Hilary Putnam, who was a student of Reichenbach, who was a student of Emmy Noether.
spoonless
Aug. 2nd, 2007 06:20 pm (UTC)
That's neat. I guess if you go back far enough, everything was philosophy (after all, PhD = Doctor of Philosophy).

In the course of exploring stuff, I noticed that the [Bad username: http://en.wikipedia.org/wiki/University_of_Leipzig]University of Leipzig</a> will be celebrating 600 years of operation in 2009 (since opening its doors in 1409!). That's a long time to be passing down wisdom from generation to generation.

It looks like the philosophy tree (https://webspace.utexas.edu/deverj/personal/philtree/philtree.html) is only 8% as big as the math genealogy project. Is that because there are a lot more mathematicians than philosophers, or just because philosophers don't care as much about history (surely not the latter--or are they just slower to establish a web presence?)

Looks like the 3 biggest lines of philosophers start with Leibniz, Christian Hermann Weisse, and Otto Mencke. Although the strange thing is, I notice Christian Hermann Weisse is 19th century, so shouldn't it be possible to figure out who his advisor was?
easwaran
Aug. 2nd, 2007 06:24 pm (UTC)
I'm pretty sure it's the former - despite philosophy having been around longer, it was never as big as math.

The other strange thing about the philosophy tree is that there have been a decent number of people that are prominent despite never getting a PhD. I think this is the case with Camo Jackson, the father of Frank Jackson and the academic ancestor of a large part of the Australian philosophy community. And several prominent British philosophers of the middle of the century are like that too, like I think Strawson and Sellars.

Also, the philosophy genealogy project isn't really set up to deal with multiple advisors, so it can't represent the fact that the great David Lewis was a student of both Quine and Putnam, for instance.
spoonless
Aug. 2nd, 2007 07:01 pm (UTC)
I was going to ask about the possibility of "not having an advisor"... maybe that is Weisse's case.

It sounds really weird, but I guess if you went to grad school for a few years you could know enough to go off and publish a bunch and establish a reputation without ever getting a PhD. In physics or math, it even seems possible, although I don't know if it's happened... it would be pretty damn hard to get a professorship and have students without ever getting a PhD, but if you were doing impressive enough work hopefully they would bend the rules somewhere. I nominate pbrane as most likely to try something like this... and actually pull it off :)

When people have "multiple advisors" does that mean they switched halfway through? Or are they just listing different people on their dissertation committee? Where I am, it's pretty common to work with several professors while you're in grad school, but you always have one person who is "officially" your advisor, as far as I know.
easwaran
Aug. 2nd, 2007 07:16 pm (UTC)
I think the academic world was different in philosophy in the '50s.

I think in some places it's standard to actually have two. And other times people have two that are both playing something like the primary role, even at other places. I think it's also common if there's someone not officially at your university that you're primarily working with.
spoonless
Aug. 2nd, 2007 06:22 pm (UTC)
Oh, and that's really cool about Emmy Noether being the grand-advisor of Hilary Putnam. I had no idea! It's always cool when students cross over into other fields from their advisors.
mathemajician
Aug. 2nd, 2007 09:03 am (UTC)
Erdos -> A.S. Marcus -> C.S. Calude -> me

:-)
spoonless
Aug. 2nd, 2007 06:39 pm (UTC)
Nice! So you're a 3... want to publish something together (to help me lower my number by 1)? j/k

(to be consistent with my notation, you should have used -'s instead of ->'s)
geheimnisnacht
Aug. 5th, 2007 08:01 am (UTC)
On the wikipedia Erdos Number page, there's mention of another scheme. Take into account all co-authorship pathways to Erdos, thus creating an analog to an electrical network where you are interested in, as I would call it, your Erdos Resistance. Give each co-authorship connection unit resistance. While interesting, it would be much harder to calculate of course.

I also saw that on the Erdos-Bacon number wikipedia page, the example of a low number was 6. So if you had gone to BU, you would have potentially been the minimal case. And you could entertain math/physics students at parties for perhaps a minute or two.

Hmm, I had a thought about proposing Sexual Erdos-Bacon numbers, but I have a feeling the Erdos side wouldn't hold up. Famous scientists need more groupies.
(Deleted comment)
spoonless
Mar. 23rd, 2009 07:33 am (UTC)
Re: ah, that's what I get for skipping out to industry rather than doign grad work in math
Cool... Richard Brauer was one of the people I was just reading about in Symmetry and the Monster, a great popular book about the history of the discovery and classification of finite groups, leading all the way up to the Monstrous Moonshine conjecture.

In the end, I'm skipping out to industry too... I like the idea of actually getting that piece of paper first, in case it comes in handy some day (like, realistically maybe I'll be stranded on a desert island and need it for kindling?).
(Deleted comment)
spoonless
Mar. 23rd, 2009 04:20 pm (UTC)
Re: ah, that's what I get for skipping out to industry rather than doign grad work in math
Oh, wow... then I'm sure he will be discussed in this book, I just haven't gotten to the end yet. I think the last thing I read was about Mathieu finding the Mathieu groups.
( 11 comments — Leave a comment )

Profile

blueshirt
spoonless
domino plural

Latest Month

May 2017
S M T W T F S
 123456
78910111213
14151617181920
21222324252627
28293031   

Tags

Powered by LiveJournal.com
Designed by Lizzy Enger