The Real Reasons Why Philosophers Shouldn’t Use LaTeX

Josh Parsons (Oxford) has written a widely discussed post on “The LaTeX cargo cult,” explaining why he discourages philosophy students from using LaTeX.  He makes some interesting points.  But what he has left out is the overarching principle that you should simply always use the best tool for the purpose at hand – and “best” should take into account lots of things: cost (in money and time you need to invest to become proficient in the use of the tool), ease of use, functionality, and the needs of the prospective audience.

For a long time, LaTeX had the upper hand over available alternatives (i.e., Microsoft Word).  It produced high quality output (Word didn’t), it was free (Word wasn’t), it could do lots of things Word couldn’t do (like bibliographies), it was an open format (Word wasn’t).  Well, times have changed. There are more alternatives, and the alternatives now can do lots of things they didn’t use to be able to. The latest Word document format is open, and based on open standards like XML. There are free, open source alternatives to Microsoft, such as LibreOffice. The alternatives have gotten better at typesetting, and you can now do most of the things in which LaTeX had the upper hand for a long time, e.g., bibliographies and reference management, through plug ins and add-ons (both non-free like Endnote, and free, open, cross-platform like Zotero or JabRef).

So while at one point “well, I use bibliographies and references a lot, and I want to have a nice-looking hardcopy” were sufficient reason to use LaTeX and spurn Word, that’s no longer the case.

Given this fact, other considerations should probably play a more important role now when deciding whether to learn LaTeX and when to use it.

  • LaTeX still has a steep learning curve and you can run into complex issues (and simple issues that are hard to solve).  If you have limited amounts of time – say if you’re a grad student writing a dissertation – then becoming proficient at and writing everything in LaTeX will probably be a distraction.
  • LaTeX on its own is very bad at revision control and commenting, but Word and LibreOffice are very good at it.  If your piece of writing requires others to read, comment on, and make revisions to it – say, if you’re a grad student writing a dissertation with an advisor who doesn’t use LaTeX and would like to easily comment on your drafts – then don’t use LaTeX. (The same goes for writing any kind of administrative document that anyone else in your institution has to open, comment on, reformat, reuse, or revise!)
  • LaTeX is very good at producing print-based output, but pretty bad at producing output that can easily be reused in other formats – say, on a web page or in a form – so if you need to use your piece of writing in settings where formatted or unformatted text is needed – say, if you’re a grad student preparing funding applications via web-based forms – think twice about using LaTeX.
  • LaTeX is very good at making your writing conform to a given format (e.g., a thesis or journal layout), but it can be very time consuming to make LaTeX output conform to a format for which no class or style package exists.  So if there’s an Word (or PowerPoint or whatever) template for what you need but no LaTeX style file – then it’ll probably be easier to just use that.  (E.g., I wouldn’t dream of writing letters of recommendation in LaTeX given that there’s an institutional letterhead template.)

Of course all this doesn’t mean that you should never use LaTeX, and I think it also doesn’t mean that we should discourage students from learning (about) it.  In fact, I think it would be a mistake to do so. There are lots of scenarios in which LaTeX is the best option.  And there are good reasons grad students should at least have a passing familiarity with LaTeX.

  • Do you work in a (sub)field where LaTeX use is prevalent (logic, physics, math)? Then you should probably learn and use LaTeX. (Parsons acknowledges this! But even if all you do is TA intro to formal logic once, learning and using LaTeX can pay off immensely!)
  • Does the thing you’re writing need any of the powerful features that LaTeX has but, say, LibreOffice doesn’t?  Use LaTeX.
  • Does your advisor use LaTeX and invite you to co-author a paper with her? Learn LaTeX.

There are other reasons to use LaTeX. There are other reasons to not use LaTeX (and scenarios where other tools are better).  But don’t not use or learn LaTeX because it’s a cargo cult – it isn’t – or because it’s a proprietary format – it isn’t – or because it’s not a “declarative language.”  It’s a powerful tool that’s useful in certain contexts. If you find yourself in such a context, learn it, and use it. And given that it is relatively widely used, at least learn what it is so you can make an informed decision. And perhaps encourage your students to do so, too.

Bringing Logic (and Philosophy, CS) to the Masses

At this year’s Vienna Summer of Logic the organizers did something I haven’t seen done before, and which I think should be emulated: over the course of the two weeks that 2,400 logicians were gathered in Vienna, they organized a Logic Lounge in seven instalments.  For an hour each, one or more conference participants engaged in a moderated conversation in front of a general audience in a café near the conference venue. The moderators were well-prepared, and the discussants all had interesting things to say: about what logicians “do,” about important results and why they were important, about connections between logic and other areas.  There was a session with Georg Gottlob about how logic regained a foothold in Austrian intellectual discourse and in Vienna’s universities in the 1980s (due in large part to people like Peter Weibel, a high-profile Austrian media artist), one on Gödel’s theorems, a conversation with Christos Papadimitriou (among other things about Logicomix), one with Moshe Vardi on the ethics of AI, and one on women in logic with Ruzica Piskac and Magdalena Ortiz .  (I unfortunately had two miss the events featuring Roderick Bloem and Byron Cook.) These, I thought, were very successful and engaging ways to bridge the gap between the rarefied and technical academic program of the conferences making up the VSL and the public.

It was a rewarding experience for me: both as a member of the audience, and as the guy that got to explain what the incompleteness theorems are about.  It wasn’t something teaching prepares you for: in the classroom, you have audiovisual materials, you can rely on a textbook, you can expect your students to have some background.  In the Logic Lounge, it was basically like explaining Gödel to someone you just met in a bar.  You can’t presume anything, you can’t use any jargon or formulas, and you have to make sure you give the “big picture” and explain why anyone should care.  I hope I did a decent job.

The organizers put a lot of effort into the events and the “public” aspect of the VSL in general, and I find that very laudable. Logic isn’t something you learn about in high school or even in university unless you take a course in it.  It’s something the general public only has pretty vague ideas about – but something the specialists think (with good reason) is important and should attract more interest, students, and funding.  The same can be said for philosophy and probably at least for some areas of computer science (theory) and (pure) mathematics.  So why don’t more conferences do that sort of thing?

Organizing a conference is a lot of work.  But it’s also a valuable opportunity to publicize the value of what we do in academia to the “outside” world.  You’ll have a number of able and hopefully willing eminent people available to participate, you don’t have to worry about attendance (since at least some of the conference goers will be curious), and you have a chance to raise the profile of your discipline and perhaps the local university department to the public.  Topics aren’t hard to find. The ASL could have a keynote speaker chat about infinity or hypercomputation. The PSA could have a philosopher of science talk about scientific evidence and climate change. At LICS you can have someone talk about security and verification of programs, at PLoS about why the language in which you code is important.  And at the APA: any neat topic that would wow students in intro courses: philosophy of travel, existence of god, paradoxes, justice, science and free will  — but don’t just do it like an intro course: have your participants also talk about what they’re writing on.  

You need a venue willing to cooperate: since the people attending will all order at least a drink and you’re probably running the event before dinner, that should be easy to find.  And you need someone to moderate and ask questions, and that person should probably be someone who isn’t an expert – else you run the risk of the conversation ending up at too high a level.  Ask a local journalist, or someone who already has some experience running events like these, e.g., a local Science Café. And then get some publicity: send out a press release (or have the local University send out one).

Thanks to the organizers of the VSL Logic Lounge, Oliver Lehmann and Helmut Veith, to Thomas Kramar (Die Presse) for the keeping us on track, and everyone who came and asked questions!

Proof Formalization in Mathematics: Guest Post by Jeremy Avigad

There’s a discussion going on at the Foundations of Mathematics mailing list about the purpose and value, actual and potential, for formalized proofs in mathematics.  Harvey Friedman asked Jeremy Avigad to comment; he sent this super-useful list of references, republished here with his approval.

John Harrison and I recently wrote a survey on formalized mathematics, for computer scientists:

Here are some slides from a related talk that I presented at the AMS Joint Meetings in Baltimore earlier this year:

Tom Hales recently wrote a nice piece for the Bourbaki seminar:

He has another lovely survey here (which discusses computation in mathematics more generally):

There is an issue of the Notices of the AMS from 2008 dedicated to formal proof:

I recently wrote a review of Hales’ nice book on the proof of the Kepler conjecture, to appear in the Bulletin of Symbolic Logic:

Here is a survey of mine on some of the difficulties involved in making sense of ordinary mathematical language and notation:

There is now a substantial literature on formal mathematics, and writeups of formalizations regularly appear in conferences like Interactive Theorem Proving (ITP), as well as journals like the Journal of Automated Reasoning. Homotopy type theory has also gotten a lot of press lately, in parts for its interest as a new framework for verification.

Here are some formalizations I personally have worked on:

Most of these have something to say about the current challenges and difficulties.

There are number of good interactive theorem provers out there. I am currently involved in the design and library development for a new one, Lean, being developed by Leonardo de Moura at Microsoft (it is an open source project):

There are slides describing it.  It is in its early stages, and not yet fully functional, but I am excited about it. We are aiming for a public “release” early next year.

As indicated in many of the publications just listed, progress is needed before interactive theorem provers are commonly used, though I am absolutely certain it will eventually happen. This includes things like developing better user interfaces, automated support, and libraries. A student of mine, Rob Lewis, and I are working on a heuristic method of proving real-valued inequalities, described here:

Jeremy Avigad, Robert Y. Lewis, Cody Roux, 2013, “A heuristic prover for real inequalities,” arXiv:1404.4410 and LNCS 8558 (2014)

Many others are developing other types of automation, both for the purposes of supporting the verification of mathematical proof, and for supporting the verification of hardware and software.

I apologize for over-emphasizing my own projects; there is a tremendous amount of work being done in the area of now, and mine is just a small part of it. I once heard Natarjan Shankar say that this is “the golden age of metamathematics,” and I agree.  This is a really exciting time to be working with formal methods.

Edward Nelson, 1932-2014

Just found out that Edward Nelson died last month.

SotFoM II: Competing Foundations

The second installment of SotFom (Symposium on the Foundations of Math) is asking for papers by Halloween:

FINAL CFP and *EXTENDED DEADLINE*: SoTFoM II `Competing Foundations?’, 12-13 January 2015, London.

FINAL CFP and *EXTENDED DEADLINE*: SoTFoM II `Competing Foundations?’, 12-13 January 2015, London.

The focus of this conference is on different approaches to the foundations of mathematics. The interaction between set-theoretic and category-theoretic foundations has had significant philosophical impact, and represents a shift in attitudes towards the philosophy of mathematics. This conference will bring together leading scholars in these areas to showcase contemporary philosophical research on different approaches to the foundations of mathematics. To accomplish this, the conference has the following general aims and objectives. First, to bring to a wider philosophical audience the different approaches that one can take to the foundations of mathematics. Second, to elucidate the pressing issues of meaning and truth that turn on these different approaches. And third, to address philosophical questions concerning the need for a foundation of mathematics, and whether or not either of these approaches can provide the necessary foundation.

Date and Venue: 12-13 January 2015 – Birkbeck College, University of London.

Confirmed Speakers: Sy David Friedman (Kurt Goedel Research Center, Vienna), Victoria Gitman (CUNY), James Ladyman (Bristol), Toby Meadows (Aberdeen).

Call for Papers: We welcome submissions from scholars (in particular, young scholars, i.e. early career researchers or post-graduate students) on any area of the foundations of mathematics (broadly construed). While we welcome submissions from all areas concerned with foundations, particularly desired are submissions that address the role of and compare different foundational approaches. Applicants should prepare an extended abstract (maximum 1,500 words) for blind review, and send it to sotfom [at] gmail [dot] com, with subject `SOTFOM II Submission’.

Submission Deadline: 31 October 2014

Notification of Acceptance: Late November 2014

Scientific Committee: Philip Welch (University of Bristol), Sy-David Friedman (Kurt Goedel Research Center), Ian Rumfitt (University of Birmigham), Carolin Antos-Kuby (Kurt Goedel Research Center), John Wigglesworth (London School of Economics), Claudio Ternullo (Kurt Goedel Research Center), Neil Barton (Birkbeck College), Chris Scambler (Birkbeck College), Jonathan Payne (Institute of Philosophy), Andrea Sereni (Universita  Vita-Salute S. Raffaele), Giorgio Venturi (CLE, Universidade Estadual de Campinas)

Organisers: Sy-David Friedman (Kurt Goedel Research Center), John Wigglesworth (London School of Economics), Claudio Ternullo (Kurt Goedel Research Center), Neil Barton (Birkbeck College), Carolin Antos-Kuby (Kurt Goedel Research Center)

Conference Website: sotfom [dot] wordpress [dot] com

Further Inquiries: please contact
Carolin Antos-Kuby (carolin [dot] antos-kuby [at] univie [dot] ac [dot] at)
Neil Barton (bartonna [at] gmail [dot] com)
Claudio Ternullo (ternulc7 [at] univie [dot] ac [dot] at)
John Wigglesworth (jmwigglesworth [at] gmail [dot] com)

The conference is generously supported by the Mind Association, British Logic Colloquium, and Birkbeck College.

Cambridge Graduate Conference on the Philosophy of Mathematics and Logic

If you’re in that part of the world (or will be in January), you might be interested to know that registration for the 8th Annual Cambridge Graduate Conference on the Philosophy of Mathematics and Logic (17-18 January 2015) is now open:

The conference will be held in St. John’s College, Cambridge. There will be two keynote speakers and six talks from graduate students on a variety of topics in the Philosophy of Mathematics and Logic, broadly construed. The graduate papers will have respondents, and the talks will be followed by open discussion. The keynote speakers this year are Prof Alan Weir (Glasgow) and Dr Mary Leng (York)To register for the conference please click here. Registration will close on Monday the 1st of December 2014.

For more information please see the conference website:

Putnam Blogging on Tarski on Truth

Hilary Putnam is writing on Tarski’s theory of truth (and Field’s analysis of it) at Sardonic Comment. First two blog posts are up:

Steve Awodey gives inaugural Calgary Mathematics & Philosophy Lecture

I’m very excited that Steve Awodey is on his way here to deliver the first Calgary Mathematics & Philosophy Lecture tomorrow! He’s speaking on “Univalence as a New Principle of Logic.” If you’re in Calgary, you should come.  It’ll be exciting. Thursday, 3:30 pm, in Engineering Building A aka ENA 101 on the UofC campus. Here’s the abstract:

It is often convenient or useful in mathematics to treat isomorphic structures as the same.  The Univalence Axiom for the foundations of mathematics elevates this idea to a foundational principle in the setting of Homotopy Type Theory.  It states, roughly, that isomorphic structures can be identified.  In his talk, Prof. Awodey will explain this principle and how it can be taken as an axiom, and explore the motivations and consequences, both mathematical and philosophical, of making such an assumption.

Steve will give a more technical talk in the Math Department on Friday at 2pm.

Also check out the sweet poster: