Home > Books and culture > What I’ve been reading

What I’ve been reading

April 3rd, 2005

I’ve been badly overstretched for the last few months, with an excessive amount of travel, and one symptom has been a failure to keep up with regular features like this one. I’m gradually getting my life under some control, so I thought I’d try to restart some conversations about books and writing. I have a huge pile of books on my desk, some of which I’ve promised to review. In this category, theres“Freakonomics : A Rogue Economist Explores the Hidden Side of Everything” (Steven D. Levitt, Stephen J. Dubner), “In Defense of Globalization” (Jagdish Bhagwati), and Diversity in Development: Reconsidering the Washington Consensus.

I’ve also become interested in philosophical issues relating to causality, which are closely linked to my concerns about uncertainty (in a world with no uncertainty, causality is essentially trivial). Books I’ve found helpful include“Reasoning about Uncertainty” (Joseph Y. Halpern)“Facing the Future: Agents and Choices in Our Indeterminist World” (Nuel D. Belnap, Michael Perloff, Ming Xu, Nuel Belnap)“Causation and Counterfactuals (Representation and Mind)” (The MIT Press) and“Causality : Models, Reasoning, and Inference” (Judea Pearl)

On the leisure front, I’ve been enjoying Iain Banks series of novels about The Culture, most recently Excession

I hope to extend this post during the week, adding a range of comments on these books over the fold. But feel free to jump in first with your thoughts, recommendations for further reading etc.

Categories: Books and culture Tags:
  1. harry clarke
    April 3rd, 2005 at 20:27 | #1

    I don’t have comments re the books you have read.

    My own recent expedition over recent weeks has been into older works of James Joyce (namely Ulysses and Finnegans Wake) which I have, in the past, found to be difficult and/or almost incomprehensible.

    My crib, or aid, has been the, just released, excellent NAXOS cd’s in which large slabs of these books are read by Irish actors Jim Norton and Marcella Riordan. They are superb and more than just ‘bleeding chunks’.

    Hearing these classics read by Irish people helps understanding and enjoyment of these novels of perhaps flawed genius.

    Anthony Burgess describes Ulysses as the “greatest novel of the century”. Is that right? I don’t know but the novel, when read aloud, is penetrating, fiendishly accurate, though fairly straight (internal) narrative. It is just a delight — pure pleasure. Finnegans Wake, the comic masterpiece, more difficult even when spoken, and until I get the train of thought enough to relax and enjoy what seem at times to be jaberwocky ramblings I will leave assessing it to the literary ‘experts’.

    Also planning to reread “Portrait of Artist as a Young Man” (available in full on NAXOS though I have not listened to it) and “Dubliners” (I first read over 20 years ago) and, to complete the Ellmam’s biography of James Joyce (Burgess also describes this as the “greatest literary biography of the century”). Joyce was an admirable if imperfect man as well as mindblasting writer.

    This should keep me going for a year or so. What I really wanted to emphasise is that I probably would not have got started were it not for the spoken word efforts by NAXOS.

  2. Homer Paxton
    April 3rd, 2005 at 20:41 | #2

    JQ where do you find the time to read all this this stuff, work, be a father and husband and blog?

    I am flat out reading the classic biography of Hitler again!

  3. Peter McBurney
    April 3rd, 2005 at 21:45 | #3

    Nice to see an economist reading AI literature, John (Halpern, Pearl). Halpern has been at the forefront of logical models of uncertain reasoning, useful if computers are to reason coherently about real-world domains. The early expert systems (eg, MYCIN) arguably reasoned incoherently about uncertainty, since the methods used to represent and reason about uncertainty were not well-founded theoretically.

    Pearl’s main work in AI has been on Causal Influence Diagrams (aka Bayesian Belief Networks, aka Bayesnets). What this book of his lacks, IMO, is any well-founded theory of causality. This is not surprising, since statisticians have, as a species, been suspicious of cause since the mis-identification of the causes of cholera in the mid-nineteenth century. (His book is also, to my taste, badly structured. The chapters are mainly previously-published papers, and they are presented in the order they were published. The book would work better if the chapter order was reversed.)

    Pearl’s book will tell you how to model causal effects, but not what cause is. A better guide to theories of causality is this introductory volume:

    Ernest Sosa and Michael Tooley (Editors): Causation. Oxford Readings in Philosophy. OUP, 1993.

    And, as a tangent, you may not be aware that Judea Pearl is the father of Daniel Pearl, Wall Street Journal journalist killed by supporters of the Taliban in Pakistan or Afghanistan after 9/11.

  4. April 3rd, 2005 at 22:30 | #4

    Overstretched…. you lead three lives to my one!

  5. Bill Cushing
    April 4th, 2005 at 00:01 | #5

    Do you ever sleep, JQ?

    Me–I’ve started (again) on Proust. As time now permits. Dipped a madeleine in my Lygon St. coffee on Saturday morning.

  6. Peter McBurney
    April 4th, 2005 at 03:00 | #6

    “(in a world with no uncertainty, causality is essentially trivial).”

    Is this true? There are many reasons for uncertainty, and not all of these relate to causal determinism. For example, we can be uncertain about the outcome of some process because:

    – The process is nondeterministic (ie, there is a random element in the process).

    – The process is entirely deterministic but complex, and we lack knowledge of the different causes, and/or of their precise interactions, and/or of their initial conditions.

    – We do not have perfect knowledge of the outcome, even if it is deterministic, for instance, for reasons related to imprecision of measurement.

    – Our conceptual model of the world and/or our language are unable to represent the outcome, or not able to represent it with the appropriate degree of precision.

    Only if we had no uncertainty in the sense of the first case here would I think that causality might be trivial. In the other cases, even eliminating uncertainty could still leave use with non-trivial causality.

  7. John Quiggin
    April 4th, 2005 at 08:15 | #7

    I’d put it the other way around, Peter. If any of these sources of uncertainty is present, causality is non-trivial. If none are present, causality is trivial.

  8. Paul Norton
    April 4th, 2005 at 11:05 | #8

    I’m re-reading Alec Nove’s The Economics of Feasible Socialism, after a break of 16 years.

  9. April 4th, 2005 at 11:08 | #9

    Unless you define uncertainty tautologously, how can you state that causality is trivial if uncertainty is absent? We can construct deterministic “toy universes”, e.g. certain cellular automata, to bring out these points. If nothing else, we can easily get hidden variables from a surfeit of information. Chaitin’s Omega exists, in a certain sense, which means that everything is in a certain sense knowable; but the problems just move around. And so on.

  10. stephen bartos
    April 4th, 2005 at 17:58 | #10

    john, I’m surprised you’re not reading The Algebraist given your fame as an algebraist. though to be honest I don’t think it one of Bank’s best, and his earlier culture novels (eg player of games, use of weapons)were much better.

  11. John Quiggin
    April 5th, 2005 at 13:36 | #11

    I plan to get to it, Stepen, but I’ve heard other negative reports, and I thought I’d try to read the novels in something like chronological order. I liked Player of Games a lot.

  12. Stephen Hill
    April 8th, 2005 at 10:56 | #12

    Herman Melville’s “Pierre or the Ambiguities” at the moment.

    Bank’s “Walking on Glass” wasn’t bad as I recall

  13. September 23rd, 2005 at 23:00 | #13

    HOUSTON – Hurricane Rita roared toward the Texas and Louisiana coasts early Friday, a major Category 4 storm that spurred a traffic-snarled exodus toward higher ground and fears it could cripple the heart of the nation’s petrochemical industry.
    oem Software

Comments are closed.