Volume 4, Number 9September 2010

www.thereasoner.orgISSN 1757-0522

Contents

§1 Editorial 133

§2 Features 134

§3 News 137

§4 What’s Hot in . . . 143

§5 Events 144

§6 Courses and Programmes 147

§7 Jobs and Studentships 148

§1Editorial

I am delighted to return as guest editor of The Rea-soner. Exactly one year ago, I opened the September is-sue with an interview with Theo Kuipers from the Uni-versity of Groningen. In our conversation, we touchedupon a number of topics, in particular scientific real-ism, progress, truth, verisimilitude, and the method ofphilosophy of science. Nearly the same topics are thecentral ones of this month’s interview. This is due notonly to my lack of imagination, but also to the fact that,in my view, these are some of the most important prob-lems in the philosophy of science. For this reason, I

decided to ask Professor Ilkka Niiniluoto for an inter-view. During the last forty years, Niiniluoto has pub-lished an impressive number of papers on virtually allfields of contemporary philosophy of science, focussingparticularly on probability and inductive logic, truth andverisimilitude (also known as truthlikeness or truth ap-proximation), realism and scientific progress, and theevaluation of theories within natural and social sciences.Let me briefly survey his main contributions, which ex-plain why he is the right person to answer my questions.

Niiniluoto is presently Chancellor of the Univer-sity of Helsinki, where he teaches Theoretical Phi-losophy, and Chairman of the Philosophical Societyof Finland. He is one of the mostprominent figures of the philosoph-ical school going back to EinoKaila, Georg H. von Wright, andJaakko Hintikka. The developmentof this tradition has been recentlyanalyzed in a volume, AnalyticPhilosophy in Finland (Rodopi,2003), edited by Niiniluoto himself(together with Leila Haaparanta),for which Niiniluoto wrote also theintroductory, historical chapter. Within this tradition,the Finnish School of inductive logic takes the lion’sshare. Niiniluoto’s first important contribution to theSchool’s research programme is Theoretical Conceptsand Hypothetico-Inductive Inference (Reidel, 1973),written together with Raimo Tuomela, in which the au-thors use Hintikka’s inductive logic in order to defendcritical scientific realism. A few years later, Niiniluotoand Tuomela also edited The Logic and Epistemology

133

of Scientific Change (Acta Philosophica Fennica, 30,1979), a collection of papers devoted to four very recent(at that time) methodological research programmes:structuralism, cognitive decision theory, verisimilitude,and the logical theory of belief change.

Sometimes, errors do trigger progress in the historyof ideas. It is well-known that Popper’s attempt (in Con-jectures and Refutations, 1963) to explicate the notionof verisimilitude (construed as similarity or closenessto the comprehensive truth about a target domain) wastechnically flawed, as Pavel Tichy and David Miller in-dependently proved in 1974. This failure opened theway to the post-Popperian theories of verisimilitude,which constitute a lively research programme in formalphilosophy of science. The most developed and well-known theory of verisimilitude is the so-called “simi-larity approach”, proposed since 1974 by Pavel Tichyand Risto Hilpinen, and subsequently developed by Ni-iniluoto, Tuomela and Graham Oddie. About a decadelater, Niiniluoto published two books: Is Science Pro-gressive? (Reidel, 1984), a collection of essays devotedto explicating scientific progress in terms of increasingverisimilitude, and Truthlikeness (Reidel, 1987), a pre-sentation of his own theory of verisimilitude, as well asa detailed discussion of the history, importance, and ap-plicability of this notion, and a defence against its crit-ics. Truthlikeness is often referred to as “the Bible ofverisimilitude”, since it contains virtually all you needto know for seriously studying the subject (in this case,you will also find useful Niiniluoto’s survey article on“Verisimilitude: The third period” in the British Journalfor Philosophy of Science, 49, 1998).

In his most recent book on these themes, CriticalScientific Realism (Oxford University Press, 1999), Ni-iniluoto offers a detailed and up-to-date presentationof his philosophical outlook. The book subtly com-bines a coherently fallibilist view of human knowledgewith an uncompromising defence of realism in ontol-ogy, semantics, epistemology, theory construction, andmethodology. Niiniluoto carefully distinguishes all themain forms of realism, antirealism, and relativism dis-cussed in the recent literature, so that Critical Scien-tific Realism can be read also as a high-level textbook,containing one of the most learned and complete expo-sitions of the subject to date. The reader interested inthe recent debate on Niiniluoto’s philosophy of scienceshould read Approaching Truth (College Publications,2007), a Festschrift edited by Sami Pihlstrom, PanuRaatikainen, and Matti Sintonen, collecting a number ofpapers by leading scholars devoted to his work in threewide areas: 1) philosophy of logic, of language, and ofmathematics; 2) induction, truthlikeness, and scientificprogress; and 3) epistemology, culture, and religion.

Before starting the interview, I wish to thank Feder-ica Russo, Jon Williamson, and Lorenzo Casini for theirinvitation to open this issue and for their editorial work;

and Roberto Festa and Luca Tambolo for many conver-sations on the topics of the interview, which inspiredmost of the questions below.

Gustavo CevolaniPhilosophy, Bologna

§2Features

Interview with Ilkka Niiniluoto

Gustavo Cevolani: First of all, thank you for agreeingto be this month’s interviewee. As usual, let me startby asking you about your intellectual history. How didyou first get into research in logic and philosophy ofscience? Who had the greatest influence on your philo-sophical career?

Ilkka Niiniluoto: I did my Master’s degree in math-ematics in 1968, specializing in probability theory andBayesian statistics with ProfessorGustav Elfving. At the same time,I had started to study philosophyand mathematical logic with Pro-fessor Oiva Ketonen. The philo-sophical devotion and personal in-tegrity of Academician Georg Hen-rik von Wright made a strong im-pression on me. I had already de-cided to move from mathematics totheoretical philosophy when Professor Jaakko Hintikkaappointed me his research assistant in the summer of1971. Hintikka’s distributive normal forms and his mea-sures of inductive probability and semantic informationprovided me the basic tools for the study of theoreticalterms and inductive inference. The most inspiring andvaluable lessons in philosophical methodology I havelearned from Hintikka.

GC: You were trained in the tradition of the “Finnishschool” of inductive logic, pursuing the research inau-gurated by Eino Kaila, von Wright and Hintikka. Inretrospective, what have been the most important con-tributions of this school? How lively is this traditiontoday?

IN: Von Wright’s treatment of eliminative induc-tion was an important contribution, but it is not muchdiscussed today. The greatest achievement of theFinnish school was Hintikka’s system of inductive gen-eralization which improved Carnap’s inductive logicby showing how universal statements in infinite do-mains can receive non-zero probabilities. Hintikka’sstudents—including Risto Hilpinen, Raimo Tuomela,Juhani Pietarinen and myself—then developed and ap-plied this insight in various directions. My own work

134

on truthlikeness and its estimation is also a continua-tion of this tradition. Hintikka himself downplayed therole of induction in his interrogative model of inquiryin the 1990s. It is a little disappointing to see thatHintikka’s achievement is often ignored by Bayesianscholars who either reject inductive logic or work withinthe Carnapian framework. The models of induction inArtificial Intelligence are less sophisticated than Hin-tikka’s system. But I am happy that there are philoso-phers in other countries—among them Theo Kuipersand Roberto Festa—who have made progress along thelines of the Finnish school.

GC: The post-Popperian research programme ontruthlikeness (aka verisimilitude, or truth approxima-tion) is a distinguished approach to the central prob-lems of contemporary philosophy of science. Your 1987book, Truthlikeness, is a milestone of this programme,exploring both the logical definition of verisimilitudeand its methodological applications. However, the en-tire programme failed, at least until now, to gain wideacceptance and visibility among philosophers of sci-ence. First, truthlikeness is still often conflated withdifferent concepts like probability, approximate truth,partial truth, and so on. Second, its role for the axi-ology and the methodology of science is largely under-estimated and sometimes plainly ignored. What are thereasons of this situation, in your opinion?

IN: Truthlikeness is a fascinating topic which haskept me active already for 36 years, and there is stillmuch research to be done in this area. The commu-nity of logicians working seriously on this theme is rel-atively small, and there is a lot of disagreement aboutthe right approach. Many philosophers who are fondof the realist idea of truth approximation have foundthe logical treatments of truthlikeness too technical or“Carnapian”—Popper himself never gave me any ref-erence or personal communication, even though I suc-ceeded to save the notion of verisimilitude with an ex-plication that satisfies all the central Popperian desider-ata. As there is no consensus so far about the basicprinciples of truthlikeness, it may be difficult for otherphilosophers of science to appreciate the important con-ceptual distinctions, so that they simply work with anintuitive notion of “approximate truth”. It is also easyfor them to ignore the potential of the precise conceptof truthlikeness for wider issues in the philosophy ofscience. The situation is different from the role of prob-ability: in spite of various interpretations, there is anaccepted standard mathematical explication of this no-tion. On the other hand, there are also hot disputesabout many other important concepts in the philosophyof science—such as theory, model, truth, confirmation,lawlikeness, explanation, and reduction.

GC: Verisimilitude plays a crucial role in your ownversion of “critical scientific realism”. You have beendefending scientific realism since the beginning of your

career, in the early Seventies. At that time, anti-realism(in its instrumentalist version) was widespread and be-came very fashionable shortly after, with the publica-tion of The Scientific Image by Bas van Fraassen (1980).Today, the trend may appear to be reversed. Last year,the PhilPapers website organized a survey on a numberof central philosophical questions. Among 1800 pro-fessionals (PhDs or faculty members), over 70% is re-ported to “accept or lean toward” scientific realism (al-though the figure falls toward 50% among those broadlyspecialized in philosophy of science). How do youjudge the results of this poll and, more generally, thepresent state of the realism/anti-realism debate?

IN: I am glad to hear about this relative success ofscientific realism. Arthur Fine was wrong when he an-nounced the death of realism some twenty-five yearsago. Of course one should remember, as I tried to showin my Critical Scientific Realism (1999), that there aremany interesting forms of realism and anti-realism. Onecan reliably predict that this debate will always be a vi-tal issue in the philosophy of science. During the lastdecade, structural realism has gained popularity, and in-ternal realism has lost ground. But my guess is alsothat “leaning toward scientific realism” is quite commonamong those scholars who are working within the phi-losophy of special sciences: they have to rely on somesort of realist interpretation when they take a seriouslook at the contents of the best theories in physics, biol-ogy, medicine, psychology, economics, and sociology.

GC: The survey mentioned above reports a slight ma-jority (56%) of philosophers leaning toward moral real-ism, rising to over 60% among the specialists of nor-mative ethics and meta-ethics. In your Critical Scien-tific Realism, you defend realism in ontology, seman-tics, epistemology and methodology but accept (a formof) moral relativism. Can you elaborate a bit on thispoint? In particular, if “moral facts” are human-madeand then relative to time, culture, etc., what is the dif-ference between them and other human artefacts?

IN: I have indeed defended “moral constructivism”which treats moral values and principles as human-made social artifacts. Moral facts differ from someother human artifacts by their Durkheimian coerciveforce—their power in guiding our actions and decisions.In this sense, morality is a “real” phenomenon in thePopperian World 3. It is an extremely important aspectof our life and social interaction, but it has no transcen-dent (religious or metaphysical) ground independentlyof human activity. Morality cannot be reduced to natu-ral facts about human needs or evolution, either: indi-vidually and socially, we are free to critically reconsiderand renew the moral standards prevailing in our com-munity. This sort of modest relativism is compatiblewith human responsibility, tolerance towards differentvalue systems, and attempts to reach world-wide agree-ments on human rights.

135

GC: If you had to suggest a direction of research toyoung philosophers of science starting out today, whatare the topics that you would recommend?

IN: There are still important and largely open prob-lems in inductive logic: inductive reasoning with ob-servational errors and inductive systems with relationalpredicates. The connection between truth approxima-tion and belief revision models is a promising area. An-other up to date topic is the analysis of abductive infer-ence in terms of probability and truthlikeness. A younglogician could also spend some time in going throughthe related work in the field of Artificial Intelligence.Illumination of the key ideas of scientific realism in his-torical case studies would also give opportunities for in-teresting research projects.

GC: A question of a more general nature. In the Pref-ace to your Festschrift, the reader learns that, as “one ofthe most prominent public intellectuals in Finland”, youhave “constantly defended science and reason”. Whichare, in your opinion, the worst enemies of reason today?Don’t you think that a middle course is needed betweentwo equally dangerous extremes, the “Scylla” of rela-tivism and the “Charybdis” of scientism?

IN: In my view, the most dangerous enemies of rea-son come from circles that base their beliefs and prac-tices upon irrational faith and superstition with a hos-tile attitude towards scientific inquiry. Radical forms ofpostmodern relativism may give support to such com-munities. The reliance on science is not as such danger-ous, since science itself is the critical way of formingbeliefs about the world on the basis of public evidence.The mistake of scientism lies in its overstatement of thescope of scientific reason: even though scientific knowl-edge is relevant for most urgent decision problems, thescientists have no monopoly for solving value questionswithin a free democratic society.

GC: Regarding scientism, sometimes one is underthe impression that large parts of the scientific com-munity don’t practice what they preach. On the onehand, scientists adopt a Popperian jargon, willingly as-senting to the idea that theories are always conjecturaland open to revision. On the other hand, they becomevery touchy when their pet theories are challenged, andseem often motivated by ideological, political or eco-nomic reasons. In this connection, let me mention tworecent episodes. The first is the publication of Fodor andPiattelli-Palmarini’s What Darwin Got Wrong (London,2010), which is giving rise to much debate also in mycountry (one of the authors is Italian). The secondis the so-called Climategate, concerning some hackeddocuments seemingly showing an attempt to minimizeor conceal evidence about climate change. In bothcases (admittedly very different in nature), some sci-entists violently reacted in defence of a purported sci-entific consensus about, respectively, (neo-)Darwinianevolutionism and anthropogenic global warming. In

your opinion, is there a danger of scientific dogmatism?What have philosophers to say, and to do, about theseepisodes?

IN: As a critical scientific realist, I am a fallibilistwho endorses the corrigibility of all human knowledge.Scientific dogmatism is harmful, since it is in conflictwith the self-corrective nature of science and inhibitsscientific progress. But scientists themselves should beable to estimate the reliability of their tentative conclu-sions. When the scientific community reaches a con-sensus on some question, open criticism should still beallowed, but naturally a change in the prevailing opin-ion would need strong scientific counter-arguments andnew evidence. As experts of critical thinking, philoso-phers should assist in such episodes by assessing theweight of the rival arguments and positions.

Tempus Dictum

Technological Aids to Cognitionhttp://tempusdictum.com

Kaplan on indexical logicIt is a simple matter of grammar that it is predicatesof ‘that’-clauses, rather than predicates of mentionedsentences, that are equivalent to operators on used sen-tences (see 2010a: “On Forgetting ‘that”’, The Rea-soner 4.4, 57-8). So ‘It is true/necessary that I am herenow’, for instance, in which there are operators ‘It istrue/necessary that’ on a used sentence, is equivalentto ‘That I am here now is true/necessary’, and not “‘Iam here now” is T/N’ for any predicates ‘T’ or ‘N’,of the now mentioned sentence ‘I am here now’. Thedifficulty for the main line logical tradition on this is-sue has been that there is no term forming element (like‘that’) in standard formal languages transforming a sen-tence into a nominal phrase referring to the propositionthe sentence expresses, when used. And this has led toconsiderable confusion, through attempts to make pred-icates of mentioned sentences do the job of predicates of‘that’-clauses (see 2010b: ‘What Priest (amongst manyothers) has been missing’, Ratio XXIII.2, 184-198).

The specific example of this confusion just givenarises in the work of David Kaplan. For one significant,repeated assertion of Kaplan’s is that it is the content ofa sentence (i.e., the proposition expressed) that carriesthe truth-value. Thus we find, for example, (Almog, J.,Perry, J., and Wettstein, H. (eds) 1989: Themes fromKaplan, O.U.P., Oxford, 500):

What is said in using a given indexical in dif-ferent contexts may be different. Thus if I say,today, ‘I was insulted yesterday’, and you ut-ter the same words tomorrow, what is said isdifferent. If what we say differs in truth-value,

136

that is enough to show that we say differentthings. But even if the truth-values were thesame, it is clear that there are possible circum-stances in which what I said would be true butwhat you said would be false. Thus we saydifferent things.

But, on the other hand, in his discussion of the sentence‘I am here now’, Kaplan talks about (mentioned) sen-tences being true. He says (Almog et al 1989: 508-9):

Consider the sentence (6) I am here now. . . Intuitively, (6) is deeply, and in somesense, which we shall shortly make precise,universally true. One need only understandthe meaning of (6) to know that it cannot beuttered falsely. . . . Let the class of indices benarrowed to include only the proper ones—namely those (w, x, p, t) such that in the worldw, x is located at p at the time t. Our reformhas the consequence that (6) comes out, cor-rectly, to be logically true. Now consider (8) I am here now. Since the contained sen-tence (namely (6)) is true at every proper in-dex, (8) also is true at every proper index andthus also is logically true. . . . But (8) shouldnot be logically true, since it is false. It is cer-tainly not necessary that I be here now. . . .

Because of his point here, Kaplan has claimed thatthe rule of Necessitation fails in the presence ofindexicals—the rule of Necessitation being commonlywritten as the inference from ‘`p’ to ‘`Lp’, where ‘L’ isthe operator ‘It is necessary that’.

Kaplan’s confusion here can be pinpointed by focus-ing on what he said when using the sentence ‘It is cer-tainly not necessary that I be here now’. For he did notassert then that the sentence ‘I am here now’ is not nec-essary, but instead that the proposition that he (in par-ticular) was where he was at the time in question wasnot necessary. But is the reverse of this what (8) says?

The difficulty of seeing clearly what (8) says is nothelped by the omission of quotation marks after ‘(6)’and ‘(8)’ in the text. That is a common convention inthe formal logic tradition, but in the present context itonly helps to blur the needed distinction between talkabout words and talk about what they mean. Specifi-cally, if (6) is the sentence ‘I am here now’ then is itintended that (8) be ‘ I am here now’ with ‘’ the op-erator on used sentences ‘It is necessary that’? So (8)has the form ‘Lp’? Or is (8) intended to be ‘ “I amhere now”’, with ‘’ some predicate of mentioned sen-tences? It cannot be the former that is intended, since atno proper index does it then express a proposition thatis true. So is ‘’ the predicate of mentioned sentences‘is valid’ that Kaplan later defines as ‘in every [proper]context expresses a proposition that is true’ (Almog et

al. 1989: 596)? So (8) has the form ‘V‘p”? But why,then, should (8) not express a logical truth? The sen-tence ‘I am here now’ does express a true proposition inevery proper context—by the definition of ‘proper con-text’.

So does Necessitation fail in the presence of index-icals? As before, Necessitation is the inference com-monly written as above; but that is only because of theconvention about omitting quotes, since ‘`’ (unlike ‘L’)is a meta-linguistic symbol. So the rule would be betterput as the inference from ‘`‘p” to ‘`‘Lp”. Now cer-tainly if ‘`’ was Kaplan’s ‘is valid’ then the inferencewould be invalid, if ‘p’ was ‘I am here now’. But ‘`’ isnot ‘is valid’, since the former is not confined to propercontexts, and even Kaplan acknowledges that the lattermust be distinguished from ‘it is necessary that’ (Al-mog et al. 1989: 596). Thus (6) is not ‘logically true’ asKaplan claims above. Obviously ‘Necessitation’ doesnot hold if its premise is re-interpreted as not showingsomething is necessary! In fact it is false both that `‘p’,and that `‘Lp’, for the given ‘p’. So Necessitation doesnot fail in the presence of indexicals.

Hartley SlaterPhilosophy, University of Western Australia

§3News

Degrees of Belief vs Belief, 14–15 May

Here is a summary of the invited talks of the workshopDegrees of Belief vs Belief, which took place on 14th-15th May and was hosted by the University of Stirling.

David Christensen (Brown): “Rational Reflection”.The main question of the paper concerned the relation-ship between what it is rational for one to believe andwhat it is rational for one to believe about what it is ra-tional for one to believe. Christensen approached theissue mainly in the context of graded belief. He formu-lated a rational reflection principle: Cr (A/Pr(A)=n)=n,where Cr stands for the agents credences, and Pr standsfor the credences that would be maximally rational forsomeone in that agents epistemic situation. Christensenshowed how this prima facie plausible principle leadsto some puzzling results and explored a number of dif-ferent reactions to the puzzling cases.

Peter Milne (Stirling): “Belief, Degrees of Belief,and Assertion”. On the basis of an in-depth discus-sion of the role of assertion and the relationship be-tween assertion and belief, Milne went on to discusstwo puzzling issues: The first concerned the assertionof indicative conditionals. Much recent research in thepsychology of reasoning supports two theses: that peo-

137

ple ascribe to the indicative conditional the so-calleddefective truth-table in which an indicative conditionalwith false antecedent is deemed neither true nor false;and that people assign as probability to an indicativeconditional, the (conditional) probability of consequentconditional upon antecedent. The second issue Milnediscussed concerned the relationship between assertionand degrees of belief. Certain well known cases sug-gest that a high degree of belief is neither necessary norsufficient for a willingness to assert.

Brian Weatherson (Rutgers/Arche): “Rational Be-lief and Rational Action”. Weatherson discussed theprinciple, defended by Jeremy Fantl and Matthew Mc-Grath, that if S’s belief that p is justified, then it is per-missible for S to use p as a premise in practical rea-soning. Weatherson argued that the principle is false.In cases where the agent has some rational beliefs andsome irrational beliefs, it might be that justification isinsufficient to ground action. Weatherson went on todiscuss a novel suggestion regarding the relationshipbetween rational beliefs and rational degrees of belief.

Alan Hajek (ANU): “Staying Regular”. Hajek firstreviewed several arguments for regularity as a norm ofrationality. There are two ways an agent could violatethis norm: by assigning probability zero to some dox-astic possibility, and by failing to assign probability al-together to some doxastic possibility. Williamson andEaswaran have argued forcefully that the former kindof violation may be unavoidable. Hajek’s discussionfocused on violations of the second kind. Both kindsof violations of regularity have serious consequencesfor traditional Bayesian epistemology, in particular con-ditional probability, conditionalization, probabilistic in-dependence and decision theory.

Carl Hoefer (Barcelona): “Connecting Objectiveand Subjective probability: How to Justify the Princi-pal Principle”. Hoefer focused on David Lewis’ Prin-cipal Principle and the claim that “Truth is to belief asagreement with objective chance is to degree of belief”.The plausibility and defensibility of these theses de-pend, clearly, on how we understand objective chance.The paper introduced an account of objective chance:a Humean/reductive account closely related to Lewis’own. One of the key virtues of this account of chance isthat it allows us to demonstrate that the PP is indeed arequirement of rationality. Hoefer also argued for a sec-ond thesis on which agreement with actual frequencies,rather than chances, serves as the core virtue for degreesof belief.

Branden Fitelson (Berkeley) & Kenny Easwaran(USC) “Partial Belief, Full Belief, and Accuracy-Dominance”. The paper had two main aims: (1)to make some (cautionary) remarks about the set-upand interpretation of some recent accuracy-dominancebased arguments for probabilism (with respect to par-tial beliefs), and (2) to discuss some interesting (for-

mal and informal) analogies (and disanalogies) betweenpartial belief and full belief, when it comes to the phe-nomenon of accuracy-dominance. In particular, Fitel-son and Easwaran discussed the case of an extremalagent (who can only assign degrees of belief 0 or 1) andshowed that there are non-probabilistic extremal func-tions that are not even weakly dominated by any proba-bilistic extremal function (using the Brier score).

Philip EbertDepartment of Philosophy, University of Stirling

Martin SmithDepartment of Philosophy, University of Glasgow

Logic and Knowledge, 16–19 June

The conference, reminiscent of Russell’s classical bookLogic and Knowledge, took place from June 16 to 19in the Faculty of Philosophy at La Sapienza Univer-sity of Rome. It was organized by Emiliano Ippoliti,Carlo Cellucci, Emily Grosholz, and aimed to explorethe connection between logic and knowledge, in partic-ular the interconnections among epistemology, philoso-phy of logic and philosophy of mathematics, in order toprovide a more comprehensive and articulated view ofthe topics dealt with.

The cooperation between speakers and discussants,as well as the depth of the discussions, allowed the con-ference to fulfilling its major aim, i.e. to offer new viewpoints on logic, epistemology and their relations.

The conference was opened by Donald Gillies’ (Uni-versity College London) talk on the empiricist view ofmathematics. Gillies, criticizing Quine, suggested thatquantum logic does not give a decisive argument infavour of this view, but that such an argument is pro-vided by the successful application of non-classical log-ics in Artificial Intelligence.

Mario De Caro (University of Roma Tre) in his talkargued for the conclusion that if we have reasons for be-lieving that neurobiology can enrich our understandingof the features of the human mind, there is no soundreason for thinking it will ever explain them all, and hediscussed the case of free will to support his conclusion.

After the lunch break, Michael Detlefsen (Universityof Notre Dame), in his talk about rigor, logic and intu-ition considered various conceptions of rigor, the bene-fits of rigor so conceived, or supposed to be, what roleslogical reasoning has been taken to play in the attain-ment of rigor and whether and/or under what conditionsit may indeed serve in such a roles.

Goran Sundholm (University of Leiden) in his paperoffered a comparison between the tree-like representa-tions for the grounding of knowledge and of truths that

138

are offered by Frege and Bolzano.Carl Posy (The Hebrew University of Jerusalem)—

starting from David Hilbert’s famous lecture “On theInfinite” which invokes Kant’s philosophy of mathemat-ics to endorse the need for intuitive finitary reasoning inmathematics—discussed Kant’s analysis of the notionof finite grasp and the issues that arise when we mixcertain ideal ideas of reason into the realm of the empir-ically real.

The second day was opened by RiccardoChiaradonna (University of Roma Tre) talk aboutGalen’s medical epistemology. Chiaradonna fo-cused on Galen’s views about the epistemic status ofmedicine, pointing out that Galen argues that demon-strative methods have an intrinsic heuristic value and heseeks to transpose the ‘analytical’ geometrical methodof resolution of problems into the domain of medicine.

In her talk on the importance of sight and hearing in17th and 18th century logic, Mirella Capozzi (Univer-sity of Roma La Sapienza) pointed out that the relationbetween logic and sight was favored by many authorsof logical calculi. She also argued, however, that someof such authors believed that logic should investigatemethods of discovery by taking inspiration from the artsof discourse, so strictly connected to hearing and to thevague meanings of spoken words.

The morning session ended with the talk of Jan vonPlato, (University of Helsinki), on the dimensionality ofdeductive arguments. Von Plato showed through histor-ical examples from Aristotle, Hilbert and Bernays, andGentzen, that the tree form has decisive advantages overa linear arrangement.

Timothy Williamson (University of Oxford) consid-ered and questioned Michael Dummett’s contention, inThe Logical Basis of Metaphysics, that semantic theo-ries should be formulated in such a way that the logic ofthe object-language is maximally insensitive to the logicof the meta-language. Examining the status of the Bar-can formula and its converse in quantified modal logicas a case study, Williamson argued that the insensitiv-ity in question is a less desirable feature of a semantictheory than Dummett suggests.

Cesare Cozzo (University of Roma La Sapienza)discussed and questioned Williamson’s contention, inKnowledge and its Limits, that “knowing is the mostgeneral factive stative attitude, that which one has toa proposition if one has any factive stative attitude toit at all”. According to Cozzo, Williamson does notprove that his principle that factive-stative attitudes en-tail knowledge and he considered a counterexample inorder to show that. Emiliano Ippoliti (University ofRoma La Sapienza) ended the day with a talk aboutthe issue of ampliation of knowledge. He argued thatthe generation of knowledge requires revising the no-tions of inference, logic and knowledge from an infor-mational point of view, examining how information is

generated, extracted, processed and transferred. He dis-cussed the Feynman Integral Path as case study.

The third day was opened by Dag Prawitz (Universityof Stockholm), who argued for the need of the Aris-totelian distinction between perfect and imperfect syl-logisms, in order to say what a proof is and to explainhow inferences can generate knowledge, and he tried tooffer a systematic approach to the issue.

Reuben Hersh (University of New Mexico) startedhis talk playing with Jody Azzouni a dialogue betweena teacher of mathematics and a student and then he dis-cussed, in mathematical practice, the notion of mathe-matical intuition and the validity of heuristic and com-putational reasoning in mathematics—both pure and ap-plied.

After the lunch break, Carlo Cellucci (University ofRome La Sapienza), gave his talk about classifyingand justifying inferences. He offered a new classifi-cation and justification of inferences (deductive, non-deductive and abductive) which takes into account theirrole in knowledge. He argued that the justificationof deductive, non-deductive and abductive inferencesraises similar problems and is to be approached muchin the same way.

Emily Grosholz (The Pennsylvania State University)talk offered a philosophical critique of logic. In fact,she argued, logic must treat its terms as if they werehomogeneous, to exhibit valid inferences, but the kindsof representations that make successful reference possi-ble and those that make successful analysis possible inmathematics and the sciences often juxtapose heteroge-neous terms.

Robert Thomas (University of Manitoba), in his talkassimilated mathematics to science and discussed theoperation of assimilation. He pointed to assimilations inhistory that would not happen today, abandoned assim-ilations failing to make important distinctions, and cur-rent assimilations that are controversial. He suggestedthat the lack of assimilations in mathematical practice isan important reason for the dependability of arbitrarilylong chains of reasoning uniquely in mathematics.

The conference program closed on Saturday withtwo talks. Roberto Cordeschi (University of Roma LaSapienza), offered a discussion of Herbert Simon’s ap-proach to the study of rational choice, and he regarded itas more than a starting point for evaluating an integratedapproach to the study of behaviour.

The last talk was given by Jody Azzouni (Tufts Uni-versity), who criticized the idea that there is the waythe world is, and that such a way can be characterizedin purely nominalistic terms. He tried to argue that theindispensability of mathematics to the languages of thesciences shows that this is false and he tried to refutethe possibility that the indispensability of mathematicsscreens us off from what the world itself is like, show-ing that we can characterize aspects of the world that

139

are really out there.

Emiliano IppolitiFaculty of Philosophy, “La Sapienza” University of

Rome

Square of Opposition, 20–22 June

Two concepts and one famous name strikingly come toone’s mind when opposition is dealt with: incompati-bility and negation, on the one hand; Aristotle, on theother hand. Can there be more to be said about thewell-known theory of opposition, or is it just an old-fashioned by-product of logic for undergraduate stu-dents? In order to address this question, the philoso-pher and logician Jean-Yves Beziau created a periodicmeeting since 2007 (in Montreux, Switzerland); conse-quently, the Second World Congress on the Square ofOpposition has taken place in the University of Corte(Corsica, France) from 20th to 22nd June 2010. Despiteits somehow restrictive title, this worldwide event pur-ported to give an updating overview of the concept ofopposition as such, whether organized in a basic squareor any more complex structure. More generally, thepoint was to see how opposition can be entertained orapplied to so many various disciplines like philosophy,linguistics, mathematics, psychology, computer scienceor cognitive science.

The Congress consisted in two distinct sorts of talks,whether from invited speakers or contributors. Amongthe four invited speakers, Pierre Cartier proposed an in-vestigation of infinity and its paradoxical import, thusleading to the topic of paraconsistency and contradic-tion in mathematics. Stephen Read recalled the legacyof the medieval logician John Buridan with respect tothe theory of opposition, and he exemplified this witha special theory of truth where temporal and modal ex-pressions result in an intriguing logical octagon. Slid-ing from the Middle Age to the 20th century, DamianNiwinski proposed a reconstruction of Zermelo’s theo-rem about chess games in terms of fixed-point calculusand modal oppositions, while Hartley Slater returned tothe famous existential import problem and showed howthis problem could be viewed from another perspectivethan the usual quantification.

As to the fifty-four contributors, the various sorts oftalks helped to give clues to several fundamental ques-tions about the theory of opposition: how it came tobe within the area of philosophy, how it has been de-veloped and improved through the ages and scientificdisciplines, what is it its expected utility for solving sci-entific problems. The linguists notably emphasized theambiguous use of negation in ordinary discourse andthe plurarity of its geometrical representations, whilethe mathematicians showed how opposition is deeply

rooted through group theory or category theory. Thephilosophical and logical contribution purpoted to ex-emplify the theoretical development of the theory ofopposition after Aristotle’s square and Robert Blanche(and Augustin Sesmat)’s hexagon: fuzzy logic gives riseto an indefinite seriation of logical oppositions; accord-ingly, an extension of opposite terms gives rise to an in-definite sequence of polygons whose geometrical prop-erties respect the scientific criteria of rigor and fruitful-ness (Alessio Moretti’s n-opposition theory: N.O.T.); fi-nally, the philosophical roots of logical opposition don’tprevent this concept from being reduced to an algebraicapproach and introduced into an alternative logic whereinconsistency is the central property. In a nutshell, anumber of topics from history of philosophy to philos-ophy of logic have been considered from the point ofview of the theory of opposition, among which: non-classical logics, the theory of negation, (dynamic) epis-temic logic, logical pluralism, Eastern logics, Arabiclogic, categoricity, speech act theory, universal logic.

The Third Congress on the Square of Opposition isannounced for 2012 and should take place in Crete. Be-yond the huge variety of topics to be addressed, a cru-cial question that covers these is about the scientificvalue of the concept of opposition: how relevant is itfor scientific thought, and what does its proper contri-bution amount to in the history of ideas? A tentative an-swer has been already given through the first two events,pending the third one.

Fabien SchangDresden University of Technology, Germany

Inductive Logic Programming, 27–30 June

The 20th International Conference on Inductive LogicProgramming (ILP 2010) has been held in Firenze,Italy, chaired by Paolo Frasconi (Universita degli Studidi Firenze) and Francesca A. Lisi (Universita degliStudi di Bari “Aldo Moro”). The program featured 31accepted papers (see here), 3 invited talks (here) and 2tutorials (here). Contributions spanned from theory topractice, mainly within the stream of the so-called Sta-tistical Relational Learning (SRL). Here is a summaryof the principal new ideas (presenters’ names are in ital-ics).

Best Student Paper: “From Inverse Entailment toInverse Subsumption” (Yamamoto, Inoue, Iwanuma)shows how IE can be reduced to inverse subsumptionby preserving its completeness.

“Approximate Bayesian Computation for the Pa-rameters of PRISM Programs” (Cussens) presents aBayesian method for approximating the posterior dis-tribution over PRISM parameters.

“Applying the Information Bottleneck Approach to

140

SRL: Learning LPAD Parameters” (Riguzzi, Di Mauro)adopts the Information Bottleneck approach for learn-ing LPAD parameters.

“Extending ProbLog with Continuous Distributions”(Gutmann, Jaeger, De Raedt) extends ProbLog withabilities to specify and infer over continuous distribu-tions.

“Probabilistic Rule Learning” (De Raedt, Thon) up-grades rule learning to a setting in which both the ex-amples and their classification can be probabilistic.

“Boosting Relational Dependency Networks”(Natarajan et al.) proposes the use of gradient treeboosting in RDNs.

“Multitask Kernel-based Learning with First-OrderLogic Constraints” (Diligenti et al.) defines a generalframework to integrate supervised/unsupervised exam-ples with background knowledge in the form of FOLclauses into kernel machines.

“Stochastic Refinement” (Tamaddoni-Nezhad, Mug-gleton) introduces the notions of stochastic refinementoperator and search.

“Hypothesizing about Networks in Meta-level Ab-duction” (Inoue, Doncescu, Nabeshima) deals withcompleting causal networks by means of meta-level ab-duction.

“Learning Discriminant Rules as a Minimal Satura-tion Search” (Lopez, Martin, Vrain) defines a non-blindbottom-up search strategy for hypotheses.

“Speeding up Planning through Minimal Generaliza-tions of Partially Ordered Plans” (Cernoch, Zelezny)presents an ILP framework for planning which exploitsexisting plans in new similar planning tasks.

“Exact Data Parallel Computation for Very Large ILPDatasets” (Srinivasan, Faruquie, Joshi) shows how dis-tributed computing can be used effectively in ILP.

“Automating the ILP Setup Task: Converting UserAdvice about Specific Examples into General Back-ground Knowledge” (Walker et al.) introduces sometechniques to automate the use of ILP systems for a non-ILP expert.

“Fire! Firing Inductive Rules from Economic Ge-ography for Fire Risk Detection” (Vaz, Santos Costa,Ferreira) provides an elegant and powerful approach tospatial data mining by coupling Spatial-Yap with an ILPengine.

The invited talks were very inspiring. Kifer arguedthat RIF, a W3C recommendation for the exchange ofSemantic Web rules, is a major opportunity to rekindlethe interest in logic programming. Pfeffer presented anew probabilistic programming language named Figarothat is designed with practicality and usability in mind.Poole claimed that SRL and ILP need to be a foundation

of the Semantic Web.

Francesca A. LisiDipartimento di Informatica, Universita degli Studi di

Bari “A. Moro”

Work in Progress in Causal and Probabilis-tic Reasoning, 28-29 June

Federica Russo and Phyllis McKay Illari from theUniversity of Kent recently organised an internationalworkshop titled ‘Work in Progress in Causal and Prob-abilistic Reasoning’. The workshop took place in theafternoon of June 28th and the morning of June 29th.Paris, more precisely, the campus of the Universityof Kent at Paris, formed the congenial venue for themeeting, with its wood-panelled rooms and flower-filledcourtyards.

The aim was to form a hub for exchanges betweengraduate students and experienced researchers in thefield. Jon Williamson, Phyllis McKay Illari and Fed-erica Russo from Kent, Julian Reiss from Rotterdam,Margherita Benzi from Vercelli, and Bert Leuridan andLeen De Vreese from Gent were the more experiencedresearchers in the field who took part in the workshop,fostering the exchange of thoughts through feedback,additional presentations, and the leading and feedingof discussions. But it was primarily the attendanceand the contributions of the participating young re-searchers from Kent, Rotterdam, London, and Paris,which made the workshop a success. Taken all together,a wide range of topics related to causation were cov-ered during the two half-days of talks: causal plural-ism (Leen De Vreese), causal contextualism (MichaelWilde), Glennan-style mechanisms and complex sys-tems (Lorenzo Casini), causal evidence in economics(Francois Claveau), causation and explanation (ConradHughes), causation and constitutive relevance (AdamWhite and Bert Leuridan), causal inference in the socialsciences (Attilia Ruzzene), and invariantism and proba-bility (Eric Raidl). Plenty of time was reserved for feed-back and discussion on purpose, to give graduate stu-dents the experience of academic debate at a researchlevel.

The whole organizational set-up clearly formed afruitful venue, in which junior researchers could learn inan informal setting, not only about the content of theirown and others’ research work, but also about present-ing and giving feedback more generally. A follow-upworkshop with the same aims, to continue this work,would be highly recommended.

Leen De VreeseCentre for Logic and Philosophy of Science, UGent

141

Automated Reasoning about Context andOntology Evolution, 16–17 August

In computer science an ontology is a formal represen-tation of the knowledge used by computer applications.Contexts play a key role in ontology dynamics.

An ontology can be compared with its different ver-sions under many respects. For instance, the choice ofthe specification language matters: one should strivefor the appropriate balance between expressiveness andcomplexity. Description logics (DLs) are widely usedthanks to the good balance they strike. Other languages,though, may be used to support more expressivenessor ease of use, like higher-order logics, modal logics,RDF, among others. On the other hand, the same ontol-ogy can be compared with itself relative to its evolutionover time, dictated by new requirements or changes inthe domain knowledge.

Understanding how to automate the evolution of on-tologies and the role that contexts play in it was the cen-tral theme of ARCOE-10, held at ECAI-10, in Lisbon.The two-day workshop had two invited talks, nine regu-lar presentations, a participants and a panel discussion.

Prof. Meyer gave an invited talk on ontology dy-namics and its relationship with belief change and non-monotonic reasoning. Prof. Giunchiglia’s invited talkcovered the notion of knowledge diversity and stressedthe importance of relating localized semantics to sharedsemantics.

Two contributed talks by Ribeiro and Wassermannand by Meyer, Moodley and Varzinczak brought tech-niques from belief revision to the ontology case. Withsimilar motivations, Nguyen, Alechina and Loganshowed how to use assumption-based truth maintenancesystems for debugging ontologies. Lehmann, Bundyand Chan analyzed the evolution of physics theories byqualitative causal analysis of experimental data. Wal-lace and Naz focused on the role of contexts in ontologyconstruction; Chan, Bundy and Lehmann used contextsin detecting conflicts between ontologies; d’Amato andFanizzi presented a machine learning approach to com-paring entities in a given context. Finally, Pease andBenzmueller presented in one talk the historical evolu-tion of the SUMO ontology and in another talk a higher-order logic approach to embedding formulas in SUMO.

Two discussions about the relationships between thetalks made clear the value of a multi-disciplinary forumlike ARCOE. During the panel discussion Giunchigliapointed out the fundamental problems of the SemanticWeb endeavor and of its deductive approach. ThomasMeyer stressed the differences between semantic weband semantic technologies: even if the semantic webendeavor might eventually fail, this does not preventus from taking advantage of specific existing semantictechnologies, which have proven to be useful. Fanizzicommented on how to bring methods from machine

learning to ontology evolution and also pointed out themain issue involved in semantic technology: where toget the relevant data from?

There are plans to continue the ARCOE collabora-tion and to reach out other communities interested inontologies, context and reasoning. More informationon the website.

Alan BundySchool of Informatics, University of Edinburgh, UK

Jos LehmannSchool of Informatics, University of Edinburgh, UK

Guilin QiSchool of Computer Science and Engineering,

Southeast University, China

Ivan Jose VarzinczakCSIR Meraka Institute, Pretoria, South Africa

Calls for PapersLogic and Natural Language: special issue of StudiaLogica, deadline 3 September.The ExtendedMind: special issue of Teorema, deadline1 October.Recurrence, Provability and Truth: special issue ofLogos Architekton, deadline 15 October.From Embodied Cognition to Free Will: special issueof Humana.Mente, deadline 30 October.AILACT Essay Prize: in Informal Logic / CriticalThinking / Argumentation Theory, with publication onInformal Logic, deadline 31 October.Philosophical History of Science: special issue of TheMonist, deadline 31 October.Categorical Logic: special issue of Logica Universalis,deadline 1 November.Philosophy & Technology Best Paper Prizes: winningpapers are published in Philosophy& Technology, dead-line 1 November.Concepts of Tradition in Phenomenology: special issueof Studia Phaenomenologica, deadline 15 November.Social Cognition: Mindreading and Alternatives: spe-cial issue of the Review of Philosophy and Psychology,deadline 1 December.Visual Reasoning with Diagrams: special issue of Log-ica Universalis, deadline 15 December.Experimental Philosophy: special issue of The Monist,deadline 30 April 2011.C. L. Hamblin and Argumentation Theory: special is-sue of Informal Logic, deadline 30 June.Formal and Intentional Semantics: special issue of TheMonist, deadline 30 April 2012.

142

§4What’s Hot in . . .

We are looking for columnists willing to write piecesof 100-1000 words on what’s hot in particular areasof research related to reasoning, inference or method,broadly construed (e.g., Bayesian statistical inference,legal reasoning, scientific methodology). Columnsshould alert readers to one or two topics in the par-ticular area that are hot that month (featuring in blogdiscussion, new publications, conferences etc.). If youwish to write a “What’s hot in . . . ?” column, either ona monthly or a one-off basis, just send an email to fea-tures@thereasoner.org with a sample first column.

. . . Logic and Rational Interaction

Conference season has started in the area of Logic andRational Interaction as well, and reports on a number ofrecent events are already available on LORIWEB. TheParis-Amsterdam Logic Meeting of Young Researchers(PALYMR) took place in Amsterdam this year. InesCrespo and Lucian Zagan provide information aboutthe talks given. The Workshop on Solution Conceptsfor Extensive Games (SCEG ’10) was organized by theCenter for Algorithm Game Theory at Aarhus Univer-sity and took place June 22-25, 2010. Peter Bro Mil-tersen reports on the workshop, which brought togethercomputer scientists and game theorists. Paolo Turrinidiscusses his impressions gathered at the tenth editionof the International Conference on Deontic Logic inComputer Science. And Jean-Yves Beziau providesinformation about the second world congress on thesquare of opposition, which took place in Corsica thisJune.

Readers of The Reasoner may also be interested inthe Justification Logic Bibliography, which has beenupdated with several new publications recently. RomanKuznets provides details.

As usual, we would like to invite all readers ofThe Reasoner to contribute to LORIWEB with newson events, job openings or publications in the area ofLogic and Rational Interaction. Please contact RasmusRendsvig, our web manager, or write to the loriweb ad-dress.

Ben RodenhauserPhilosophy, Groningen

. . . Algebraic, abstract algebraic and behav-ioral approaches to logical systems. Part II

In the previous contribution to The Reasoner it was ob-served that there are deductive systems which do not

possess a biconditional (↔) determining logical equiv-alence. Consequently, it was claimed that these logicalsystems are not susceptible to the Lindenbaum-Tarskiprocess of forming the quotient algebra from the for-mula algebra Fm. Hence, algebraically oriented logi-cians introduced the concept of the Leibniz congruence(Ω(T )) and by using it the Lindenbaum-Tarski processcan be fully generalized and applied to any deductivesystem C.

Recall that Ω(T ) defined on Fm over a theory T ischaracterized in the subsequent way: for any pair α, βof terms, α ≡ β (modulo Ω(T )) if for every formulaϕ and any variable p occurring in ϕ, it is the case thatϕ(p/α) ∈ T if and only if ϕ(p/β) ∈ T . Then α and β aresaid to be Leibniz-congruent (modulo T ). Identifyingany theory T of C with the deductively closed set offormulas Φ it can be asserted that if α and β are Leibniz-congruent (modulo Φ), then

Φ ∪ α `C β and Φ ∪ β `C α.

This interderivability condition is equivalent toΦ−indiscernibility of two Leibniz-congruent (moduloΦ) formulas. If Φ is the set of theorems of C, then thisrelation is said to be synonymy relation. It turns out thatdefinition of Ω(T ) depends only on the structural (i.e.,grammatical) properties of the language of C. Henceif α ≡ β (moduloΩ(T )), then α and β are interchange-able with respect to T in every context represented byany formula ϕ ∈ Fm. It should be stressed that the def-inition of the Leibniz congruence is associated to themethod of defining the identity relation in second-orderlogic that is due to Gottfried Leibniz. Recall that ac-cording to the famous Leibniz Law the identity can beexpressed by the formula

∀a∀b [a = b↔ ∀P (P (a)↔ P (b))]

where P is a variable running over all unary predicates.This ontological principle states that two objects a andb are identical (i.e., they are the same entity), if theyhave all their properties in common. This means thata and b are identical if any predicate possessed by a isalso possessed by b and vice versa. Consequently, it isobvious that according to this law the identity of twosubstances is defined by their indiscernibility.

In order to formulate this principle algebraically letus introduce the notion of a logical matrix. This model-theoretic concept generalizes the well-known (from el-ementary logic) notion of truth tables. Logical matricesare regarded as models for a logical language L. Anylogical matrix has the form M = 〈A, F〉 where A is anabstract algebra of type L and F ⊆ A is the set of des-ignated values. The fundamental function of algebra inlogic is to constitute semantic correlates of sentences.The term ‘semantic correlate’ is synonymous with suchconcepts as ‘meaning’ and ‘denotation’. It is assumed

143

that each formula ϕ ∈ Fm has a uniquely determined in-terpretation in A depending on the values in A that areassigned to its variables. Then the interpretation of ϕcan be expressed algebraically as h(ϕ), where h is a ho-momorphism (i.e., an assignment) from Fm to A map-ping each variable of ϕ into its assigned value. Then Ais identified with the universe of all possible interpreta-tion, i.e., semantic correlates for each ϕ ∈ Fm. Eachsuch homomorphism h : Fm→ A becomes a possiblesemantic correlate function from the language L to A.Then any subset F ⊆ A is said to be a truth predicate(or a truth set). Therefore, it is said that a sentence ϕ istrue inM if h(ϕ) ∈ F. Otherwise ϕ is false inM.

Using logical matrices it is possible to reformulateLeibniz’s second-order definition of identity. Namely itfollows that

ΩA(F): =

〈a, b〉 : ϕA(a, c0, ..., cn−1) ∈ F iff ϕA(b, c0, ..., cn−1) ∈ F

for all ϕ(x, z0, ..., zn−1) ∈ Fm and all c0, ..., cn−1 ∈ A

.

This is the first-order analogue of Leibniz’s defini-tion. The notation ϕ(x, z0, ..., zn−1) indicates that each ofthe variables from ϕ occurs in the list x, z0, ..., zn−1 andϕA(a, c0, ..., cn−1) is the algebraic interpretation of ϕ inA, i.e., h(ϕ) where h is any homomorphism h : Fm→ Asuch that h(x) = a and h(zi) = ci for all i < n. The stringof variables c0, ..., cn−1 ∈ A is regarded as the sequenceof parameters. Also it follows that ΩA(F) is a congru-ence on A. Observe that the definition of ΩA(F) is com-pletely independent of any logic C. It depends entirelyon the grammatical properties of L, i.e., it is definedintrinsically with respect to A and F. Then it is obvi-ous that according to ΩA(F) two entities are identicaliff they are indiscernible with respect to each propertyexpressed by any first-order formula ϕ ∈ Fm. It turnsout that ΩA(F) is the main metalogical tool of AbstractAlgebraic Logic (AAL).

Recall that in the framework of AAL we consider onlyone-sorted languages and algebras. While in the case ofBehavioral Abstract Algebraic Logic (BAAL) it is nec-essary to use many-sorted languages and algebras. Amany-sorted signature is a pair Σ = 〈S , F〉 where S isa set of sorts and F = Fwsw∈S ∗,s∈S is an indexed fam-ily of sets of operations. For simplicity, it is writtenf : s1...sn → s ∈ F for an element f ∈ Fs1...sn s. Then theformula algebra FmΣ in BAAL is generated by a sortedfamily X = Xss∈S of variable sets. The fact that x ∈ Xs

is denoted by x : s. Given a signature Σ = 〈S , F〉 it issaid that a Σ−algebra (similar to many-sorted language)is a pair A =

⟨Ass∈S , A

⟩, where each As is a non-void

set (i.e.,the carrier of sort s) and A assigns to each oper-ation f : s1...sn → s a function f

A: AS 1×...×AS n → AS .

An assignment over A is a S−sorted family of functionsh = hs : Xs → Ass∈S . Given a Σ−algebra A, a formulafrom FmΣ, i.e., ϕ(x1 : s1, ..., xn : sn) and 〈a1, ..., an〉 ∈

As1 × ...×Asn , then we denote by ϕA(a1, ..., an) the valueh(ϕ) that ϕ takes in A under an assignment h such that

h(x1) = a1, ..., h(xn) = an. In our approach to BAAL it isassumed that the set of sorts is split in two disjoint sets Vand H which are called visible and hidden sorts. Then ahidden many-sorted signature is a triple 〈Σ,V,E〉 whereΣ = 〈S , F〉 is a many-sorted signature, V ⊆ S is the setof visible sorts and E is the set of available experiments.These experiments are identified with terms of visiblesort of the form t (x : s, x1 : s1, ..., xn : sn) where x is adesignated variable of hidden sort, i.e., s ∈ H = S \V .It follows that in BAAL it is possible to reason abouthidden data only indirectly using behavioral indistin-guishability with respect to the available experiments.Intuitively speaking, it is necessary to evaluate equa-tions involving hidden values using only their visibleproperties. Such we arrive at the following definition:

Definition 1. Consider a hidden signature 〈Σ,V,E〉and a Σ−algebra A. Then given a hidden sort s ∈ H it issaid that two values a, b ∈ As are E−behaviorally equiv-alent, symbolically a ≡E b, if for every experiment t(x :s, x1 : s, ..., xn : sn) ∈ E and every 〈a1, ..., an〉 ∈ As1×...×Asn it is the case that tA(a, a1, ..., an) = tA(b, a1, ..., an).

Concluding, it can be stated that in BAAL it is possi-ble to observe the behavior of terms of hidden sorts bytheir indirect impact on the truth-values of the formulaswhich involve them.

Piotr WilczekMathematics, Poznan University of Technology,

Poland

Introducing . . .

If you would like to write one or more shortintroductions to concepts, topics, authors or books

connected with reasoning, inference or method, or ifyou have an editorial project to collate such pieces and

would like to print some of them here, please emailfeatures@thereasoner.org with your proposal.

§5Events

September

ICTAC: 7th International Colloquium on TheoreticalAspects of Computing, Natal, Brazil, 1–3 September.KSEM: 4th International Conference on KnowledgeScience, Engineering and Management, Belfast, North-ern Ireland, UK, 1–3 September.FEW: 7th Annual Formal Epistemology Workshop,Konstanz, 2–4 September.CMM Graduate Conference: University of Leeds, 3September.The Cartesian “Myth of the Ego” and the Ana-lytic/Continental Divide: Faculty of Philosophy, Rad-boud University Nijmegen, 3–4 September.

144

TIME: 17th International Symposium on TemporalRepresentation and Reasoning, Paris, France, 6–8September.CP: Principles and Practice of Constraint Programming,St. Andrews, Scotland, 6–10 September.Principles and Methods of Statistical Inference withInterval Probability: Durham, 6–10 September.Causation and Disease in the Postgenomic Era: 1st Eu-ropean Advanced Seminar in the Philosophy of the LifeSciences, Geneva, Switzerland, 6–11 September.FEW: Formal Ethics Week, University of Groningen,7–10 September.Logic, Algebra and TruthDegrees: Prague, Czech Re-public, 7–11 September.Pluralism in the Foundations of Statistics: Universityof Kent, Canterbury, UK, 9–10 September.Economics and Naturalism: Kazimierz Dolny, Poland,11–15 September.CNL: 2nd Workshop on Controlled Natural Languages,Marettimo Island, Sicily, Italy, 13–15 September.PGM: 5th European Workshop on Probabilistic Graph-ical Models, Helsinki, Finland, 13–15 September.Epistemic Aspects ofMany-valued Logics: Prague, 13–16 September.RSS: Royal Statistical Society International Confer-ence, Brighton, United Kingdom, 13–17 September.Vagueness andMetaphysics: Barcelona, 16–17 Septem-ber.Levels of Processing: Foundations of Social Cogni-tion: University Club Bonn, 16–18 September.Non-Classical Logic. Theory andApplications: Torun,Poland, 16–18 September.AS: Applied Statistics, Ribno, Bled, Slovenia, 19–22September.GAMES: Annual Workshop of the ESF NetworkingProgramme on Games for Design and Verification, StAnne’s College, Oxford, UK, 19–23 September.Words and Concepts: An InterdisciplinaryWorkshopon Philosophy, Psychology, and Linguistics: Universityof Granada, Spain, 20–21 September.IVA: 10th International Conference on Intelligent Vir-tual Agents, Philadelphia, Pennsylvania, USA, 20–22September.LRR: Logic, Reason and Rationality, Centre for Logicand Philosophy of Science, Ghent University, Belgium,20–22 September.World Computer Congress: International Federationfor Information Processing, Brisbane, Australia, 20–23September.ECML: European Conference on Machine Learningand Principles and Practice of Knowledge Discovery inDatabases, Barcelona, Spain, 20–24 September.MATES: 8th German Conference on Multi-Agent Sys-tem Technologies, Karslruhe, Germany, 21–23 Septem-ber.

History, Cognition, and Visualisation in Science: TheDavid GoodingMemorialMeeting: University of Bath,UK, 22–23 September.Actual Causation: University of Konstanz, Germany,23–24 September.Truth, Knowledge and Science: 9th National Confer-ence of the Italian Society for Analytic Philosophy, Uni-versity of Padua, 23–25 September.&HPS3: Integrated History and Philosophy of Science,Indiana University, Bloomington, 23–26 September.Logic and Language Conference: Northern Institute ofPhilosophy, University of Aberdeen, 24–26 September.Workshop on Mental Causation: Durham University,27 September.SMPS: 5th International Conference on Soft Methodsin Probability and Statistics, Mieres (Asturias), Spain,28 September - 1 October.Logic or Logics?: Workshop, Arche Research Centre,St Andrews, Scotland, 30 September - 1 October.Truth, Meaning, and Normativity: Department of Phi-losophy, Institute for Logic, Language and Computa-tion, Universiteit van Amsterdam, 30 September - 2 Oc-tober.Types of Explanation in the Special Sciences: the Caseof Biology and History: University of Cologne, 30September - 3 October.

October

AP-CAP: Asia-Pacific Computing and Philosophy Con-ference, Wellington Institute of Technology (WelTec),Petone, Wellington, New Zealand, 1–2 October.E-CAP: 8th European Conference on Computing andPhilosophy,Muenchen, Germany, 4–6 October.Objectivity and the Practice of Science: Tilburg Cen-ter for Logic and Philosophy of Science, 5 October.AIAI: 6th IFIP International Conference on ArtificialIntelligence. Applications & Innovations, Ayia Napa,Cyprus, 5–7 October.Calculation, Intuition, and A Priori Knowledge:Tilburg University, The Netherlands, 5–8 October.Validation in Statistics and Machine Learning: Weier-strass Institute, Berlin, 6–7 October.

Causality in the Biomedical and Social Sciences

Erasmus University Rotterdam, 6–8 October

The Limits of Knowledge Society: Iasi, Romania, 6–9October.Integrating Complexity: Environment and History:University of Western Ontario in London, Ontario,Canada, 7–10 October.LPAR: 17th International Conference on Logic for Pro-gramming, Artificial Intelligence and Reasoning, Yo-gyakarta, Indonesia, 10–15 October.Philosophy ofMind, Reduction, Neuroscience: Univer-sity of Lausanne, Switzerland, 12–16 October.

145

SEFA: 6th Conference of the Spanish Society for Ana-lytic Philosophy, University of La Laguna, Tenerife.14–16 OctoberPhilosophy of Scientific Experimentation: A Chal-lenge to Philosophy of Science: Center for Philosophyof Science, University of Pittsburgh, 15–16 October.The Nature of Belief: The Ontology of Doxastic Atti-tudes, University of Southern Denmark, Odense, 18–19October.FMCAD: International Conference on Formal Methodsin Computer-Aided Design, Lugano, Switzerland, 20–23 October.ADT: 1st International Conference on Algorithmic De-cision Theory, Venice, Italy, 21–23 October.Workshop on Bayesian Argumentation: Departmentof Philosophy & Cognitive Science, Lund University,Sweden, 22–23 October.Field Science: 26th Boulder Conference on the Historyand Philosophy of Science, University of Colorado atBoulder, 22–24 October.Thinking and Speaking a Better World: 3rd Interna-tional Conference on Argumentation, Rhetoric, Debateand the Pedagogy of Empowerment, Faculty of Arts,University of Maribor, Slovenia, 22–24 October.NonMon@30: Thirty Years of Nonmonotonic Reason-ing, Lexington, KY, USA, 22–25 October.MWPMW: 11th annual Midwest PhilMath Workshop,Philosophy Department, University of Notre Dame, 23–24 October.IJCCI: 2nd International Joint Conference on Computa-tional Intelligence, Valencia, Spain, 24–26 October.BNAIC: 22nd Benelux Conference on Artificial Intelli-gence, Luxembourg, 25–26 October.ICTAI: 22th International IEEE Conference on Toolswith Artificial Intelligence, Arras, France, 27–29 Oc-tober.

November

ICMSC: IEEE International Conference on Modeling,Simulation and Control, Cairo, Egypt, 2–4 November.LogKCA: International Workshop on Logic and Phi-losophy of Knowledge, Communication and Action,Donostia, San Sebastian, Spain, 3–5 November.MindNetwork: 2nd meeting of the Mind Network, anetwork for Philosophy of Mind & Cognitive Science,King’s College, Cambridge, 6 November.MICAI: 9th Mexican International Conference on Arti-ficial Intelligence, Pachuca (near Mexico City), Mexico,8–12 November.Causation, Coherence, and Concepts. Themes fromthe Philosophy of Wolfgang Spohn: Konstanz, 11–13November.P-NPMW: 2nd Paris-Nancy PhilMath Workshop, Paris,17–19 November.

AMBN: 1st International Workshop on AdvancedMethodologies for Bayesian Networks, Tokyo, Japan,18–19 November.LENLS: Logic and Engineering of Natural LanguageSemantics, Tokyo, 18–19 November.TAAI: Conference on Technologies and Applications ofArtificial Intelligence, Hsinchu, Taiwan, 18–20 Novem-ber 18-20.KICS: 5th International Conference on Knowledge, In-formation and Creativity Support Systems, Chiang Mai,Thailand, 25–27 November.ISDA: International Conference on Intelligent SystemsDesign and Applications, Cairo, Egypt, 29 November -1 December.

December

Semantics for Robots: Utopian and Dystopian Visionsin the Age of the ‘Language Machine’: Chicago, Illi-nois, USA, 2–4 December.MINDGRAD: Warwick Graduate Conference in thePhilosophy of Mind, University of Warwick, UK, 4–5December.CACS: International Congress on Computer Applica-tions and Computational Science, Singapore, 4–6 De-cember.NIPS: 24th Annual Conference on Neural InformationProcessing Systems, Vancouver, B.C., Canada, 6–11December.From Cognitive Science and Psychology to anEmpirically-informed Philosophy of Logic: Amster-dam, 7–8 December.MIWAI: 4th Mahasarakham International Workshop onArtificial Intelligence, Mahasarakham, Thailand, 9–10December.APMP: 1st International Meeting of the Association forthe Philosophy of Mathematical Practice, Brussels, 9–11 December.ICDM: International Conference on Data Mining, Syd-ney, Australia, 13–17 December.SILFS: International Conference of the Italian Societyfor Logic and Philosophy of Sciences, University ofBergamo, Italy, 15–17 December.Scepticism and Justification: COGITO Research Cen-tre in Philosophy, Bologna, 17–18 December.International Conference on Recent Advances in Cog-nitive Science: Varanasi, India, 18–20 December.

January

LogICCC Meets India: Delhi University, India, 7–8January.ICLA: 4th Indian Conference on logic and its Applica-tions, New Delhi, India, 9–11 January.Graduate Conference in Epistemology: Miami, FL,13–15 January.

146

Philosophy of Science Colloquium: Durban, SA, 18January.The Notion of Form in 19th and Early 20th CenturyLogic and Mathematics: International graduate work-shop, Vrije Universiteit Amsterdam, 20–21 January.SODA11: ACM-SIAM Symposium on Discrete Algo-rithms, San Francisco, California, USA, 23–25 January.ICISD: International Conference on Intellingent Sys-tems & Data Processing, Vallabh Vidyanagar, Gujarat,India, 24–25 January.ICAART: 3rd International Conference on Agents andArtificial Intelligence, Rome, Italy, 28–30 January.CCA: Computability and Complexity in Analysis, CapeTown, South Africa, 31 January - 4 February.

February

AIA: 11th International Conference on Artificial Intel-ligence and Applications, Innsbruck, Austria, 14–16February.PhDs in Logic: Graduate Conference and WinterSchool, Brussels, 17–18 February.

March

STACS: 28th International Symposium on TheoreticalAspects of Computer Science, Dortmund, Germany,10–12 March.NAFIPS: 30th North American Fuzzy Information Pro-cessing Society Annual Conference, El Paso, Texas,USA, 18–20 March.AI and Health Communication: Stanford University,California, 21–23 March.

April

SpringSim: Spring Simulation Multi-conference,Boston, MA, USA, 4–9 April.The Authority of Science: University of Sydney, Aus-tralia, 8–10 April.AIML: ICGST International Conference on ArtificialIntelligence and Machine Learning, Dubai United ArabEmirates, 11–14 April.AICS: 22nd Midwest Artificial Intelligence and Cogni-tive Science Conference, Cincinnati, Ohio, USA, 16–17April.NFM: 3rd NASA Formal Methods Symposium,Pasadena, California, USA, 18–20 April.

§6Courses and Programmes

CoursesSIPTA: 4th school of the Society for Imprecise Prob-ability: Theories and Applications, Durham, UK, 1–6

September.Logic or Logics?: Mini-course, Arche Research Centre,St Andrews, Scotland, 27–29 September.BLT: Bochum-Lausanne-Tilburg Graduate School: Phi-losophy of Language, Mind and Science on Calcula-tion, Intuition, and A Priori Knowledge, Tilburg Uni-versity, The Netherlands, 5–8 October; Philosophy ofMind, Reduction, Neuroscience, University of Lau-sanne, Switzerland, 12–16 October.SELLC: Sino-European Winter School in Logic, Lan-guage and Computation, Guangzhou, China, 3–18 De-cember.

ProgrammesDoctoral Programme in Philosophy: Language, Mindand Practice, Department of Philosophy, University ofZurich, Switzerland.HPSM: MA in the History and Philosophy of Scienceand Medicine, Durham University.Master Programme: Philosophy of Science, Technol-ogy and Society, Enschede, the Netherlands.MA in Cognitive Science: School of Politics, Inter-national Studies and Philosophy, Queen’s UniversityBelfast.MA in Logic and the Philosophy of Mathematics: De-partment of Philosophy, University of Bristol.MA in Metaphysics, Language, and Mind: Departmentof Philosophy, University of Liverpool.MA in Mind, Brain and Learning: Westminster Insti-tute of Education, Oxford Brookes University.MA in Philosophy: by research, Tilburg University.MA in Philosophy of Biological and Cognitive Sci-ences: Department of Philosophy, University of Bristol.MA in Rhetoric: School of Journalism, Media andCommunication, University of Central Lancashire.MA programmes: in Philosophy of Language and Lin-guistics, and Philosophy of Mind and Psychology, Uni-versity of Birmingham.MRes in Methods and Practices of Philosophical Re-search: Northern Institute of Philosophy, University ofAberdeen.MSc in Applied Statistics and Datamining: School ofMathematics and Statistics, University of St Andrews.MSc in Artificial Intelligence: Faculty of Engineer-ing, University of Leeds.

MA in Reasoning

An interdisciplinary programme at the University ofKent, Canterbury, UK. Core modules on logical,causal, probabilistic, scientific, mathematical and

machine reasoning and further modules fromPhilosophy, Psychology, Computing, Statistics, Social

Policy, Law, Biosciences and History.

MSc in Cognitive & Decision Sciences: Psychology,University College London.

147

MSc in Cognitive Science: University of Osnabruck,Germany.MSc in Cognitive Psychology/Neuropsychology:School of Psychology, University of Kent.MSc inMathematical Logic and the Theory of Compu-tation: Mathematics, University of Manchester.MSc in Philosophy of Science, Technology and Soci-ety: University of Twente, The Netherlands.Master of Science: Logic, Amsterdam.ResearchMaster inCognitive Science andHumanities:Language, Communication and Organization: Institutefor Logic, Cognition, Language, and Information, Uni-versity of the Basque Country (DonostiaSan Sebastian).

§7Jobs and Studentships

Jobs

Postdoctoral Research Associate: on the project“Word Segmentation from Noisy Data with MinimalSupervision”, School of Informatics, University of Ed-inburgh, deadline 1 September.Bertrand Russell Professorship of Philosophy: Fac-ulty of Philosophy, University of Cambridge, deadline10 September.Professorship: in Mathematical Logic, Departmentof Mathematics, Stockholm University, deadline 15September.Post-Doctoral Research Fellowships: in the Hu-manities, Social Sciences, and Theoretical Sciences,All Souls College, University of Oxford, deadline 24September.Visiting International Fellowship: in social researchmethods for visits in calendar year 2011, Department ofSociology, University of Surrey, Guildford, UK, dead-line 30 September.Post-doc Stipend: in theoretical philosophy, Philoso-phy Department, Ruhr-Universitat Bochum, deadline10 October.Tenure-track position: with specialisation in philoso-phy of science, Department of Philosophy, ConcordiaUniversity, Montreal, Quebec, deadline 1 November.Wagner Risk Fellowship: Center for Philosophy of Sci-ence, University of Pittsburgh, deadline 15 November.

Studentships

10 PhD student positions: within the doctoral pro-gram “Mathematical Logic in Computer Science”, Vi-enna University of Technology (TU Wien), until filled.PhD Studentship: “Hyper-heuristics for GroupingProblems”, School of Computer Science, University ofNottingham, until filled.

PhD position: “Modelling the evolution of theory ofmind”, Institute of Artificial Intelligence (ALICE), Uni-versity of Groningen, deadline 1 September.PhD position: “A cognitive system supporting intel-ligent interaction”, Institute of Artificial Intelligence(ALICE), University of Groningen, deadline 1 Septem-ber.PhD position: “Logics for higher-order social cogni-tion”, Institute of Artificial Intelligence (ALICE), Uni-versity of Groningen, deadline 1 September.PhD positions: in the programme “Foundations of theLife Sciences and Their Ethical Consequences” (FOL-SATEC) at the SEMM (European School of MolecularMedicine), Milano, deadline 26 September.

148