Volume 12 Number 1January 2018

thereasonerorgISSN 1757-0522

Contents

Guest Editorial 1

Features 1

News 3

Whatrsquos Hot in 4

Events 6

Courses and Programmes 6

Jobs and Studentships 7

Guest Editorial

What makes the Reasoner such an interest-ing venue is that the topic ndash ldquoreasoningrdquondashis one that crosses tradi-tional disciplinary bound-aries People interested inreasoning might find them-selves in university depart-ments ranging from psychol-ogy to economics to statisticsto philosophy and beyondThis is I think a blessingand a curse A blessing be-cause it means that in the-ory we have at our disposala very broad range of toolswith which to investigate the topic A curse because there isa real danger of reasoning researchers in different departmentsnot talking to each other or talking past one another

Our interviewee this month is someone who has spent his ca-reer moving around across those disciplinary boundaries tryingto bridge the gaps Greg and I first met in the spiritual home ofthe Reasoner ndash the University of Kent ndash at a conference abouta topic of central interest to many readers of the Reasoner thefoundations of statistics Several years later we shared an officein Munich for a while It was a pleasure to interview Greg andto learn about his views on a topic that will I predict becomemore and more relevant to those of us interested in reasoning inthe coming years machine learning

Seamus BradleyPhilosophy University of Tilburg

Features

Interview with Gregory WheelerSeamus Bradley Hi Greg thanks for agreeing to be inter-viewed

GregoryWheeler My pleasure Seamus It is always goodto talk to you

SB Letrsquos start with a little background If I remember cor-rectly you worked in engineering in some capacity before turn-ing to philosophy

GW Thatrsquos right I worked for IBM and a spin-off calledStorageTek as a mechanical vibration test engineer My jobwas to simulate earthquakes and military planes landing on re-mote dirt strips to see whether the hardware we were designingwould survive But I also diagnosed vibration problems that webrought on ourselves My days were spent breaking expensiveprototypes and figuring out why they broke

SB And you did your PhD under Henry Kyburg what washe like

GW Kyburg was a lapsed engineer too Chemical And acattle farmer on the side He and his wife Sarah bought a farm

1

in the late 60s then kept buying adjoining farms adding cowsand dogs to steer them around from one pasture to another un-til they had the largest black angus cattle farm in upstate NewYork by the 1990s An unexpected life for a fellow raised inan aristocratic family in New York City and educated at Yaleand Columbia Then again there is a Kyburg castle in Zurichdating from the 11th century ldquoKyburgrdquo it turns out meansldquocow-fortrdquo

Kyburgrsquos early critiques of probabilism are rooted in hispositive theory of evidential probability which interestinglytook de Finettirsquos lamentation to not mistake mathematical con-venience for a normative standard to heart But whereas deFinetti started with qualitative comparative judgements Ky-burg started with empirical classifications and counting mem-bers of the class The result was a radically non-subjectivist the-ory of probability even more radical than the late LMU statis-tician Kurt Weichselbergerrsquos theory and also more radical thanTerry Finersquos program who was just down the road from Ky-burg at Cornell Isaac Levi Kyburgrsquos lifelong friend and PeterWalley who was Terry Finersquos student are two others who tookde Finettirsquos maxim seriously although they stuck with assess-ments of coherent judgments but dropped precision as a nor-mative requirement Teddy Seidenfeld who was a student ofHenryrsquos in the late 60s and Isaac in the early 70s invariably puthis finger exactly on where these proposals broke-down andwhy which made perfect sense to me Through Henry I metGert de Cooman and the ISIPTA community where I got theidea that Europe might be the place for me

SB Yoursquove had something of an interdisciplinary careerworking in a computer science department in Portugal and nowin a business school in Frankfurt What has that experiencebeen like

GW I prefer life on the margins of philosophy Who wasit George Polya who said that good ideas are hurt by uncrit-ical acceptance but flourish under critical examination Engi-neers and economists both have a low-tolerance for nonsensebut are often quick to jump on a good idea Many topics thatinterest me can be recast as problems squarely in these neigh-boring fields and Irsquove found over the years that the effort todo philosophy in terms familiar to scientists and engineers re-pays handsomely I am mystified by philosophers proclaimingthe relevance of philosophy to the sciences on one day only toturn their attention in the next to coining cute names to mud-dle standard terminology The term ldquoimprecise credencesrdquo forexample which has spread like kudzu is an invasive cover formushy thinking about imprecise probabilities

SB Irsquom struck by your suggestion that the barriers be-tween philosophers and economists or engineers are just a mat-ter of language You donrsquot think there are deeper conceptual ormethodological differences

GW I would put the point this way Scientists routinely runinto problematic features of their models or conceptual mud-dles in their methodologies and within each field both a loreand literature exists to try to address them These fields alreadydo philosophy in other words even if many of those authorswouldnrsquot be comfortable calling it that Now a useful contri-bution might be to show the generality of a problem they havebumped into mdash a result that that reveals that the problem theysee with the parameters in this model are connected to a prob-lem over there with that other model Sometimes insights ofthis kind will come from a philosopher Other times such in-sights will come from a scientist or engineer who stumbled one

too many times over this pair of problems and dove into thematter herself to sort it out Whatever the source both con-tributions belong to philosophy But I donrsquot care whether thepeople who make these contributions are called philosophers orsomething else My interest is philosophy rather than being aphilosopher

Philosophers are sensitive to the sting of criticism from sci-entists So let me be clear that my remarks are not to be un-derstood as a brief for scientism (Given the times perhaps amoratorium on silly debates sparked by generic quantificationis in order) Rather my point is that it is very difficult to make acontribution to scientific methodology without knowing some-thing about science So if someone takes himself to be makinga deep conceptual point about scientific methodology but is re-peatedly met by derision from scientists I would suggest thathe reassess his project instead of seeking solace from otherswho have suffered the same brusque treatment

SB Irsquove heard you saythat you think research in AIand machine learning is go-ing to have a big impact ontopics like epistemology andtheories of reasoning in gen-eral What makes you thinkthat

GW Artificial Intel-ligence has a history ofover-promising and whenI started it was a mildlydisreputable but deeplyfascinating discipline Therewere good-old fashioned logical methods which empha-sized symbolic representation and reasoning and there wereoptimization methods that dispensed with interpretable repre-sentations altogether for ldquoblack-boxrdquo optimization and signaldetection techniques And none of it really worked Therewere hybrid systems in various labs which kluged togethersymbolic representations of this and perhaps the optimizationof an objective function for that or vice versa and those messyhacked-together systems were scaled up to industrial sizeby the likes of Google Microsoft and Amazon in the early2000s But these were basically gigantic versions of projectsthat were in university labs and a few industry labs It wasa major engineering feat to scale up these systems but theywere fragile and performance improvements came mostly frombrute-force scale very roughly speaking

That all changed in the last 5 years What happened is thatthe kluged-together symbolic and non-symbolic systems of thepast have been rapidly giving way to kluged-together optimiza-tion techniques (aka ldquoDeep Learningrdquo) which are breakingthrough previous ceilings on performance are less fragile andare easier to maintain Image classification is the primary ex-ample But there are a range of successful applications thatmay not immediately appear to be amenable to the same tech-niques such as natural language translation Microsoft recentlyannounced a speech-to-text translation system that achieves a51 error rate which matches the best human performance intranscribing spoken language to written text Skype which isowned by Microsoft may soon have real-time language trans-lation capabilities allowing two people to hold a conversationin their native tongues So we have a pretty good workingtechnology for artificial perception mdash which is amazing At

2

the same time this is only one part of AIReturning to your question about epistemology Nobody mdash

setting aside the singularity fringe mdash thinks that these recentadvances in machine learning will yield up the judgment andcommon-sense reasoning that is currently missing from thesesystems That said the variety of problems that can be reducedto a perception problem is staggering where correct classifi-cation is enough to achieve desirable goal I expect that wewill continue to be surprised by the range of problems that willbe cracked by these methods Correct classification is anothername for finding the truth or making a reliable judgment Whatis surprising is the every-growing domain of problems wheretruths can be learned and reliable judgments be made withoutmuch understanding at all The link between prediction and ex-planation which underpins data models in inferential statisticsand Bayesian statistics alike and pervades epistemology hasbeen cut So much for evidentialism

Now to be sure there are good reasons to restore this con-nection between prediction and explanation if you are turneddown for a loan it is fair to ask for a reason why Indeed theEuropean General Data Protection Regulations (GDPR) thatwill go into effect next year requires that such an explanationbe made available to algorithmic decisions involving EU citi-zens But the point is that the connection between explanationand accurate prediction is strictly unnecessary We have mod-els that make effective predictions but which are incapable ofyielding an explanation

SB Do you think advances in machine learning are going tochange the debate about philosophy of mind

GW The steam-engine Telephone switching centers Theintuitive statistician The mainframe computer All of thesetechnologies were used as metaphors in psychology So per-haps it is inevitable that as machine learning techniques spreadacross society and the sciences we will see those ideas in-fluence how we understand ourselves or the world around usI saw a conference paper earlier this year that presented amodel and evidence for how neurons in our brains performsback propagation What was once a knock-down argumentagainst artificial neural networks (ANNs) having anything to dowith brain sciencemdash ANNs rely on back-propagation brainsdonrsquotmdash is now a subject of inquiry in brain science

There is some very interesting work by Facebookrsquos AI groupin creating object masks for images all built on a convolutionalneural network architecture This system can pick out occludedor partial objects from a photo and accurately identify themSo a photo with a ball the back of a personrsquos head part of aTV screen et cetera can be picked out as individual objectsmasked by a border and correctly labeled This is a big leapbeyond classifying an entire photo as one that includes a balla person a TV et cetera which was the state of the art a fewyears ago Yet this capability is precisely the sort of achieve-ment that a layman may well see and say ldquoso whatrdquo A childcould take a marker trace around objects in a photo and writedown a correct label Because people are so good at this taskit is understandable why we do not realize how difficult it isfor a machine to do this And that this has been done on topof an R-CNN architecture is incredible This is a small steptoward the missing ldquoreasoningrdquo and ldquorepresentationrdquo that mo-tivates logical approaches to AI but these capabilities are froma ldquobottom-uprdquo fashion From the point of view of analyticalphilosophy particularly those branches that remain steeped inlogic and language the details of this algorithm will appear

completely backwardsHere is one implication for the philosophy of mind in broad

strokes It is not uncommon for philosophers of mind to viewbehavior in terms of agency and to understand agency in termsof language in general and ideas about languages from the phi-losophy of language in particular Letrsquos face it analytic phi-losophy is rooted in language But the advent of systems thatbegin with effective behavior and work backwards to proto-representations reverse the implications throwing into doubtchains of reasoning that ascribe agency to robots or passivesystems on the basis of purportedly intensional behavior andmisguided ideas of what is mentally necessary to realize suchbehavior Similar to the break between explanation and predic-tion the role of language and representation in effective behav-ior will call for reevaluation Here again there are good reasonsto tie together language and action But the presumption thereare intimate and necessary links between language and practi-cal action which is a legacy of 20th century analytic philoso-phy is challenged by the performance of these systems and inany event the last centuryrsquos obsession with language will notsuit philosophy for the current century The reign of languageis over

News

Logic in the Wild 9ndash10 November

The workshop Logic in the Wild was held on November 9thand 10th in Ghent Belgium It was the sixth workshopin the Logic Reasoning and Rationality series supported bythe Research Foundation Flanders (FWO) through the scien-tific research network on Logical and Methodological Analy-sis of Scientific Reasoning Processes The network brings to-gether research groups from nine European universities carry-ing out research on relevant topics Adam Mickiewicz Uni-versity Pozna Free University of Brussels Ghent UniversityRuhr-University Bochum Tilburg University University Col-lege London University of Antwerp Utrecht University andVU University Amsterdam For the duration of the projectfrom 2015 till 2019 there are two workshops organized peryear one in spring and one in autumn

The workshop was organized by the Centre for Logic andPhilosophy of Science (Ghent University) which coordinatesthe activities of the network and the Department of Logicand Cognitive Science (Adam Mickiewicz University PoznanPoland) Its title Logic in the Wild stemmed from Keith Sten-ning and Michiel van Lambalgenrsquos seminal book Human Rea-soning and Cognitive Science (MIT Press 2008) in which theauthors both advocate for and exemplify the productivity of theparadigm called a lsquopracticalrsquo or cognitive turn in logical re-search The approach draws on enormous achievements of alegion of formal and mathematical logicians but focuses onthe Wild actual human processes of reasoning and argumenta-tion Moreover high standards of inquiry that we owe to formallogicians offer a new quality in research on reasoning and ar-gumentation In terms of John Corcorans distinction betweenlogic as formal ontology and logic as formal epistemology theaim of the practical turn is to make formal epistemology evenmore epistemically oriented This is not to say that this prac-tically turned (or cognitively oriented) logic becomes just apart of psychology This is to say that this logic aquires a new

3

the same time this is only one part of AIReturning to your question about epistemology Nobody mdash

setting aside the singularity fringe mdash thinks that these recentadvances in machine learning will yield up the judgment andcommon-sense reasoning that is currently missing from thesesystems That said the variety of problems that can be reducedto a perception problem is staggering where correct classifi-cation is enough to achieve desirable goal I expect that wewill continue to be surprised by the range of problems that willbe cracked by these methods Correct classification is anothername for finding the truth or making a reliable judgment Whatis surprising is the every-growing domain of problems wheretruths can be learned and reliable judgments be made withoutmuch understanding at all The link between prediction and ex-planation which underpins data models in inferential statisticsand Bayesian statistics alike and pervades epistemology hasbeen cut So much for evidentialism

Now to be sure there are good reasons to restore this con-nection between prediction and explanation if you are turneddown for a loan it is fair to ask for a reason why Indeed theEuropean General Data Protection Regulations (GDPR) thatwill go into effect next year requires that such an explanationbe made available to algorithmic decisions involving EU citi-zens But the point is that the connection between explanationand accurate prediction is strictly unnecessary We have mod-els that make effective predictions but which are incapable ofyielding an explanation

SB Do you think advances in machine learning are going tochange the debate about philosophy of mind

GW The steam-engine Telephone switching centers Theintuitive statistician The mainframe computer All of thesetechnologies were used as metaphors in psychology So per-haps it is inevitable that as machine learning techniques spreadacross society and the sciences we will see those ideas in-fluence how we understand ourselves or the world around usI saw a conference paper earlier this year that presented amodel and evidence for how neurons in our brains performsback propagation What was once a knock-down argumentagainst artificial neural networks (ANNs) having anything to dowith brain sciencemdash ANNs rely on back-propagation brainsdonrsquotmdash is now a subject of inquiry in brain science

There is some very interesting work by Facebookrsquos AI groupin creating object masks for images all built on a convolutionalneural network architecture This system can pick out occludedor partial objects from a photo and accurately identify themSo a photo with a ball the back of a personrsquos head part of aTV screen et cetera can be picked out as individual objectsmasked by a border and correctly labeled This is a big leapbeyond classifying an entire photo as one that includes a balla person a TV et cetera which was the state of the art a fewyears ago Yet this capability is precisely the sort of achieve-ment that a layman may well see and say ldquoso whatrdquo A childcould take a marker trace around objects in a photo and writedown a correct label Because people are so good at this taskit is understandable why we do not realize how difficult it isfor a machine to do this And that this has been done on topof an R-CNN architecture is incredible This is a small steptoward the missing ldquoreasoningrdquo and ldquorepresentationrdquo that mo-tivates logical approaches to AI but these capabilities are froma ldquobottom-uprdquo fashion From the point of view of analyticalphilosophy particularly those branches that remain steeped inlogic and language the details of this algorithm will appear

completely backwardsHere is one implication for the philosophy of mind in broad

strokes It is not uncommon for philosophers of mind to viewbehavior in terms of agency and to understand agency in termsof language in general and ideas about languages from the phi-losophy of language in particular Letrsquos face it analytic phi-losophy is rooted in language But the advent of systems thatbegin with effective behavior and work backwards to proto-representations reverse the implications throwing into doubtchains of reasoning that ascribe agency to robots or passivesystems on the basis of purportedly intensional behavior andmisguided ideas of what is mentally necessary to realize suchbehavior Similar to the break between explanation and predic-tion the role of language and representation in effective behav-ior will call for reevaluation Here again there are good reasonsto tie together language and action But the presumption thereare intimate and necessary links between language and practi-cal action which is a legacy of 20th century analytic philoso-phy is challenged by the performance of these systems and inany event the last centuryrsquos obsession with language will notsuit philosophy for the current century The reign of languageis over

News

Logic in the Wild 9ndash10 November

The workshop Logic in the Wild was held on November 9thand 10th in Ghent Belgium It was the sixth workshopin the Logic Reasoning and Rationality series supported bythe Research Foundation Flanders (FWO) through the scien-tific research network on Logical and Methodological Analy-sis of Scientific Reasoning Processes The network brings to-gether research groups from nine European universities carry-ing out research on relevant topics Adam Mickiewicz Uni-versity Pozna Free University of Brussels Ghent UniversityRuhr-University Bochum Tilburg University University Col-lege London University of Antwerp Utrecht University andVU University Amsterdam For the duration of the projectfrom 2015 till 2019 there are two workshops organized peryear one in spring and one in autumn

The workshop was organized by the Centre for Logic andPhilosophy of Science (Ghent University) which coordinatesthe activities of the network and the Department of Logicand Cognitive Science (Adam Mickiewicz University PoznanPoland) Its title Logic in the Wild stemmed from Keith Sten-ning and Michiel van Lambalgenrsquos seminal book Human Rea-soning and Cognitive Science (MIT Press 2008) in which theauthors both advocate for and exemplify the productivity of theparadigm called a lsquopracticalrsquo or cognitive turn in logical re-search The approach draws on enormous achievements of alegion of formal and mathematical logicians but focuses onthe Wild actual human processes of reasoning and argumenta-tion Moreover high standards of inquiry that we owe to formallogicians offer a new quality in research on reasoning and ar-gumentation In terms of John Corcorans distinction betweenlogic as formal ontology and logic as formal epistemology theaim of the practical turn is to make formal epistemology evenmore epistemically oriented This is not to say that this prac-tically turned (or cognitively oriented) logic becomes just apart of psychology This is to say that this logic aquires a new

3

task of ldquosystematically keeping track of changing representa-tions of informationrdquo as Johan van Benthem puts it and thatit contests the claim that the distinction between descriptiveand normative accounts of reasoning is disjoint and exhaus-tive From a different than purely psychological perspectivelogic becomesmdashagainmdashinterested in answering Deweyrsquos ques-tion about the Wild how do we think This is the new alluringface of psychologism or cognitivism in logic as opposed tothe old one which Frege and Husserl fought against And thiswas the area of research to which this workshop was devoted

The workshop brought together 23 participants who pre-sented talks on applications of logic to analyses of natural lan-guage and everyday reasoning phenomena The keynotes weredelivered by Iris van Rooij (Radboud University) Keith Sten-ning (University of Edinburgh) and Christian Strasser (RuhrUniversity Bochum)

In her talk lsquoCognition in the wild logic and complexityrsquoIris van Rooij addressed the issue of computational intractabil-ity of models of cognition Van Rooijrsquos proposal proposal isthat cognitive science should recognize tractability as a funda-mental constraint on cognition in the wild She explained howthe tractability constraint can serve as a formal guide in theorydevelopment and furthermore illustrated how logic-based ap-proaches may especially benefit from this approach as it mayenlarge their recognized scope and relevance for cognitive sci-ence

Keith Stenning started his talk lsquoMemory is the organ of non-monotonic reasoningrsquo with a question Nothing is wilder thanthe human mind He outlined a program of research whichuses Logic Programming (in a particular flavour) as a modelof human semantic memory in the service of nonmonotonicreasoning to an interpretation He claimed that applying LPto memory will serve as an example of a relation between logicand the mind and hopefully motivate some researchers of a log-ical bent to collaborate with the kind of empirical work whichneeds to go on Stenning warned that there is a great danger onboth sides of the cognitivelogical fence of underestimating thedensity of the problems which live down this crack The psy-chologist who denies the relevance of logics lsquonormativersquo sys-tems is as numerous as the logician who thinks that his (usuallybut not always lsquohisrsquo) newly invented logic is straightforwardlya contribution to how human reasoning works

Christian Strasserrsquos talk lsquoReasoning by cases (RbC) in thenonmonotonic wildernessrsquo was concerned with is an inferencescheme especially apt for situations in which we deal with in-complete information He discussed some challenges for defea-sible accounts of RbC highlighted shortcomings of approachesto RbC from the literature on non-monotonic logic and pre-sented a new account of a defeasible variant of RbC based onformal argumentation

Rafal UrbaniakErikWeber

Ghent UniversityMariusz Urbanski

Adam Mickiewicz University Poznan

Calls for Papers

Disagreement Perspectives from Argumentation Theory andEpistemology special issue of Topoi deadline 31 JanuaryDecision Theory and the Future of Artificial Intelligence

special issue of Synthese deadline 15 FebruaryDefeasible andAmpliative Reasoning special issue of Interna-tional Journal of Approximate Reasoning deadline 15 Febru-aryNon-Classical Modal and Predicate Logics special issue ofLogic Journal of the IGPL deadline 30 April

Whatrsquos Hot in

Mathematical Philosophy

Reviving the present column is a good resolution for 2018The plan is that members and friends of the Munich Centerfor Mathematical Philosophy (MCMP) will take turns to writeit

To kickstart this Irsquod like to share some thoughtsabout one decision-theoretic issue that has been bug-ging me recently The issue arises in Savagersquos frame-work At first it looks like a terminological puzzle ofsorts But it proves to be more substantial than that

In Savagersquos frameworkthe options between whichthe decision-maker choosesare acts ie functions froma set of states to a set of pay-offs (aka consequences)Assume that the name ofthe game is the followingYoursquore supposed to observethe decision-makerrsquos choicesbetween Savagian acts andbased on that information toidentify her beliefs about the likelihood of the states and herpreferences between the payoffs As is well known in Savagersquosown take on this identification exercise the decision-makerrsquosbeliefs are quantified by a subjective probability function herpreferences by a utility function and her observed choices con-form to the rule of maximizing subjective expected utility As isequally well known there are many troublesome cases whichSavagersquos work was instrumental in identifying where such amodel is not applicable

One such troublesome case seems to be indifferently re-ferred to as ldquoact-state dependencerdquo or ldquomoral hazardrdquo in mostof the literature The intuition is as follows In some casesthe decision-makerrsquos beliefs about the likelihood of the stateswill somehow depend on the Savagian act under considera-tionmdashwhence ldquoact-state dependencerdquomdashand this in turn is bestunderstood with reference to the ldquomoral hazardrdquo cases stud-ied in economicsmdashie essentially cases where the decision-makerrsquos choices can somehow influence the likelihood of theevents of interest Now here comes the question to which Iwant to draw your attention Are rdquoact-state dependencerdquo andrdquomoral hazardrdquo synonyms in decision theory Irsquom not deny-ing that there are conditions under which they can be treated assuch Irsquom asking whether there are not also cases where theycannot

It turns out that there are indeed cases where act-state de-pendence and moral hazard come apart Let me start with thesimplest of the two stories which I need to tell namely thatof act-state dependence without moral hazard In fact you

4

could here take as an example many non-expected utility mod-elsmdashprovided you look at them from the right angle Considersay Gilboa and Schmeidlerrsquos familiar max-min expected util-ity model It is usually presented as a multi-prior model inwhich the decision-maker maximizes over the set of priorsminimum expected utility But this means that equivalentlyyou can think of the model as attributing to each Savagian actone act-dependent prior Thus sticking to the standard (cuttinga long story short epistemic) interpretation of the set of priorswe started from you will have act-state dependence withoutmoral hazard

Let me now turn to the other story to be told ie that ofmoral hazard without act-state dependence The situation issomewhat more subtle But the literature on moral hazard ineconomics provides inspiration The most fundamental issueis after all whether the decision-maker can exert some influ-ence on the likelihood of the states more than how she mightexert it or whether it will always be in ways that are observ-able to us Accordingly assume that there are some unobserv-able side actionsmdashcrucially do not confuse these side actionswith the Savagian actsmdashby which the decision-maker can ifshe so wishes influence the likelihood of the states Other-wise keep your favorite interpretation of the Savage game un-changed (Or if it helps you get a grip on the twist which Irsquomproposing think of each Savagian act simply as a bet on thelikelihood of the states with payoffs such as banknotes teddybears or cotton candiesmdashwhatever will cure you from the un-necessary philosophical assumptions restricting many interpre-tations of the Savage framework) Thus in general you willhave moral hazard without act-state dependence

I conclude that in general neither act-state dependence normoral hazard implies the other Therefore in general act-statedependence and moral hazard should not be used interchange-ably in decision theory I cannot elaborate here on the impli-cations of this simple observation I can only say that they arenon-trivial and arguably significant

Jean BaccelliLudwig-Maximilians-Universitat Munchen

Medieval ReasoningFor readers accustomed tothe mathematical symbolismof contemporary logicrsquos for-mal apparatus one of themost peculiar yet interestingfeatures of medieval logicis its use of an extremelyregimented version of me-dieval Latin as a logical lan-guage along with the lackof a clear-cut distinction be-tween object-language andmeta-language A few weeksago trying to introduce regimentation to the students in myclass on Buridan and 14th century nominalism I asked themldquoWhat is a formal languagerdquo One of the answers surprisedme ldquoIt is a language in its higher form as it is codified by anAcademy or a similar cultural institution ndash as the Academiefrancaise or the Oxford English Dictionaryrdquo What that studenthad in mind was the sociolinguistic notion of a formal register

for what is commonly (however misleadingly) called ldquonaturallanguagerdquo However unexpected it was certainly an interestinganswer because that sense of ldquoformalrdquo still carries the idea ofa kind of speechwriting employing a specific vocabulary andstructured rigorously according to officially sanctioned gram-matical syntactical and semantic rules While the vocabularythe rules and those who do the sanctioning are vastly differ-ent if we take a formal language to be defined as a collectionof strings on a fixed alphabet with explicitly stated formationrules then the sense in which my student intended a language tobe formal is not too far removed from the one in which a logicallanguage is formal Certainly formal logical languages havea degree of abstraction symbolisation and de-semantificationthat sets them apart from natural languages but it is not obviousat all that symbolisation and de-semantification are sufficient oreven necessary conditions to make a language formal Beyondrestating Montarguersquos and formal linguisticsrsquo thesis that natu-ral languages can be treated as formal languages one could putforth the stronger claim that there is a sense in which formalityis not an exclusive property of logical languages but one that isshared by non-symbolised non-de-semantified and non-highly-abstracted languages too If so then there isnrsquot such a big dif-ference between formalisation and regimentation or betweensymbolically formalised languages and regimented languagessuch as Medieval Logical Latin ndash or even the formal register ofan ordinary language sanctioned by an authoritative AcademyThe difference that is there looks like a difference of degreebut not a qualitative one To explore it Medieval Logical Latinappears to be the best suited case study it is as highly artificialas a a rdquonaturalrdquo language can get it has several rules extendingreforming and improving the ordinary grammatical and syntac-tical rules of Latin and regimenting its semantics and last butnot least it was actually used as a logical language TBC

Curious Stay tuned till next episode

Graziana CiolaPhilosophy Scuola Normale Superiore Pisa

Philosophy and Economics

As you read this a New Year will have begun for you As Iwrite this however I am just starting to say Goodbye to theold one You see what separates us is a crucial timespan ofjust a few days and weeks in which the old year is essentiallydone and the New Year has not yet begun Most universitiesand academic matters in the Western hemisphere slow down totheir eventual seasonal halt at the end of the year And we get totake a step back and muse over things future and past We thusspend some time in a period that does not belong to any par-ticular year as the German expression ldquozwischen den Jahrenrdquohas it We find ourselves liberated from getting things done inthe old year ndash whether you have given up simply stopped oraccomplished what there was to accomplish does not matter ndashbut we do not yet have hit the ground running in the New YearIt is your time between the years

Now that I have you in a reflective mood I would like toinvite you to take a note of some Reasoner-related philosophersand economists that we lost in 2017 read up on some of theirobituaries and make sure their work stays with us in the yearsto come I am limiting myself to mentioning two philosophersand two economists

The two philosophers are Derek Parfit and Delia Graff

5

Fara Parfit passed away on the first day of 2017 Theblog Daily Nous has a very long list of remembrances andobituaries many of which are truly fascinating Fara passedaway at a still young age but her work in the philoso-phy of language both on names and descriptions and onvagueness was already ndash and will probably continue to be ndashvery influential Here is her Princeton obituary and an en-try in the blog Daily Nous which contains another obituary

The two economists areTony Atkinson and Ken-neth Arrow Arrow mightneed less of an introductionto this particular Reasoner-crowd and you can findmany of his works on hisStanford page Atkinsonwho also passed away onthe very first day of 2017is most famous for his workon poverty and inequalityThere is a website dedicatedto his work with a well-maintained bibliography Philosopher-economists will be familiar with his 2001 article lsquoThe strangedisappearance of welfare economicsrsquo (Kyklos 54193-206) butthere is also a 2009 article lsquoEconomics as a moral sciencerdquo(Economica 76791-804) that argues in a similar directioneconomics should be understood as studying the assumptionsof normative statements The 2009 article is not only updatedbut frames the argument more broadly than its 2001 cousinand discusses the relationship between welfare economics andthe capability approach Many of Atkinsonrsquos articles containmethodological reflections but this one will be the most pro-nounced

I have been happily browsing and reading on the work andobituaries of all the aforementioned Books and papers andprojects related to the work of these four philosophers andeconomists are all candidates for going on my todo-list for2018 Unlike you I have the advantage of the period betweenthe years still ahead to figure this out But I hope some of thepointers you find here give you the opportunity to read some-thing you havenrsquot yet for a slower pace of your start into theNew Year

Conrad HeilmannErasmus Institute for Philosophy and Economics (EIPE)

Erasmus University Rotterdam

Events

January

ASET 6th World Conference on Applied Science Engineeringand Technology India 2ndash3 JanuarySEaC The View from Above Structure Emergence and Cau-sation University of Oxford 11ndash12 JanuaryPhiloMaLogi Cambridge Graduate Conference on the Philos-ophy of Mathematics and Logic University of Cambridge 20ndash21 JanuaryBigDat 4th International Winter School on Big Data Roma-nia 22ndash26 JanuaryTaD New Perspectives on Truth and Deflationism Universityof Salzburg 26ndash27 January

February

AppMathampComSci International Conference on AppliedMathematics and Theoretical Computer Science India 1ndash3FebruaryPresDat Presenting Data London 6 FebruaryBDAAIampCL Big Data Analytics AI and Machine LearningOntario Canada 7 FebruaryMathStatCompSci International Conference on Advances inMathematics Statistics and Computer Science Dubai 9ndash10FebruaryA-SRSI Agent-Specificities and Relationships in Social Inter-actions Cologne Germany 15ndash16 FebruaryOnBlf Interdisciplinary Workshop on Belief New York 15ndash16 FebruaryCMampEval Causal Modelling and Evaluation Kings CollegeLondon 19ndash23 FebruaryMiN-CLogic Doing Metaphysics in Non-Classical Logic Lis-bon 22-23 February

Courses and Programmes

CoursesSIPTA 8th School on Imprecise Probabilities Oviedo 24ndash28July

6

Computer SimulationMethods Summer School High Perfor-mance Computing Center Stuttgart (HLRS) 25ndash29 September

ProgrammesAPhil MAPhD in Analytic Philosophy University ofBarcelonaMaster Programme MA in Pure and Applied Logic Univer-sity of BarcelonaDoctoral Programme in Philosophy Language Mind andPractice Department of Philosophy University of ZurichSwitzerlandDoctoral Programme in Philosophy Department of Philoso-phy University of Milan ItalyHPSM MA in the History and Philosophy of Science andMedicine Durham UniversityMaster Programme in Statistics University College DublinLoPhiSC Master in Logic Philosophy of Science and Epis-temology Pantheon-Sorbonne University (Paris 1) and Paris-Sorbonne University (Paris 4)Master Programme in Artificial Intelligence Radboud Uni-versity Nijmegen the NetherlandsMaster Programme Philosophy and Economics Institute ofPhilosophy University of BayreuthMA in Cognitive Science School of Politics InternationalStudies and Philosophy Queenrsquos University BelfastMA in Logic and the Philosophy ofMathematics Departmentof Philosophy University of BristolMA Programmes in Philosophy of Science University ofLeedsMA in Logic and Philosophy of Science Faculty of PhilosophyPhilosophy of Science and Study of Religion LMU MunichMA in Logic and Theory of Science Department of Logic ofthe Eotvos Lorand University Budapest HungaryMA in Metaphysics Language and Mind Department of Phi-losophy University of LiverpoolMA inMind Brain and Learning Westminster Institute of Ed-ucation Oxford Brookes UniversityMA in Philosophy by research Tilburg UniversityMA in Philosophy Science and Society TiLPS Tilburg Uni-versityMA in Philosophy of Biological and Cognitive Sciences De-partment of Philosophy University of BristolMA in Rhetoric School of Journalism Media and Communi-cation University of Central LancashireMA programmes in Philosophy of Language and Linguisticsand Philosophy of Mind and Psychology University of Birm-inghamMRes in Methods and Practices of Philosophical ResearchNorthern Institute of Philosophy University of AberdeenMSc in Applied Statistics Department of Economics Mathe-matics and Statistics Birkbeck University of LondonMSc in Applied Statistics and Datamining School of Mathe-matics and Statistics University of St AndrewsMSc in Artificial Intelligence Faculty of Engineering Uni-versity of Leeds

MA in Reasoning

A programme at the University of Kent Canterbury UK Gainthe philosophical background required for a PhD in this area

Optional modules available from Psychology ComputingStatistics Social Policy Law Biosciences and History

MSc in Cognitiveamp Decision Sciences Psychology UniversityCollege LondonMSc in Cognitive Systems Language Learning and Reason-ing University of PotsdamMSc in Cognitive Science University of Osnabruck GermanyMSc in Cognitive PsychologyNeuropsychology School ofPsychology University of KentMSc in Logic Institute for Logic Language and ComputationUniversity of AmsterdamMSc in Mind Language amp Embodied Cognition School ofPhilosophy Psychology and Language Sciences University ofEdinburghMSc in Philosophy of Science Technology and Society Uni-versity of Twente The NetherlandsMRes in Cognitive Science and Humanities Language Com-munication and Organization Institute for Logic CognitionLanguage and Information University of the Basque Country(Donostia San Sebastian)OpenMind International School of Advanced Studies in Cog-nitive Sciences University of BucharestResearchMaster in Philosophy and Economics Erasmus Uni-versity Rotterdam The Netherlands

Jobs and Studentships

JobsTemporary Lecturer in Logic and Philosophy of Mathemat-ics University of Amsterdam deadline 8 JanuaryAssociate Professor (two) in Statistics University of War-wick deadline 10 JanuaryProfessorship in Theoretical Philosophy University of Greif-swald Germany deadline 10 JanuaryLecturer in Medical Statistics University of Leicester dead-line 11 JanuaryLectureship in Statistics Lancaster University deadline 15JanuaryLecturer in Statistics University of Otago deadline 26 Jan-uary

StudentshipsPhD in Machine Learning University of Edinburgh deadline5 JanuaryPhD in Statistics University of Oslo deadline 31 JanuaryPhD in Neurophilosophy LMU Munich deadline 15 February

7