NOMINALISM AND CONVENTIONALISM IN SOCIAL CONSTRUCTIVISM

Paul Ernest University of Exeter p.ernest(at)ex.ac.uk Introduction There are various forms of social constructivism in the social and human sciences, especially in sociology (Berger and Luckmann !""), discursive psychology (#ergen !!!) and philosophy ($acking %&& ). There are also variants in education and learning theory ('ertsch !! ), in mathematics education ('ein(erg and #avelek !)*), and even in the philosophy of mathematics ($ersh !!*). $o+ever, +hat is meant (y the use of the term here is that version of social constructivism as a philosophy of mathematics given its fullest expression to date in my eponymous (ook (Ernest !!)). To avoid cum(ersome circumlocutions, no reference +ill (e made to the range of extant or possi(le variations or alternatives, although it is not intended to assert the superiority or supremacy of the singular version explicated here. ,ocial -onstructivism is put for+ard as a philosophy of mathematics +ith the primary aim of offering an account of mathematical practice, including also the social structuring and the historical development of mathematics. .t is a naturalistic philosophy and so it has many ela(orated characteristics relating to its descriptive and social aspects. $o+ever although ontological and epistemological issues are discussed in Ernest ( !!)), . +ish to focus here on its relationships to traditional ontological and epistemological positions in the philosophy of mathematics. 'hile social constructivism has (een contrasted strongly +ith /latonism and mathematical realism, on the one hand, and +ith foundationalist and a(solutist positions, on the other hand, here . +ish to ela(orate its positive location in philosophical traditions. This paper cautiously descri(es social constructivism as nominalist, +ith respect to ontology, and conventionalist +ith respect to epistemology and the foundations of kno+ledge. The caution is due to the particular variants or interpretations of nominalism and conventionalism that are attri(uted to social constructivism, for there are dominant traditions +ithin each perspective +ith +hich social constructivism is not identified or su(sumed under. By asserting that the o(0ects of mathematics are signs, rather than purely psychological or mental entities (the claim of conceptualism), material entities (the claim of materialism) or o(0ective self1su(sistent entities (the claim of realism) +hat is espoused is claimed to (e a variety of nominalism. $o+ever, this differs from the most common forms of nominalism, for it is (ased on a conception of sign that re0ects the representational theory of truth. ,igns are part of a cultural realm that is intersu(0ective. .t transcends the perceptions and understanding of any one individual (ut does not transcend the kno+ledge and practices of humankind as a +hole and thus does not (elong to any extra1human reality. Else+here . have termed this realm 2o(0ective3, (ut this re4uires a ne+ definition of the term, differing from the received, traditional usage and its ontological presuppositions (Ernest !!)). The related claim that the concepts, terms, theorems, rules of proof and logic, truths and theories of mathematics are socially constructed cultural entities constitutes a form of conventionalism. $o+ever, +hat is not asserted is that the concepts and truths of mathematics are the result of ar(itrary, +himsical or even ideologically motivated decisions and choices.

5any conventions in mathematics are not conscious decisions (ut reflections of historical practices laid do+n for very good reasons. 6urthermore, +here conscious decisions are made in mathematics they are usually to complete or extend existing rules and practices +ithin mathematics that result in general, simple, elegant, practical and consistent systems. 7s such, such choices although not actually forced or necessary, for if so they +ould not (e choices +ithin the sense meant here, are nevertheless the choices that come closest to (eing re4uired (y past traditions. Nominalism ,ocial constructivism is claimed to (e a nominalist philosophy of mathematics (ecause it asserts that the o(0ects of mathematics are signs. . understand signs in the semiotic sense of having or (eing composed of (oth a signifier, that +hich represents, often a material representation, and a signified, the meaning represented. Thus unlike in $il(ert3s formalism, the signs of mathematics are not 0ust detached and empty sym(ols (signifiers) (ut al+ays have meanings (signifieds), even if precisely specifying the characteristics and ontology of these entities is difficult and complex, and may involve am(iguity and multiplicity. 6ollo+ing /eirce3s seminal +ork in semiotics, as +ell as modern semiotic philosophy (Eco !)8, 9errida !*)), these signifieds are understood to (e further signs. 7lthough /eirce developed a complex tripartite treatment of signs, he is unam(iguous in asserting that :The meaning of a representation can (e nothing (ut a representation.; (/eirce !< 1=), >ol. , ,ection <<!). .n /eirce3s system the 2interpretant3 of a sign corresponds in many respects to the meaning or interpretation of a sign. ,o there is an infinite regression here. 6inally, the interpretant is nothing (ut another representation to +hich the torch of truth is handed along? and as representation, it has its interpretant again. (/eirce !< 1=), >ol. , ,ection <<!). 7lthough there is potentially an infinite regression in this theory, this is no more a vicious circle than looking up meanings in a dictionary. Ultimately the dictionary meaning of a +ord can only (e given in terms of relationships (et+een other +ords and their meanings. ,o too, . claim, the meanings or signifieds of virtually all signs in mathematics are themselves further signs. . do not claim that every meaning is itself a sign (ecause . +ant to leave open t+o further possi(ilities@ first, that signifieds can (e actions or operations on signs, and second, that they can (e tangi(le o(0ects in the real +orld. . do not at this stage think that the second case is relevant to mathematics, for if, for example, one points to a set of three o(0ects as an instance of the num(er three, in my vie+ this indicated plurality or set constitutes a sign itself, rather than 0ust the meaning of a sign. .ndicating a set of tangi(le o(0ects in effect makes it a sign itself. 6rom a nominalistic perspective, it +ould not (e possi(le to consistently adopt any other
'ittgenstein ( !=<) uses a (roader notion of choice in follo+ing a rule, claiming that any rule follo+ing involves an agreement or decision, in the sense that choosing to participate in a language game and form of life and maintain its rules and conventions, such as respecting 5odus /onens in mathematical proof, is al+ays a matter of choice, not necessity. 'ithout challenging this, the choices, conventions and decisions . am referring to are those that come a(out +hen there is no une4uivocal and unam(iguous already laid do+n rule to follo+. Thus 'illiam $amilton in inventing, constructing and defining the system of Auaternions chose to a(negate commutativity in the (inary operation in his system (i.0 B 10.i 0.i) to o(tain the (est system +ith the properties he sought, although other options +ere open to him (/ickering !!=). This +as not a permitted move in the contemporary language games of alge(ra, and thus led to the formation of ne+ language games that have proved very fruitful in mathematics.

definition of meanings except in terms of other signs. 6or any other definition +ould s+iftly lead out of a nominalist theory into something else. .f meanings +ere in general located in the physical +orld, this +ould give rise to an empirical realist ontology. .f the meanings +ere located in some a(stract realm, the outcome +ould also (e a realist ontology of some a(stract sort, like /latonism for example. Like+ise meanings located in the mind +ould give rise to a form of conceptualism. Thus the ontological position of social constructivism is nominalism. The o(0ects of mathematics are signs, and furthermore the meanings of these signs are typically yet further signs. $o+ever, as indicated a(ove, this is not to say that all that exists in mathematics is signs. There are in addition sign related activities@ idealiCed human action on signs. Thus the numeral 2<3 connotes (oth the act of esta(lishing a one1to1one correspondence +ith prototypical triplet set (cardinality), and the act of enumerating a triplet set (ordinality). Each of these connotations presupposes some elements of threeness, re4uiring the use of a representative triplet set. But the explicit definition of < as the successor of % does not. 6ormally this definition can (e represented as < B def ,%, +here , is the one place operation of successor, a primitive in the theory of /eano arithmetic. 7lthough this identity asserts a static relationship, it contains an operation, namely that of applying the successor operation to %. Traditionally mathematics is understood as timeless, since the idealiCed act of transforming % into its successor ,% (commonly named 2<3) is reported after the idealiCed act, +ith no temporal frame+ork. Devertheless, the admission and representation of change signifies an a(stracted and idealiCed analogue or counterpart to time. Ef course the idealiCed analogue of time admitted in mathematics differs in one ma0or respect from time in its usual sense, it is reversi(le. ,1 < B %, 0ust as truly as ,% B <, +here , 1 denotes the inverse operation of , (the predecessor operation). .n learning the meaning of the e4uals sign (B) children usually first understand it as a sign of se4uential transformation (see, e.g., Fieren !!%). ,o %G< B = is read as % added to < makes =, in +hich the e4uals sign signifies the move to the end product of an operation. -onse4uently, initially children often have difficulty in grasping its symmetric form = B %G< since this is not easily understood in these terms.% 7 necessary step in the successful learning of mathematics is to grasp the symmetric property of e4uality, as +ell as its reflexivity and transitivity. Like+ise, these three properties must also (e understood to apply to the analogous relation of logical e4uivalence that applies to formula and sentence pairs. Thus developing an understanding of the signs of mathematics involves a(stracting from the time se4uential feature of operations, (ecause of the a(stract and timeless character of the signified space of mathematical operations and o(0ects.< Ef course the signs of mathematics are nuanced enough in their meanings so that potentially irreversi(le se4uences of operations can (e represented too, through (oth ordered relations on terms and formulas (using, e.g., H and , respectively) and reductions in complexity of terms and formulas (using, e.g., su(stitutivity of terms). The spaces of transformations of mathematical signs can have (oth a(elian and non1a(elian group1like properties (although in most case they do not strictly speaking form a group). The discussion of this very simple example (egins to illustrate some of the richness and complexity of mathematical signs and their meanings, understood semiotically and
% This is related to the need for premature closure noted and defined (y -ollis ( !*=). .t represents the empirically noted desire of children in the early stages of learning mathematics to achieve syntactical simplicity (y deriving a single ans+er in +orking a mathematical task. < .ndeed /iaget ( !=%) had the prescience to identify the stage of (eing a(le to reverse informal mathematical and logical operations as a crucial step in the development of children3s mathematical thinking. The achievement of .nvariance in this sense signals the transition to the stage of -oncrete Eperations.

nominalistically. There are signs, sign rules, and meanings for (oth signs and their rules. 6urthermore these meanings occur in t+o forms, first as agreed (y the mathematical and educational communities (shared or conventional meanings), as +ell as learners3 o+n developing and sometimes idiosyncratic interpretations of them. To understand even a part of a simple mathematical topic involves mastering a complex set of inter1relationships (et+een signs, sign rules, and sign meanings, +hich themselves em(ody complex relationships (et+een yet further signs and rules of sign use. Thus the nominalistic claims of social constructivism are not purely theoretical, (ut also descri(e some of the dimensions of coming to kno+ and understand in mathematics, that are presupposed if not al+ays ackno+ledged (y mathematicians and successful users of the su(0ect. 7CCouni ( !!8) distinguishes t+o forms of nominalism. The first form simply denies that there are a(stract mathematical o(0ects. This form leads to a variety of pro(lems. 7s Auine ( !=<, !"!) has argued, allo+ing 4uantifiers to range over classes of mathematical o(0ects, such as sets or num(ers, and then denying that these o(0ects exist, is pro(lematic. To (e carried through consistently this position leads to cum(ersome circumlocutions in +hich the o(0ects are used (ut then defined a+ay, as mere faons de parler in place of complex and multi1layered definitions. This is not the nominalist position of social constructivism. The second form of nominalism does not deny that mathematical o(0ects exist, (ut instead concerns :the identification of them +ith some of the notation supposedly referring to them; (7CCouni !!8@ 8*). This more or less corresponds +ith the position adopted here. 7CCouni goes on to call this 2nominalism on the cheap3 (ecause it avoids the complex or in his metaphor, costly, circumlocutions of the first form. $e makes a final telling remark +hich dra+s him close to the ontological position of social constructivism@ :mathematical o(0ects are posits, and posits are not, strictly speaking, independent of their positors.; (7CCouni !!8@ % 8? original emphasis throughout). Like+ise, the nominalist position adopted here is that mathematical o(0ects are the meanings of signs, themselves signs or actions and operations upon them. But then this raises the further pro(lem of +hat signs are and in +hat space they are to (e found. 7s the previous 4uotation suggests, the social constructivist position is that they are humanly made posits, (ut this of itself does not clarify their ontology. 7 useful starting point for classifying spaces or domains of existence is /opper3s ( !*!) < +orlds, the physical, mental and o(0ective. 'hile . do not su(scri(e to any a(solute tripartite division such as this, it provides a useful frame+ork for discussing signs and kno+ledge that captures many of the traditional and still +idespread notions of ontology. 7s signs, the o(0ects of mathematics do not fall neatly into any one of /opper3s < +orlds. 'orld is the material and physical +orld. ,igns have material representations in this domain, and cannot exist +ithout them, (ut they are more than these representations. 6or example t+o utterances of the same sign each have uni4ue material representations and yet are representations of the same sign. The relation of identity that holds (et+een them, and a fortiori (et+een any t+o tokens of the same type, is not materially present except as represented in a third possi(ly un+ritten sign. 7t the next higher level of a(straction, t+o apparently identical signs in different contexts often +ill have different meanings. Thus there is more to a sign than its mere material representations. There are also relationships (et+een signs manifested as human understandings and rules of sign use. 'orld % is the mental or psychological +orld. ,igns undou(tedly give rise to personal

meanings +hich some might +ish to locate in this +orld.8 But if meanings +ere solely located here there +ould (e pro(lems +ith communication and agreement. = 'ittgenstein ( !=<), in his /rivate language argument, makes the case that +e do not and cannot have private languages that refer to our private sensations and experiences. Languages are primarily pu(lic, as they are deployed and developed in social language games, +hich are part of shared forms of life. Enly after +e (egin to ac4uire mastery of language through pu(lic use and performance do +e internaliCe and appropriate them to our personal +orld of meanings. 'orld < is the o(0ective +orld of concepts, meanings, pro(lems, kno+ledge, etc. The positing of this +orld as the repository of o(0ective meanings and the o(0ects of mathematics provides the (asis for realism and /latonism. This is an elegant and self1consistent family of ontological positions that has satisfied scholars from /lato to 6rege, #Idel, and /opper, as +ell as most mathematicians and philosophers of mathematics. /art of its appeal is that it accommodates the undenia(ly o(0ective and impersonal aspects of mathematics so much (etter than Empiricism and -onceptualism. The social constructivist solution is to adopt 'orld < as a domain of o(0ective kno+ledge, (ut to redefine it as social and cultural. This is to adopt, at least in part, the social theory of o(0ectivity proposed (y Bloor ( !)8), $arding ( !)"), 6uller ( !))) and others. 'hat . mean (y saying that o(0ectivity is social is that the impersonal and sta(le character that attaches to some of our (eliefs, and the sense of reality that attaches to their reference, derives from these (eliefs (eing social institutions. (Bloor !)8@ %%!). Bloor argues that /opper3s +orld < can defensi(ly and fruitfully (e identified +ith the social +orld. $e also argues that not only is the three1fold structure of /opper3s ontology preserved under this transformation, (ut so are the connections (et+een the three +orlds. $o+ever, this social interpretation does not preserve /opper3s meaning for o(0ectivity, although it accounts for most features of o(0ectivity@ the autonomy of o(0ective kno+ledge, its external thing1like character, and its independence from any kno+ing su(0ect3s su(0ective kno+ledge. 7pplying this to the o(0ects of mathematics, the claim is that these o(0ects are signs and their operations. ,igns and sign operations have material representations ('orld manifestation), they give rise to and evoke meanings in people ('orld % manifestation), (ut they are social institutions ('orld < manifestation). ,igns and their operations are social institutions (ecause they rest on shared functions, rules and agreements and these govern their uses and shared meanings. ,ign usage and meanings are (oth learned and validated in pu(lic, in 'ittgensteinian terms, in language games situated in social forms of life. $o+ever, the pu(lic rules and agreements underpinning the use of signs are not all explicit. ,ome of them are tacit, em(edded in custom and accepted practices, and ac4uired implicitly through social participation in language games. 7 good analogy, although it is a special case, is the learning of the rules of grammar through participation in linguistic practice. 5any grammatical speakers cannot explicitly state the rules of grammar. Jather they have induced them, as if (y osmosis, from the patterns of accepted (and criticiCed) language use they have experienced as speakers and listeners. -onse4uently, they have learned them implicitly as patterns that guide the production and reception of speech, and similarly, of +ritten expressions.
8 5y cautious phrasing is (ecause in social theories of mind the personal need not (e identified +ith the psychological. ,ee, for example, $arrK and #illett ( !!8) >ygotsky ( !*)) and 'ertsch ( !!*). = This is a key pro(lem for versions of .ntuitionism that claim that the o(0ects of mathematics are personal concepts, (ut that different person making individual acts of construction arrive at identical concepts.

Unlike +ritten alpha(etic language, mathematical signs are not simply constructed to express or tell a story, +ith superfluities and decorations and the possi(ility of many su(stitutions of +ords or the addition of phrases or sentences. 5athematical signs are closely +orked in transformational se4uences to express operations on signs, (e they primarily calculations or proofs. To fulfill these roles, satisfy the underlying rules, and correctly express the compound operations involved, there are very tight constraints on the selection and uses of signs. Each sign must (e 0ustifia(ly linked through rule applications to its predecessors and its successors in the se4uence of signs that constitutes a calculation or proof. ,uch rule applications ensure that there is a continuity and preservation of meaning, such as numerical value or truth value, throughout the length of the sign se4uence. 7(ove all, a se4uence of mathematical signs is goal directed, it is intended to end +ith a calculated value or proved theorem. The meanings underlying mathematical signs, themselves signs, are the o(0ects of mathematics, in the case of terms, or statements of relationships (et+een the o(0ects of mathematics, in the case of formulas and sentences. There is no other realm of o(0ects to +hich the language of mathematics signs points. The analogy +ith alpha(etic language fails, for the latter points to the lived +orld of human (eings, as +ell as their imagined +orlds, although it too can point to the +orld of signs." Like mathematics itself, signs are o(0ective, they transcend the individual and are o+ned and created (y humankind. .ndividuals may partake of either, and can add a little (it to the uses of either that may, if adopted more generally, (e incorporated into the larger pattern o+ned (y humankind. This parallel, . claim is not a coincidence. The +orlds of mathematics and signs are so similar (ecause they are the same in certain respects, in that the o(0ects of mathematics are signs. Ef course the converse does not hold, (ecause there are non1mathematical signs such as street signs, medical symptoms, paintings and novels. The relationship (et+een human (eings and the cultural and social +orlds of signs and, a fortiori, of mathematics is not a symmetric one. 6ocusing on mathematics alone, it can uncontroversially (e stated that mathematics is much larger than any one person or group of persons. 6or although individuals and groups can make even large contri(utions that (ecome accepted into mathematics (y common consent, such contri(utions are al+ays tiny +hen vie+ed against the notional totality of the discipline that has evolved over the past =&&& years. .n learning mathematics, that is appropriating some of its sign and meaning systems, no1one can take on (oard the totality. .n making mathematics, the novel sign uses, patterns and meanings +ill al+ays (e a small addition in comparison to that +hich +as appropriated. The enduring vastness of the discipline gives it its o(0ectivity and external thing1like character. Ene of the dangers of /opper3s ( !*!) tripartite ontology is that it splits human action into three separated realms, and this inevita(ly has an impact on the interpretation of signs. -learly human (eings have a physical and material (asis, and their mental activities, . +ould argue, are insepara(le from this (asis. But they are also social and hence are (oth constituted (y their relationship +ith signs, culture and kno+ledge, as +ell as creating these inha(itants of 'orld <. 6urthermore, the relationship (et+een these three +orlds is not static, (ut a dynamic and interpenetrating functional dance or conversation. That is, they all run together seamlessly in the +orld, including all activities, such as, for example, me the +riter +riting this text and you the reader reading it. ,igns are thus not static o(0ects, (ut functions and tools that are continuously at +ork, used (y
" /honetic language also points to the +orld of human vocal sounds, (ut this is irrelevant here.

human (eings in their communicative and other practices. Thus +hen . claim that mathematical o(0ects are operations on signs and sign functions as +ell as the signs themselves, this is not an optional add on. ,igns only have meaning +ithin the nexus of human sign related activities. .n his analysis of the semiotics of mathematics Jotman ( !!8) distinguishes several categories of mathematical signs. 6irst he distinguishes the alpha(etic from the numeric (or mathematical) sign. Dumeric signs are further analysed into the ideogrammatic and the diagrammatic. .deograms include numerals, B, G, and so on, signs that can (e inserted +ithin the one dimensional flo+ of alpha(etic signs. 9iagrammatic signs are complex, relational, and typically cannot (e inserted in the one dimensional flo+ of alpha(etic signs, as they are multi dimensional. To use such signs in mathematical texts re4uires a (reak in the alpha(etic flo+. 7lthough the origin of such (reaks lies in ancient mathematical and scientific texts, +e no+ take for granted the insertion of ta(les, figures, diagrams, etc., in texts of all sorts. .ndeed, in hypertexts even videos and links to other texts are inserted in this +ay. Jotman does not use the dimensional distinction applied here. $o+ever, Lemke (%&&<) distinguishes typological versus topological semiosis in mathematics, +hich loosely corresponds to the alphanumeric versus diagrammatic distinction in Jotman, and invites the dimensional distinction. Jotman, Lemke and others argue for the necessity and irreduci(ility of the diagrammatic and topological modes of semiosis in mathematics. This provides an argument against Logicism. 6or if mathematics could (e reduced to logic, then the diagrammatic +ould (e dispensa(le in mathematics. 6or despite 6rege3s ( )*!) t+o dimensional -oncept 'riting, mathematical logic is expressi(le alphanumerically, that is in terms of alpha(etic and ideogrammatic sym(ols. 5athematical logic could (e expanded to accommodate the diagrammatic signs of mathematics, (ut this contradicts the claims of Logicism. Damely, that the concepts of mathematics can (e defined in purely logical terms and the theorems of mathematics can (e derived solely from logical axioms. But diagrammatic signs are not a part of modern mathematical logic. .f the diagrammatic and topological modes are inelimina(le in mathematics, then mathematics cannot (e reduced to logic. Logicism has already failed to esta(lish its second claim in that the theorems of mathematics cannot (e derived from purely logical axioms. $o+ever, this argument means that Logicism has also failed to esta(lish its first claim in that the concepts and signs of mathematics cannot (e defined in purely logical terms.* Everall, +hat . am claiming is that the o(0ects of mathematics are to (e found in the domain of mathematical signs. 5athematical o(0ects as are not only named (y signs (ut are also (rought into (eing through sign functioning. Each mathematical theory as it is defined opens up a domain of mathematical discourse, and this domain is populated (y mathematical signs, the o(0ects of mathematics. En a metalinguistic level these theories themselves are mathematical o(0ects too. 'ithin mathematical theories definitions (ring ne+ o(0ects of mathematics into (eing, and the permitted modes of definition are strictly regulated. 6or
* . have to (e careful +hat . assert here, (ecause the diagrammatic and topological modes are representa(le in terms of the alphanumeric and the typological, 0ust as pictures can (e digitiCed. This follo+s from 9escartes3 ground(reaking linking of alge(ra and geometry. $o+ever in making such translations and reductions a human faculty of kno+ing is eliminated. 'e have (oth spatio1visual and logico1linguistic modes of kno+ing (often identified +ith right and left (rain hemisphere activities), and the elimination of one in favour of the other loses some of the (alance, complementarity and po+er of our thought. The fact that all kno+ledge can (e represented in (inary code does not mean that human kno+ing is enhanced (y actually representing all kno+ledge this +ay. ,uch reductions threaten or even destroy meaning and understanda(ility.

example, in /eano arithmetic 2G3 is defined inductively, that is (y means of the induction axiom. Typically, this definition is as follo+s@ 2G3 is a (inary operation defined on the set of natural num(ers, so that for all natural num(ers x, xG& B x and xG,n B ,(xGn). This definition creates the addition operation on D, and although it is not explicitly defined so that it can (e eliminated in terms of its definiens in a single linguistic move, the definition is primitive recursive and hence constructive. That is, (y iterative su(stitutions xGn is e4uivalently transformed into ,,L,x, the successor of x repeated n times. 7t each stage the su(stitution is explicitly defined, and there is an initially given limit to the num(er of su(stitutions re4uired. 'hat even this very simple example illustrates is ho+ mathematical o(0ects are mathematical processes or their end products. .n the +ords of 5achover ( !)<) mathematical o(0ects are reified constructions, that is they are the reification or nominalisation of constructive procedures and processes. That is the operations on mathematical o(0ects themselves (ecome mathematical o(0ects (ut at the next higher level of a(straction or definition. This is reflected in the mathematically primordial transition from ordinal counting and num(ers to cardinal num(ers, (ut it is also reflected at every level in mathematics as ne+ o(0ects are defined in terms of old. Dot every definition involves a reification, unless one counts the collecting together o(0ects to form a finite set. (Evidently forming an infinite set does involve a reification). $o+ever, unlike 5achover . do not limit the domain of meaningful constructions to those accepta(le to constructivists or .ntuitionists. 6rom a semiotic perspective, forming a completed infinite set is no more o(0ectiona(le in principle than forming a finite one, even though the constructive processes involved cannot (e completed (efore the reification into a ne+ o(0ect. Jotman ( !!<) in his ground(reaking semiotic theory and philosophy of mathematics has 4ualms a(out admitting unfinisha(le imaginary actions on mathematical signs. But my vie+ is that in the virtual reality of mathematical signs and o(0ects, provided our sign practices are consistent and conservative of the underlying meanings, no types of actions on signs need (e ruled out ab initio. Evidently part of the strength of mathematical sign systems is that they can incorporate infinities and unfinisha(le procedures and actions, (ecause they are purely semiotic systems. They are not constrained (y the finiteness of matter, energy limitations and entropy +e accept of the physical +orld. .t might (e o(0ected that the semiotic interpretation of mathematical o(0ects as signs that . have sketched is not a form of nominalism, (ut a version of mathematical realism or some other doctrine. 6or . do not re0ect or deny the existence of a(stract mathematical o(0ects, (ut locate them in the cultural and discursive domain of semiotics. But traditional philosophy typically 2looks through3 language to find the domain of o(0ects to +hich it refers, analogous to the correspondance theory of truth or the picture theory of meaning. . am asserting that there is no other reality in +hich to find the o(0ects of mathematics other than the domain of mathematical signs and their uses and functions in the domain of discourse. This certainly preserves many of the properties of mathematical realism and /latonism, 0ust as the social interpretation of o(0ectivity preserves properties of /opper3s 'orld <. ,ince nominalism is the doctrine that the o(0ects of mathematics are 0ust names, or more generally signs, this seems closest to the spirit of a semiotic, social constructivist philosophy of mathematics. $o+ever, if nominalism is understood to mean a position that denies the existence of a(stract mathematical o(0ects such as num(ers and sets in any respect, this is not the position adopted here. /utnam ( !*%) characteriCes nominalistic language as formaliCed languages +hose varia(les range only over individual things, and +hose predicates stand only for ad0ectives and ver(s

applied to individual things. ,uch languages cannot make reference to num(ers, sets or other a(stract o(0ects in any respect. $e criticiCes such language as inade4uate for either mathematics or science. By implication, this is a criti4ue of any strong version of nominalism that restricts itself to nominalistic languages. This is clearly not the version of nominalism . am endorsing here, +hich is supposed to support and recognise existing mathematical practice, +hich clearly involves the a(stract o(0ects of mathematics, rather than restricting it or re4uiring its reform. The nominalism adopted here is supposed to address the issue of +hat these o(0ects are, and +here they are to (e found, rather than to try to define them a+ay. Burgess ( !)<) argues against all forms of nominalism, +hich he characteriCes into three types@ instrumentalist, hermeneutic and revolutionary nominalism. Each of these relates to the role of a(stract mathematical o(0ects in science. .nstrumentalist nominalists adopt an instrumental philosophy of science, hermeneutic nominalists argue that on linguistic analysis the need for a(stract mathematical o(0ects like num(ers is elimina(le, and revolutionary nominalists claim that a ne+ kind of science is possi(le in +hich no existential claims a(out a(stract mathematical o(0ects are made. Burgess criticiCes and re0ects each of these three positions, (eginning +ith instrumentalism. This ends up, he argues, through re0ecting the truth of scientific theories leading to the re0ection of common sense (eliefs as +ell. .n addition, he dou(ts +hether it is possi(le to truly dis(elieve scientific theories that make reference to mathematical o(0ects. But the position . +ish to support here is precisely the instrumentalist one he re0ects, although (y no means expressed in his terms. ,ocial constructivism is instrumentalist a(out mathematics, science, language, and all of human cultural creations. To assert this re4uires me to anticipate the conventionalism . endorse in the next section. Damely, that our intellectual creations are shared and 0ointly created conceptual tools for understanding and operating in the physical, social and cultural +orlds +e inha(it. 5y interpretation of 'ittgenstein supports the semiotic versions of nominalism espoused here. 6or as 'ittgenstein asks@ .s it already mathematical alchemy, that mathematical propositions are regarded as statements a(out mathematical o(0ects 1 and mathematics as the exploration of these o(0ectsM .n a certain sense it is not possi(le to appeal to the meaning of the signs in mathematics, 0ust (ecause it is only mathematics that give them their meaning. ('ittgenstein !*)@ !!) 'ittgenstein suggests that +e need to look +ithin mathematics itself for the meaning of the signs of mathematics, and to pro0ect them out+ard into some realist or /latonist domain of mathematical o(0ects is a delusion (rought a(out (y our linguistic ha(its. 7lthough the version of nominalism . am espousing here has not received much attention in either the philosophy of mathematics or in philosophy in general, it offers a num(er of explanatory (enefits. 6irst of all, it opens the door to an evolutionary epistemology and genetic epistemology approach for understanding mathematics. 5athematical signs, terms, concepts, and theories have gro+n more complex and ela(orate over the course of =&&& years of recorded history, and so too has the range of a(stract entities signified (y mathematics. Understanding mathematical o(0ects nominalistically in terms of signs and sign use explains ho+ the realm of a(stract o(0ects of mathematics can have gro+n +ith the course of history. .f the o(0ects of mathematics are already pre1existent in an o(0ective realm it seems strange that +e perceive them and relate to them only +hen +e are a(le to construct them. ,econdly, persons from school children to adult mathematicians have access to a varying and

developmentally gro+ing range of signs, rules and meanings. The mastery of signs and their rules and meanings is central to all communicative activity, so it is not surprising that it is central to mathematical understanding too. To vie+ mathematical o(0ects as existing in some o(0ective realm (eyond the grasp of our actions 0ust seems less plausi(le than locating it in our communicative practices. These are not very persuasive arguments to someone +ho re0ects nominalism as descri(ed here. But they do represent (enefits to someone open to the possi(ility of this means of accounting for mathematical o(0ects in that they allo+ the resultant philosophy of mathematics to fit together +ith historical and developmental (psychological) accounts of mathematics to give a coherent overall vie+ of the field. Conventionalism -onventionalism has (een descri(ed as :the vie+ that a priori truths, logical axioms, or scientific la+s have no a(solute validity (ut are disguised conventions representing one of a num(er of possi(le alternatives;. (Dorton !!*@ % ) This captures one of the key claims of social constructivism, namely that the concepts, axioms, truths, theorems, theories and standards of mathematics have no a(solute validity (ut represent one set of choices or possi(ilities out of a num(er of possi(le or imagina(le alternatives. This is not to criti4ue or denigrate the excellently fruitful and valua(le choices that +e inherit from history, or ela(orate in contemporary mathematics. 6ar from it, the choices and their outcome, the discipline of mathematics, represent one of the pinnacles of creative flo+ering of the human spirit. 7s free creations of humanity, rather than something pre1existing forced upon us (y inevita(le necessity, the ideas and results of mathematics are all the more remarka(le for their (eauty and elegance, yet po+erful generality and practical utility.) Thus, the claim that . am making is that mathematics rests on human conventions, choices and historical practices in a +ay that . shall ela(orate. There is a conventionalist tradition in the philosophy of mathematics +hich ackno+ledges conventions, fundamentally social agreements of one sort or another, as providing the (asis for logical and mathematical proof and truth. Elements of a conventionalist philosophy of mathematics are to (e found in /oincarK ( !&=), +ho asserts that certain geometrical hypotheses are freely (ut not ar(itrarily adopted conventions. ,imilar elements are also to (e found in 70dukie+icC, +ho termed himself a radical conventionalist, and argued that the linguistic (asis of kno+ledge significantly determines its content (#iedymin !)%). Logical empiricists such as -arnap, $empel, Dagel, and philosophers such as 7yer, Auine, as +ell as others, espouse versions of conventionalism. There are different forms of conventionalism, +hich hinge on different interpretations of the concept of convention. 6uller ( !))) makes a key distinction (et+een t+o senses of the term. 6irst, there is the more artificial sense of convention as an explicit agreement on a definition, assumption or a rule. This might (e termed a rational convention. The standard versions of conventionalism in the philosophy of mathematics referred a(ove adopt this sense and propose in one +ay or another that the conventions on +hich mathematics rests, i.e., the foundations of mathematics, are chosen for pragmatic reasons. This is closely related to instrumentalism, in +hich theories are tools chosen to serve specific purposes.
) The creations of mathematics are free in a strictly regulated and restricted sense that except in exceptional cases conserves existing meanings, rules and structures.

7ccording to +hat might (e termed rational conventionalism, the (ody of mathematical kno+ledge, the superstructure, follo+s (y logical means, i.e., proofs, from its conventional (asis. .n the extreme form +hich 9ummett ( !=!) termed 2full1(looded3 conventionalism and incorrectly attri(utes to 'ittgenstein, there is no (ase1superstructure division and all mathematical kno+ledge is directly adopted (y convention. This latter position is untena(le, given the universally ackno+ledged importance of inference in esta(lishing mathematical kno+ledge, and no philosophers su(scri(e to it.! 6uller3s ( !))@ =") second sense of convention is :a practice that has emerged largely +ithout design yet continues to (e maintained;. -onvention in this sense is close to +hat is implied (y participation in 'ittgensteinian language games and forms of life, and in 6oucault3s 2discursive practices3. 6or to participate in a form of life and its language games is to follo+ the roles and norms and engage in the expected practices, i.e., to o(serve the conventions of the form of life. This sense of convention is at least partly implicit, since usually no explicit statement of the conventions +ill (e made, rather participants must infer the conventions from o(served (ehaviours and from others3 corrections of their o+n infractions. This might (e termed historical convention, since it is (ased on pre1existing practices. The previous discussion of the role of grammar in language use fits +ith this conception. .n simple terms historical conventions follo+ practice rather than preceding it, as they do in rational conventions. Let me no+ clarify +hat conventionalism means in terms of epistemology. ,ocial constructivism claims that kno+ledge and truth are socially constructed. 'hat does this meanM 9oes it mean that any collection of signs, representations or information making claims a(out states of affairs 2constructed socially3 is e4ually trueM -ertainly not, for this +ould (e patently a(surd and +ould lead to the production of contradictions, as +hen the ,tate of .ndiana legislated %%N* for the value of /i. 9oes this claim mean that any historic group reaching agreement a(out a state of affairs has the right to call its claims true or esta(lished kno+ledgeM $ere . must agree +ith #oldman ( !!=@ ) that :truth must not (e e4uated +ith consensual (elief;. 6or if this criterion +as admitted, +e +ould have to ackno+ledge false and pernicious claims as the truth. 6or example, the Fu Flux Flan agrees that (lack persons are inferior to +hite persons in a num(er of specifia(le +ays. ,imilarly the DaCi3s propounded the doctrine of the inferiority of the Oe+s. The key point here is that such groups had a +idespread internal agreement, (ut that this agreement did not make the claims true. Their dogmas are unfounded claims that social constructivists along +ith other li(eral or humane thinkers re0ect as pernicious lies. But this raises the 4uestion of ho+ else truth is socially constructed. Truth is not e4uated +ith consensual (elief. .t is not simply created (y fiat, agreement, or convention. .nstead, truth claims are su(0ected to criteria for acceptance. Truth claims need to (e +arranted (y something other than simply group agreement for acceptance. .n mathematics, a ne+ kno+ledge claim, a +ould1(e theorem or ne+ result, must (e put for+ard +ith a proposed +arrant for the claim, that is, +ith a proof. The proof is an argument that is persuasive in esta(lishing the truth, or rather the proven nature of the claim, to experts in the field. & The criteria for the acceptance of ne+ theorems do not exist independently of humankind, ho+ever, for it re4uires a group of experts to deploy them. .n addition, the
! Ence again . am putting aside 'ittgenstein3s ( !=<) notion that all rule follo+ing is optional, discussed in Dote . & There are technical issues surrounding the relationship (et+een truth and proof in mathematics foregrounded (y #Idel3s ( !< ) seminal .ncompleteness theorems that for simplicity . shall overlook here (ut that do not invalidate my argument.

criteria are not fully explicit, and . +ould argue, cannot (e rendered fully explicit (Ernest !!!). 6or they depend in part on the experience and case kno+ledge of accepted results in mathematics of these experts. Experts serving as 0ournal and conference referees, for it is in such roles that mathematicians act as gatekeepers for the admission of ne+ mathematical kno+ledge, are not al+ays unanimous, nor even al+ays correct in their 0udgements, in the eyes of history. But it is through the application of criteria and standards of logic, mathematical practice and rhetorical form that ne+ results are +arranted in mathematics (Ernest !!)). .t is these second order criteria and standards that represent a central part of the conventional (asis of mathematical kno+ledge. -ommunities of mathematicians agree on these criteria and standards, not ar(itrarily, (y rational conventions, (ut (y historical conventions em(edded in historical practice. Thus an accepted mathematical result has a persuasive +arrant, a proof, that satisfies the appropriate group of experienced mathematicians that it meets current proof standards. $o+ever, the second order criteria and standards for the accepta(ility of mathematical kno+ledge are themselves historical conventions, developed organically and historically in communities of mathematicians and transmitted from one generation to the next partially through +ritten criteria and partially through shared meanings and mathematical practices. The vertical variations in such criteria (i.e., over time) are clear, and . claim, cannot (e explained (y the pushing (ack of the frontiers of ignorance. 'hat is less commonly ackno+ledged is the extent of the horiContal variations in such criteria (i.e., over different mathematical specialisms). .n one of the fe+ relevant studies Fnuth ( !)=) found +ide divergences in the style and rhetorical form of papers in different mathematical topics implying +idely varying criteria of acceptance in play. The criteria for acceptance of mathematical results thus vary greatly over time and specialism, and represent accepted historical conventions in place in the different communities of mathematicians involved. .n my vie+ another important factor is also at +ork. 6or . claim that there is also a mathematical metaphysics or ideology that sits in place alongside these varying conventions and practices. This asserts the o(0ectivity, universality and certainty of mathematics and its standards and (asis. .n conse4uence, mathematicians disregard the variations in conventions, standards and practices of accepta(ility in mathematics. .nstead they see these differences as surface effects that mask an underlying constancy and permanent core of meaning, o(0ectivity and truth. 6oucault3s ( !)&, !)8) concept of a 2regime of truth3 concerns the historical truth1status of a socially accepted model, perspective or +orld1vie+. 'hen for historical reasons important sectors of a community or society have come to accept such a perspective, and act as if it is true, then a 2regime of truth3 prevails. 7 regime of truth does not concern individually +arranted propositions, (ut an overall metaphysical +orld1vie+. ,uch a perspective may seem as +ell grounded as a foundationalist account of kno+ledge, (ut the (asis of such truths is the social acceptance and lived nature of the underlying presuppositions. 7 regime of truth is hegemonic, and it is held in place (y a discursive practice, a set of language games em(edded in a form of life, parallel to 'ittgenstein3s notions. The metaphysics of mathematics that sees the results of mathematics and the criteria for their acceptance as certain, unchanging and timeless is such a regime of truth. .t is a +idely shared +orld1vie+ that overlooks empirical evidence that contradicts it, or explains it a+ay as insignificant. 7n apparent +eakness in this account is that it sounds as if persons have (een fooled or even coerced into accepting error or a false account. But a regime of truth ena(les
.t is also typically maintained (y po+erful social structures and institutions.

people to see +hat is as +hat has to be? to identify historically grounded (ut contingent truths, not errors, as necessary truths. 7ccording to conventionalism, all kno+ledge presuppositions are the results of different forms of accepted practices, agreement or decision and can all (e 4uestioned and reconsidered. Even the fundamental La+ of (Don)-ontradiction is put aside in a limited +ay in some modern logics (Joutley et al. !)%). 5ostly the un4uestioned (asic assumptions in mathematics are an important given, the (asis for mathematical research, and certainly this for mathematical applications and education. But sometimes these un4uestioned assumptions serve not as (uttresses for the certainty and usefulness of mathematics (ut as epistemological o(stacles to further progress. The example of $amilton3s difficulties in 4uestioning the universal applica(ility of commutativity in alge(ra, and ho+ this +as an o(stacle to further progress, is discussed (elo+. 7 crucial pro(lem for a conventionalist philosophy of mathematics is to account for mathematical and logical necessity and the certainty of mathematical truths and theorems. 'ittgenstein offers an approach to this pro(lem in his later philosophy. 'ittgenstein3s vie+ of logical and mathematical necessity comes out of his theory of language games. $is vie+ is that necessity, such as that of dra+ing an inference follo+ing the la+s of deductive logic, +hich underpins so much of mathematical kno+ledge, arises from the human agreement in follo+ing a rule that is stipulated, presupposed and em(edded in a language game. The +ord 2agreement3 and the +ord 2rule3 are related to one another, they are cousins. .f . teach anyone the use of the one +ord, he learns the use of the other +ord +ith it. ('ittgenstein !=<@ )") Thus there is no extra1human or o(0ective force that compels anyone to follo+ a logical rule or to accept the conclusion of a logical deduction. .t is rather that participating in certain language games entails accepting certain rules. .f one re0ects the rule one is repudiating the game as it is understood and played (y others. :To o(ey a rule, to make a report, to give an order, to play a game of chess, are customs (uses, institutions).; ('ittgenstein !=<@ ) ) Ef course the traditional vie+ that logical necessity underpins deduction and rational thought is very firmly entrenched. 'ittgenstein anticipates the o(vious philosophical o(0ection that rule follo+ing in logic and else+here stems not from human agreement and (ut from some essential form of logical necessity, +hatever that might (e. But even to communicate disagreements a(out truth, falsity or necessity presupposes that +e agree to use the terms compara(ly in social discourse and life. :,o you are saying that human agreement decides +hat is true and +hat is falseM; 1 .t is +hat human (eings say that is true and false? they agree in the language they use. That is not agreement in opinions (ut in forms of life. ('ittgenstein !=<@ ))). 7greement in 'ittgenstein3s sense arises from our participating in shared language games (+oven into our forms of life), and does not consist of ar(itrarily adopting conventions. This gives us the shared constraints on the meanings of our language, and ultimately leads us to decide +hat counts as truth and falsehood. Thus the relation among agreement, convention and truth is far more su(tle and complex in 'ittgenstein3s philosophy than in rational conventionalism. 7n example, not from 'ittgenstein, is as follo+s. The traditional vie+ of philosophy and logic is that the follo+ing logical inference is necessary +ithout 4ualification@ given A and

A B, then B. 5y understanding of 'ittgenstein3s position on this is as follo+s. 7greeing that this inference is necessary depends on many prior implicit agreements. 6irst of all, the parties to the agreement must all share an understanding of a sophisticated language, +ritten English in this case. This in turn presupposes that the parties are part of a linguistic community and routinely communicate, interact +ith others and take part in shared social activities. ,econdly, the parties (to the agreement) must agree that 2 A3 and 2B3 are metalinguistic sym(ols denoting fixed (ut ar(itrary English propositions, and that every instance of one of them has the same denotation (+ithin an assumed (ut delimited meaning context). Third, the parties accept the rule of inference Modus Ponens as valid (i.e., +henever its premises have the truth as their value, they agree that invaria(ly the conclusion does too.) 7lthough not an exhaustive analysis, these assumptions sho+ that the logical necessity of the inference depends on shared forms of life (assumption ) and participation in language games (assumptions , % and <). Ence these assumptions are made (and most of them you the reader and . the +riter as participating mem(ers of modern literate and academic society have usually no option (ut to make %) then the conclusion is necessary. Like+ise, given another, simpler set of assumptions a(out the game of -hess, and a particular (oard configuration, check1mate in t+o moves is similarly necessary. 5athematics is the su(0ect par excellence, in +hich necessity a(ounds. Ence certain assumptions, definitions and rules are accepted the greater part of mathematics does follo+ (y logical inference, i.e., (y necessity. But that necessity rests on a set of assumptions that . claim are not themselves necessary in totality. ,ome crucial elements of mathematical kno+ledge are contingent truths, handed do+n from past practice and convention, and conse4uently the (ody of mathematical kno+ledge as a +hole is contingent truth. The social constructivist position is that there is a great deal of sta(ility in a discipline like mathematics +hile at the same time there are virtually no essential or necessary features handed on do+n through the millennia. 5athematicians often contrast necessity +ith ar(itrariness, and implicitly argue that if mathematics has no a(solute necessity and essential characteristics to it, then it must (e ar(itrary, and conse4uently anarchy prevails and anything goes. $o+ever as Jorty ( !! ) has made clear in philosophy, contingency, not ar(itrariness, is the opposite of necessity. ,ince to (e ar(itrary is to (e determined (y or arising from +him or caprice rather than 0udgement or reason, the opposite of this notion is that of (eing selected or chosen. . +ish to argue that mathematical kno+ledge is (ased on (oth contingency, due to socio1historical accident, and deli(erate choice (y mathematicians, +hich is ela(orated through extensive reasoning and practices into mathematical tradition. Both contingency and selection are active throughout the long history of mathematics. . also +ish to argue that the adoption of certain rules of reasoning and consistency in mathematics mean that much of mathematics follo+s +ithout further choice or accident, (y logical necessity, provided +e maintain the rules and conventions. . freely admit that much of mathematics follo+s (y logical necessity from its assumptions and adopted rules of reasoning. $o+ever this does not contradict the conventionalist and anti1 a(solutist position of social constructivism, for . deny that the rules, reasoning and logical necessity in mathematics are a(solute or context1independent. 5athematics consists of language games +ith very entrenched rules and patterns that are very sta(le and enduring, (ut +hich al+ays remain open to the possi(ility of change, and in the long term, do change as a totality, if not in every part.
% But note that for philosophical discussion purposes +e might temporarily suspend (elief in these or any assumptions, i.e., choose to play a different language game.

7 +ell kno+n example is commutativity in the multiplication of num(ers, so that 0kBk0. The !th century mathematician #eorge /eacock framed a la+, The Principle of the Equivalence of Permanent Forms, that stated that developments in alge(ra must al+ays respect the underlying arithmetical la+s. Thus non1commutativity +as virtually unthinka(le. Dot (ecause of /eacock3s dictum, (ut (ecause of the underlying sta(le practice that he reflected in his explicit statement. But it +as a convention, not a logical necessity. 7fter %& years of struggle to extend imaginary num(ers 'illiam $amilton made (reakthrough (y re0ecting 0kBk0, and putting 0k B 1k0 instead. This led to the important Theory of Auaternions. .n so doing, $amilton respected and conserved many of the la+s of alge(ra, (ut also made extensions and significant changes (/ickering !!=). $e overcame +hat Bachelard ( != ) called an epistemological o(stacle, a received, conventional part of the conceptual apparatus of mathematics that o(structed further progress? a contingency that appeared a necessity. ,uch dramatic changes in +hich past strictures are overturned to develop a ne+ and fruitful theories may not happen every day in the history of mathematics. But they do happen regularly in children3s development in school mathematics. .t is commonplace for teachers to extend mathematical topics re4uiring the negation of existing rules and the change of underlying meanings, through the adoption of ne+ rules. 6or example, for a young child mastering elementary calculation, the task <18 is impossi(le. But later it has a determinate ans+er@ <18 B 1 . ,imilarly < divided (y 8 (<N8) is at first an impossi(le task. Later it is not only a possi(le task, (ut P names the ans+er to it, i.e., (ecomes a ne+ kind of semiotic o(0ect, a fractional numeral. .n early multiplication tasks children learn implicitly or explicitly that :multiplying al+ays makes (igger;. Later +hen the domain of num(ers they operate on is expanded to include fractional and decimal num(ers (i.e., Jationals), or even 0ust Cero, this rule is contradicted. The difficulties caused to learners (y these changes in definitions and conventions are +ell kno+n. .n these and many compara(le cases the rule changes are necessitated (y changes in the underlying meaning of the operations. Thus su(traction, initially, is usually understood in enactive or metaphoric terms as resulting from the partitioning of a collection of concrete o(0ects and the removal of one part. $ence <18 is impossi(le, it is not 0ust a matter of learner ignorance. ,u(se4uently in learner development su(traction is commonly understood more structurally as the inverse of addition applied to an enlarged and more a(stract domain of num(ers. $ence since <1< B &, <18 B 1 . But this is a ne+ mathematical system, +hich represents an intellectual advance that took humankind hundreds of years to make and to accept. .t is very likely that the later more a(stract meaning of su(traction cannot (e developed +ithout the earlier concrete meaning, so the apparent contradiction is unavoida(le. .n examples such as these, the student has to 2unlearn3, that is relin4uish something already learned, in order to make further progress. 6ollo+ing Bachelard, these pro(lems have also (een termed epistemological o(stacles (Brousseau !!*, ,ierpinska !)*). .t is through immersion and participation in the practices of first learning mathematics, and later doing research mathematics, that mathematicians are enculturated into mathematical forms of life +ith their tacit rules, conventions and kno+ledge. These contingent features, extended and ela(orated through rigorous reasoning and proof, give mathematicians a sense of the necessity of their su(0ect. They also enculturate them into the standards and criteria for the acceptance of ne+ mathematical kno+ledge. But if only one convention or contingency is +oven into this kno+ledge, and . claim there are many more than one, then mathematics as a +hole is conventional.

Conclusion The version of social constructivism that . have (een discussing is a naturalistic philosophy of mathematics that aims to provide an account of mathematics as it is practised, cogniCant of (oth the social structures +ithin the mathematical community and the historical development of the discipline. . have argued that in a particular sense of the term it is nominalist, (ecause it regards mathematical o(0ects as signs deployed +ithin semiotic systems +ith sign rules and meanings. . have not denied that a(stract o(0ects exist, 0ust located them in the realm of culture, alongside money, literature, and other human institutions and artifacts. . have argued that mathematical kno+ledge is conventional in the sense that it is +arranted (y the rules of mathematics and the mathematicians3 understandings of logical necessity. $o+ever, . have claimed that these rules, and mathematicians decisions of accepta(ility (ased on them, is itself partly a result of historical contingency. By su(scri(ing to these limitations and deviations from the traditional ideology of the purity, o(0ectivity, and perfection of mathematics . aim to reclaim mathematics from the idealists. Bringing mathematics (ack do+n to earth, to the mundane reality of lived human life, is not to denigrate or (esmirch it. .ronically, the aim is to offer an more accurate and a truer picture of mathematics as a part of lived human experience. References 7CCouni, O. ( !!8) Metaphysical Myths, Mathematical Practice, -am(ridge@ -am(ridge University /ress. Bachelard, #. ( != ) L activit! rationaliste de la physique contemporaine , /aris@ /resses Universitaires de 6rance. Berger, /. L. and Luckmann, T. ( !"") The "ocial #onstruction of $eality% & Treatise in the "ociology of 'no(ledge, London@ /enguin Books. Bloor, 9. ( !)8) 7 ,ociological Theory of E(0ectivity, in ,. -. Bro+n, Ed., )b*ectivity and #ultural +ivergence (Joyal .nstitute of /hilosophy lecture series, *), -am(ridge@ -am(ridge University /ress, !)8@ %%!1%8=. Brousseau, #. ( !!*) Theory of +idactical "ituations in Mathematics, 9ordrecht@ Flu+er. Burgess, O. /. ( !)<) 2'hy . am Dot a Dominalist3, ,otre +ame -ournal of Formal Logic %8 ( !)<), !<1 &=. -ollis, F. 6. ( !*=) & "tudy of #oncrete and Formal )perations .n "chool Mathematics% & Piagetian /ie(point, ,ydney@ 7ustralian -ouncil 6or Educational Jesearch. 9errida, O. ( !*)) 0riting and +ifference (Translated (y 7. Bass), London@ Joutledge and Fegan /aul. 9ummett, 5. ( !=!) 'ittgenstein3s /hilosophy of 5athematics, The Philosophical $evie(, >ol. "), <%81<8). Jeprinted in /. Benacerraf, and $. /utnam, Eds., Philosophy of Mathematics% "elected readings, Engle+ood -liffs, De+ Oersey@ /rentice1$all, !"8@ 8! 1=&!. Eco, U. ( !)8) "emiotics and the Philosophy of Language, Bloomington, .ndiana@ .ndiana University /ress. Ernest, /. ( !!)) "ocial #onstructivism as a Philosophy of Mathematics, 7l(any, De+ Qork@ ,UDQ /ress. Ernest, /. ( !!!) 26orms of Fno+ledge in 5athematics and 5athematics Education@ /hilosophical and Jhetorical /erspectives3, Educational "tudies in Mathematics, >ol. <), Dos. 1<@ "*1)<.

6oucault, 5. ( !)&) Po(er1'no(ledge (edited (y -. #ordon), De+ Qork@ /antheon Books. 6oucault, 5. ( !)8) The Foucault $eader (edited (y /. Ja(ino+), London@ /enguin Books 6rege, #. ( )*!) 2egriffsschrift, in $ei0enoort ( !"*)@ 1)%. 6uller, ,. ( !))) "ocial Epistemology, Bloomington, .ndiana@ .ndiana University /ress. #ergen, F. ( !!!) &n .nvitation to "ocial #onstruction, London@ ,age. #iedymin, O. ( !)%) "cience and #onvention, Exford@ /ergamon. #Idel, F. ( !< ) R(er formal unentscheid(are ,StCe der /rincipia 5athematica und ver+andter ,ysteme .. Monatshefte fur Mathemati3 und Physi3, >ol. <), *<1 !). Translation in $ei0enoort ( !"*)@ =!%1" *. #oldman, 7. .. ( !!=) Education and ,ocial Epistemology, in Deiman, 7. Ed. 4556 Philosophy of Education "ociety 7earboo3, -hampaign, .llinois@ /hilosophy of Education ,ociety, University of .llinois at Ur(ana1-hampaign. $acking, .. (%&& ) The "ocial #onstruction of 0hat8 -am(ridge University /ress. $arding, ,. ( !)") The "cience 9uestion in Feminism, 5ilton Feynes@ Epen University /ress. $arrK, J. and #illett, #. ( !!8). The +iscursive Mind. London@ ,age /u(lications. $ei0enoort, O. van, Ed. ( !"*) From Frege to :;del% & "ource 2oo3 in Mathematical Logic , -am(ridge, 5assachusetts@ $arvard University /ress. $ersh, J. ( !!*) 0hat .s Mathematics, $eally8 London, Oonathon -ape. Fnuth, 9. E. ( !)=) 7lgorithmic Thinking and 5athematical Thinking, &merican Mathematical Monthly !% ( !)=) *&1 ) . Lemke, O. (%&&<) 25athematics in the 5iddle@ 5easure, /icture, #esture, ,ign, and 'ord3, in 7nderson, 5, ,aenC1Ludlo+, 7. Tell+eger, ,. and -ifarelli, >. >., Eds. (%&&<) Educational Perspectives on Mathematics as "emiosis% From Thin3ing to .nterpreting to 'no(ing, De+ Qork, Etta+a and Toronto@ Legas /u(lishing, %&&<@ % =1%<8. 5achover, 5. ( !)<) To+ards a De+ /hilosophy of 5athematics, 2ritish -ournal for the Philosophy of "cience, >ol. <8, 1 . Dorton, 7. L. Ed. ( !!*) The <utchinson +ictionary of .deas, Exford@ $elicon pu(lishing. /eirce, -. ,. ( !< 1=)) #ollected Papers () volumes), -am(ridge, 5assachusetts@ $arvard University /ress. -91JE5 version pu(lished in /ast 5asters series, -harlottesville >irginia@ .nterLex -orporation. /iaget, O. ( !=%) The #hild s #onception of ,umber (Translation, #attegno U $odgson), London@ Joutledge and Fegan /aul. /ickering, 7. ( !!=) -oncepts and the 5angle of /ractice@ -onstructing Auaternions, "outh &tlantic 9uarterly, >ol. !8, Do. %, 8 *18"=. /oincarK, $. ( !&=) "cience and <ypothesis. Translation (y 2'. O. #.3, London@ 'alter ,cott /u(lishing -ompany, Ltd. /opper, F. J. ( !*!) )b*ective 'no(ledge. Jevised Edition, Exford@ Exford University /ress. /utnam, $. ( !*%) Philosophy of Logic, London@ #eorge 7llen U Un+in. Auine, '. >. E. ( !=<) From a Logical Point of /ie(, De+ Qork@ $arper Torch(ooks. Auine, '. >. E. ( !"!) )ntological $elativity and )ther Essays, De+ Qork@ -olum(ia University /ress. Jorty, J. ( !! ) #ontingency, .rony and "olidarity% Philosophical Papers /olume = , -am(ridge@ -am(ridge University /ress. Jotman, B. ( !!<) &d .nfinitum The :host in Turing s Machine% Ta3ing :od )ut of Mathematics and Putting the 2ody 2ac3 in, ,tanford -alifornia@ ,tanford University /ress. Jotman, B. ( !!8) 5athematical 'riting, Thinking and >irtual Jeality, in /. Ernest, Ed. ( !!8) Mathematics, Education and Philosophy% &n .nternational Perspective, London, The 6almer /ress, *"1)".

Joutley, J. +ith 5eyer, J. F., /lum+ood, >. and Brady, J. ( !)%) $elevant Logics and their $ivals 4, 7tscadero, -alifornia@ Jidgevie+. ,ierpinska, 7. ( !)*) $umanities students and epistemological o(stacles related to limits. Educational "tudies in Mathematics, 4>, <* V<!*. >ygotsky, L. ,. ( !*)) Mind in "ociety. The development of the higher psychological processes. -am(ridge, 5assachusetts@ $arvard University /ress (Edited (y 5. -ole et al.) 'ein(erg, 9. and #avelek, O. ( !)*) 7 ,ocial -onstructivist Theory of .nstruction and the 9evelopment of 5athematical -ognition, Proceedings of .nternational :roup for Psychology of Mathematics Education #onference 44, 5ontreal, >ol. <, <8"1<=%. 'ertsch, O. >. ( !! ) /oices of the Mind, London@ $arvester 'heatsheaf. 'ertsch, O. >. ( !!*) Mind &s &ction, Exford@ Exford University /ress. 'ittgenstein, L. ( !=<) Philosophical .nvestigations (Translated (y #. E. 5. 7nscom(e), Exford@ Basil Black+ell. 'ittgenstein, L. ( !*)) $emar3s on the Foundations of Mathematics (Jevised Edition), -am(ridge, 5assachusetts@ 5assachusetts .nstitute of Technology /ress. This paper first appeared in Philosophica >ol.*8 (%&&8) pp. <1%!.