<?xml-stylesheet type="text/xsl" href="to.xsl"?>
<TEI xmlns="http://www.tei-c.org/ns/1.0">
   <teiHeader>
      <fileDesc>
         <titleStmt>
            <title>'ALIA EST RERUM ALIA TERMINORUM DIVISIO': ABOUT AN UNPUBLISHED MANUSCRIPT OF LEIBNIZ</title>
            <author><name>Massimo</name>
               <surname>Mugnai</surname>
            </author>
         </titleStmt>
         <publicationStmt>
            <authority>ILIESI-CNR</authority>
            <availability>
               <p>Biblioteca digitale Progetto Agorà</p>
            </availability>
         </publicationStmt>
         <sourceDesc>
            <bibl>
               <title level="m">'ALIA EST RERUM ALIA TERMINORUM DIVISIO': ABOUT AN UNPUBLISHED MANUSCRIPT OF LEIBNIZ</title>
               <author>Massimo Mugnai</author>
               <title level="a"/>
               <publisher>Leo S. Olschki Editore</publisher>
               <editor/>
               <pubPlace>Roma</pubPlace>
               <idno type="isbn"/>
               <biblScope> pp. 257-269 (Collana Lessico Intellettuale Europeo, LXXXIV)</biblScope>
               <date/>
            </bibl>
         </sourceDesc>
      </fileDesc>
     </teiHeader>
   <text>
      <front>
         <titlePage>
            <docAuthor>Massimo Mugnai</docAuthor>
            <docTitle>
               <titlePart>'ALIA EST RERUM ALIA TERMINORUM DIVISIO': ABOUT AN UNPUBLISHED MANUSCRIPT OF LEIBNIZ</titlePart>
            </docTitle>
         </titlePage>
      </front>
      <body>
         <pb n="257" facs="UNITA/UNITA_257.jpg"/>
            <p>From the beginning of his long philosophical career to the end of his<lb/>life,
               Leibniz seems to have never dismissed the intent - which, after various<lb/>pauses,
               he renewed with the regularity of a recurring dream - to compile<lb/>long lists of
               definitions. These lists had, in principle, two different aims: to<lb/>establish a
               firm ground for the building of a new encyclopedia and to re-<lb/>form the
               traditional doctrine of categories. Often in the same “table of defi-<lb/>nitions”
               these two aims intertwine. Sometimes the lists are quite short and<lb/>heterogeneous
               in content; sometimes they are long and homogeneous, but<lb/>with many repetitions;
               and, in general, they all exhibit an unaccomplished,<lb/>tentative quality which,
               however, do not diminish their philosophical inter-<lb/>est. These lists have been
               published in the <hi rend="italic">Vorausedition</hi> or elsewhere; only<lb/>a
               handful of them is as yet unpublished.</p>
            <p>In this paper I intend to draw attention to an unpublished list of defini-<lb/>tions
               which Leibniz wrote after 1700 and which almost exclusively contains<lb/>logical
               items and reflections on logical ontology. I will first give the original<lb/>Latin
               text (without the critical apparatus); then I will propose an English<lb/>translation
               of the most relevant passages; and finally I will present a short<lb/>commentary to
               the translated passages.</p>
            <p>
               <hi rend="italic">The Latin text</hi> (LH 7 C Bl. 76 v. A translation will be given
               of the<lb/>passages comprised between brackets “&lt;” and “&gt;”).</p>
            <p>Alia est rerum alia terminorum divisio.</p>
         <p><hi rend="italic"> Terminus </hi>est cogitabile quod potest esse subiectum propositionis verae<lb/>estque
            vel implicans contradictionem, nempe <hi rend="italic">chimaera</hi>, vel possibile seu<lb/>non implicans,
               nempe reale.</p>
         <p> <hi rend="italic">Reale</hi> est terminus possibilis seu distincte cogitabilis, ut homo, doctus,<lb/>causa,
               actio.</p>
            <p>&lt;Rursus terminus sumitur vel materialiter, et ita est ens, cuius
               diversi<lb/>termini sunt eadem res ut rex et propheta in Davide, nempe idem est
               rex</p>
         <pb n="258" facs="UNITA/UNITA_258.jpg"/>
            <p>qui est propheta. Vel terminus sumitur formaliter, ut rex qua rex, nempe<lb/>etsi non
               sit alius rex quam propheta, tamen aliud est regem esse aliud<lb/>prophetam esse.
               David qua rex differt a Davide qua propheta. Diversi ergo<lb/>sunt termini sed ens
               idem. Interim in terminis quoque formaliter spectatis<lb/>distinctio est. Interdum
               enim in idem resolvuntur veluti figura triangula<lb/>et figura trilatera, item
               rectangulum aequilaterum et quadrilaterum ae-<lb/>quiforme, ubi nodo est eadem;
               interdum vero diversa notio est, ut aureum<lb/>et rotundum, etsi monetae conveniant
               quam ducatum vocamus. Termini<lb/>ergo differunt re, cum diversa entia spectant,
               forma cum ad diversas no-<lb/>tiones spectant, consideratione, cum diversa dicunt seu
               cogitanda exhibent.<lb/>Idem est quidam rex et quidam propheta, omne triangulum et
               omne tri-<lb/>laterum. Stemus ergo in rebus ipsasque notiones secundas ut res
               considere-<lb/>mus&gt;.</p>
         <p> <hi rend="italic">Terminus</hi> est vel <hi rend="italic">Nihil</hi> ut <hi rend="italic">Blitiri</hi>, vel aliquid.</p>
         <p><hi rend="italic"> Aliquid</hi> est subiectum propositionis verae, estque vel impossibile, quod<lb/>dicitur
               Chimaera, vel ens.</p>
         <p><hi rend="italic"> Ens </hi>est aliquid possibile, ut Deus, homo, scientia, scriptio, motus.</p>
         <p>Ens vel est <hi rend="italic">subsistens</hi> quod tantum subiectum esse potest, vel
            <hi rend="italic">attribu-<lb/>tum</hi>, quod est constituens praedicati alterius entis ut scientia constituit
            ut<lb/>aliquis sit sciens. Actio ut agens. Sed nonne datur tertium? &lt;Sic tempus,
            lo-<lb/>cus, neque est subsistens nec attributum. Idem est de numero, ordine.
            Sic<lb/>decem non est attributum ullius rei. Neque enim de aggregato neque
            de<lb/>singulis dici potest numerus denarius. Idem est de relatione quae commu-<lb/>nis,
            v. g. similitudo duorum. Datur itaque attributum quod est simul in<lb/>pluribus
            subiectis. Talia ergo sunt ordo, adeoque tempus et locus&gt;.</p>
            <p>A possible translation of the first passage between brackets:</p>
            <p>Again a term is taken either materially, and thus it is a being the differ-<lb/>ent
               terms of which are the same thing, as king and prophet in David - i. e.<lb/>the same
               person who is a king is a prophet as well; or a term is taken for-<lb/>mally as a
               king insofar as he is a king - i. e. even though the king is not a<lb/>different
               person from the prophet, yet to be a king is different from being a<lb/>prophet.
               David, insofar as he is a king, differs from David insofar as he is a<lb/>prophet.
               Therefore the terms are different, but the being is the same. In<lb/>some
               circumstances, however, even the terms considered formally admit of<lb/>a
               distinction. In fact, by means of analysis, they give rise, sometimes, to
               the<lb/>same - as in the case of a triangular and a trilateral shape or in the case
               of<lb/>an equilateral rectangle and an equiform quadrilateral, where the notion
               is<lb/>the same; sometimes the notion is different, as in the case of “golden”
               and<lb/>“round”, even if both apply to the coin which we call “ducat”.
               Therefore<lb/>terms differ in the following way: about the thing - insofar as they
               concern<lb/>different beings; about the form - insofar as they concern different
               notions;<lb/>about the way of considering - insofar as they mention or show
               different<lb/>thoughts. A certain king and a certain prophet are the same, and so are
               ev-</p>
         <pb n="259" facs="UNITA/UNITA_259.jpg"/>
            <p>ery triangle and every trilateral. Thus, we have to stay with things and
               to<lb/>consider as things all second notions as well.</p>
            <p>Commentary</p>
            <p>As is well known, Leibniz’s use of “term” is quite puzzling. In our<lb/>text (in the
               part I have not translated), he defines a <hi rend="italic">term</hi> as
               «something<lb/>which can be thought and which can be the subject in a true
               proposition».<lb/>This seems to be quite in agreement with the definition of a term
               as the<lb/>meaning of a word, which we find in a letter to Des Bosses of 1712.<note
                  place="foot" xml:id="ftn1" n="1"> Cf. GP II, p. 470. </note> But in<lb/>the text
               under consideration, Leibniz plainly writes that <hi rend="italic">Blitiri</hi>,
               which was<lb/>traditionally mentioned as a typical linguistic expression with no meaning,<lb/>
               <hi rend="italic">is</hi> a term. Without excusing Leibniz’s inaccuracies in the use
               of the term<lb/>“term”, I think that we can easily explain these inaccuracies if we
               take into<lb/>account his peculiar attitude towards nominalism. In the period that
               dates<lb/>from the composition of the <hi rend="italic">Dialogue</hi>
               <note place="foot" xml:id="ftn2" n="2"> Cf. GP VII, pp. 190-191. </note> (1677) to
               that of <hi rend="italic">New Essays</hi>
               <note place="foot" xml:id="ftn3" n="3"> Cf. A VI, vi, p. 77. </note>
               <lb/>(1704-5), Leibniz firmly claims that men cannot think without signs
               -<lb/>whether they be written or spoken words, letters or numbers, or
               simply<lb/>mental images or perceptions of some sort. At the same time, Leibniz
               rec-<lb/>ognizes that words and other signs are meaningful only if they express
               some<lb/>idea or ideas. The point is that men cannot have - with the exception of
               a<lb/>handful of cases - direct access to ideas (in particular if the ideas are
               suffi-<lb/>ciently complex): the psychological act of grasping an idea - an act
               which<lb/>gives rise to what Leibniz calls a “concept” - is always associated in
               our<lb/>mind with the use of words or other signs.<note place="foot" xml:id="ftn4"
                  n="4">
                  <hi rend="italic">Ibid</hi>. </note> Thus language is something<lb/>which
               interposes itself between the realm of ideas and us; and our thoughts<lb/>are, for
               the most part, “blind thoughts” <hi rend="italic"> (cogitationes caecae) </hi> made
               of words<lb/>(or of other signs).<note place="foot" xml:id="ftn5" n="5"> Cf. GP IV,
                  p. 423. </note> In this sense, words <hi rend="italic">are</hi> concepts or,
               better said, substi-<lb/>tutes for concepts. It is not inappropriate to think that
               this point of view in-<lb/>duced Leibniz to use the expression “term” in a quite
               inaccurate way,<lb/>which blurs the differences between conceptual content and its
               linguistic<lb/>expression. At any rate it seems not too far from the spirit - if not
               from the<lb/>literal meaning - of Leibniz’s writings to consider a term as a
               meaningful<lb/>linguistic expression. Therefore we can re-define a term in a
               Leibnizian<lb/>sense as a linguistic expression associated with a meaning, which can
               be the</p>
         <pb n="260" facs="UNITA/UNITA_260.jpg"/>
            <p>subject (or predicate) in a sentence. That an expression is meaningful does<lb/>not
               necessarily imply, however, that it is associated with a concept or an<lb/>idea. If I
               read or hear the compound name “round-square”, I am certainly<lb/>able to understand
               the meaning of the two composing words “round” and<lb/>“square” and I am also able to
               understand that the compound refers to<lb/>something which is, simultaneously, round
               and square. But in this case nei-<lb/>ther a thing nor an idea corresponds to the
               compound word, because the<lb/>expression “round-square” gives rise to a
                  contradiction.<note place="foot" xml:id="ftn6" n="6"> GP IV, p. 424 and p. 450.
               </note>
            </p>
            <p>Terms - as Leibniz writes in our text - may be considered in three<lb/>different
               ways: 1) from the extensional point of view, simply taking into ac-<lb/>count the
               object (or objects) which fall under them; 2) in a formal way,<lb/>looking at their
                  <hi rend="italic">intension</hi> or conceptual content; 3) according to the
               pecu-<lb/>liar way in which they «mention or show different thoughts». Leibniz
               char-<lb/>acterizes the first way of considering terms by means of the
               expression<lb/>“materially” (<hi rend="italic">materialiter</hi>), which is strongly
               reminiscent of the “material<lb/>supposition” and of the analogous “taking a term
               materially” of the<lb/>scholastic logicians.<note place="foot" xml:id="ftn7" n="7">
                  Cf. Ludovicus Carbo, <hi rend="italic">Introductionis in Logicam, sive Totius
                     Logicae Compendii Absolutis-<lb/>simi,</hi> Libr. VI, Venetiis, apud Io.
                  Baptistam et Io. Bernardum Sessam, 1597, p. 35: «Est ergo sup-<lb/>positio
                  materialis illa qua terminus significat se ipsum, seu vocem; ut “homo est vox”,
                  homo dici-<lb/>tur supponere materialiter, quia seipsum, seu vocem ipsam
                  significat: quae suppositio reperitur in<lb/>vocibus etiam non significativis».
               </note> However “materially” here alludes not to the linguis-<lb/>tic structure or
               grammatical form of the term itself - according to its pre-<lb/>vailing meaning in
               medieval times - but to the subject matter or “thing” to<lb/>which the term properly
               refers. Even though terms - with the exception of<lb/>proper names - name properties
               or peculiar aspects of things - as, for in-<lb/>stance, “yellow” or “round” or “man”
               - they refer, insofar as we consider<lb/>them materially, to the things themselves to
               which the properties are at-<lb/>tributed. As clearly emerges from <hi rend="italic"
                  >General Inquiries</hi>, for Leibniz there is no<lb/>difference between adjectives
               and substantives: both have the same logical<lb/>structure, the same “predicative
               character”, and hence both presuppose a<lb/>thing (or several things) of which they
               are the predicates.<note place="foot" xml:id="ftn8" n="8"> Cf. G. W. Leibniz, <hi
                     rend="italic">Allgemeine Untersuchungen über die Analyse der Begriffe und
                     Wahrhei-<lb/>ten,</hi> Lateinisch-Deutsch, Herausgegeben von F. Schupp, Felix
                  Meiner, Hamburg, 1982, p. 2 (C, p.<lb/>356). </note> The logical<lb/>structure of
               “yellow” or of “man”, for example, is interpreted by Leibniz as<lb/>implying
               something - a concrete thing - which is respectively yellow or a<lb/>man. Thus,
               different terms may refer to the same individual. Leibniz’s ex-<lb/>ample is that of
               “king” and “prophet”, which, in the case of the Biblical<lb/>David, have identical
               denotation.</p>
         <pb n="261" facs="UNITA/UNITA_261.jpg"/>
            <p>Terms, however, even though referring to the same object, may differ<lb/>
               <hi rend="italic">formally</hi> - i. e. regarding form, insofar as they mention
               different properties<lb/>or different aspects of the given object. Even though David,
               the individual,<lb/>is both a king and a prophet, being a king is a different
               property from that<lb/>of being a prophet. Here Leibniz uses “formally” <hi
                  rend="italic"> (formaliter</hi>) in sharp con-<lb/>trast to “materially” (<hi
                  rend="italic">materialiter</hi>) with the aim of distinguishing the form -<lb/>or
               intension or conceptual content - of a term, from its matter or extension.<lb/>In a
               short commentary to a book by the Jesuit Aloys Kümmet (edited in<lb/>1706), Leibniz
               sketches an analogous distinction between things and<lb/>concepts:</p>
            <p>Thus things which are really distinct usually are distinguished by<lb/>means of the
               senses, whereas things which differ in concept - i. e. those<lb/>which differ in
               their <hi rend="italic">formalities</hi>, even though they are not really different -<lb/>are distinguished
               by the mind. On a plane, for example, the triangle and<lb/>the trilateral are not two
               different things, but differ in concept only; there-<lb/>fore they are the same thing
               really, not formally. <note place="foot" xml:id="ftn9" n="9"> VE 5, p. 1086. Cf. also
                  here (same page): «Quicquid Subiecto inhaeret, formalitas dici po-<lb/>test, et
                  denominatio». </note>
            </p>
            <p>Thus, terms are said to differ <hi rend="italic">formally</hi>, because they differ
               in their <hi rend="italic">formali-<lb/>ties.</hi> The word “formality” <hi
                  rend="italic"> (formalitas) </hi> is typical of the Scotist tradition, but<lb/>it
               cannot be assumed as evidence of any scotistic influence on Leibniz’s
               phi-<lb/>losophy. As is well known, in Leibniz’s time the current scholastic
               doctrines<lb/>were characterized by an eclectic attitude which blurred the borders
               be-<lb/>tween the different philosophical positions. Therefore, even if an
               expression<lb/>is typical of a certain philosophical milieu, from its use one cannot
               automat-<lb/>ically infer that the user adhered to the theses commonly held by people
               be-<lb/>longing to that milieu.</p>
            <p>In a revealing passage of the <hi rend="italic">New Essays,</hi> which is also very
               important for<lb/>setting the terminological distinction between <hi rend="italic"
                  >extension</hi> and <hi rend="italic">intension</hi> of a<lb/>term, “formality” is
               explicitly held by Leibniz as a synonym of the word “idea”:</p>
         <p>... when I say <hi rend="italic">Every man is an animal</hi> I mean that all men are included<lb/>amongst
               all the animals; but at the same time I mean that the idea of animal<lb/>is included
               in the idea of man. “Animal” comprises more individual than<lb/>“man” does, but “man”
            comprises more ideas or more<hi rend="italic"> formalities</hi>: one has<lb/>more instances, the other more
               degrees of reality: one has the greater exten-<lb/>sion the other the greater
               intension. <note place="foot" xml:id="ftn10" n="10"> Cf. A VI, vi, p. 486 (english
                  transl. by P. Remnant and J. Bennett, Cambridge, Cambridge<lb/>University Press,
                  1981). </note>
            </p>
         <pb n="262" facs="UNITA/UNITA_262.jpg"/>
            <p>Hence we may paraphrase what Leibniz writes on formalities saying that<lb/>terms
               which differ formally are different in virtue of their conceptual con-<lb/>tent and
               that not all conceptual distinctions give rise to real distinctions (a<lb/>conclusion
               which maintains a Scotist flavour as well).</p>
            <p>Till now I have attempted to illustrate what Leibniz properly means<lb/>by the first
               two ways of considering a term, out of the three mentioned<lb/>above. The task has
               been not particularly difficult, because there are many<lb/>other texts in which
               Leibniz proposes or discusses the distinction between<lb/>these two different ways.
               Problems arise, however, with the “third way”,<lb/>which seems to have been
               explicitly recognized only in our text. As Leibniz<lb/>observes, terms considered
               formally may express the same conceptual con-<lb/>tent in a different guise. And what
               the “third way” of considering a term<lb/>takes into account is precisely the guise
               or manner in which a term “ar-<lb/>ranges” or “presents” a given conceptual content.
               Clearly, this “arrange-<lb/>ment” or “presentation” must depend partly <hi
                  rend="italic">on language</hi> - i. e. on the dif-<lb/>ferent possibilities we
               have of expressing the different aspects of an idea. At<lb/>the same time it seems to
               depend also on the fact that a term expresses a<lb/>complex idea or formality. If two
               terms express the same simple idea - i. e.<lb/>an idea which cannot be further
               analyzable - then they <hi rend="italic">name</hi> the same<lb/>thought in different
               ways and do not «mention or show different thoughts»<lb/>(as Leibniz writes).</p>
            <p>Leibniz tries to clarify this point with his standard example of the
               terms<lb/>“triangle” and “trilateral”. “Triangle” and “trilateral” differ from
               each<lb/>other because they represent different ways of considering the same idea
               (i.<lb/>e. that of a geometrical figure with three sides and three angles).
               “Triangle”<lb/>and “trilateral” are not different <hi rend="italic">formalities</hi>,
               but simply different ways of<lb/>thinking of the same formality. That the formality
               is the same - Leibniz<lb/>claims - can easily be shown by means of conceptual
               analysis: if we proceed<lb/>to analyse the concept corresponding to “triangle” we
               individuate a set of<lb/>basic notions or ideas which characterize the concept
               corresponding to “tri-<lb/>lateral” as well.</p>
            <p>In our text the <hi rend="italic">formalities</hi> are objective, independent of the
               different<lb/>ways in which we can consider them, whereas the “way of
               considering”<lb/>gives rise to <hi rend="italic">different expressions</hi> of the
               same formality. Thus, the third way<lb/>of considering a term seems to be determined
               by the way (or ways) in<lb/>which the <hi rend="italic">formalities</hi> are
               considered by the mind. As I observed above, this<lb/>is apparently the only text in
               which Leibniz makes this claim, and what is<lb/>puzzling about it is that it seems to
               be in neat contrast with Leibniz’s main<lb/>doctrine on formalities. If we look again
               at the passage previously quoted<lb/>from Leibniz’s commentary on Kümmet’s book, we
               see that the two terms</p>
         <pb n="263" facs="UNITA/UNITA_263.jpg"/>
            <p>“triangle” and “trilateral” are there assumed plainly to express <hi rend="italic"
                  >two differ-<lb/>ent formalities</hi>, not two different ways of considering the
               same formality. In<lb/>Leibniz’s words, “triangle” and “trilateral” «are the same
               thing really, not<lb/>formally».<note place="foot" xml:id="ftn11" n="11"> VE 5, p.
                  1086. </note> One can hope for more light looking at another passage from<lb/>
               <hi rend="italic">New Essays</hi>, in which the “triangle-trilateral” issue is
               explicitly considered:</p>
         <p>... someone who said<hi rend="italic"> The triangle and the trilateral are not the same</hi><lb/> would be
               wrong, since if we consider it carefully we find that three sides<lb/>and three
               angles always go together... However, one can still say in the ab-<lb/>stract that
            <hi rend="italic">triangularity is not trilaterality</hi>, or that the formai causes of the<lb/>triangle and
               of the trilateral are not the same, as the philosophers put it.<lb/>They are
               different aspects of one and the same thing. <note place="foot" xml:id="ftn12" n="12"
                  > A VI, vi, p. 363 (english transl. by P. Remnant and J. Bennett). </note>
            </p>
            <p>“Triangularity” and “trilaterality” are said to be “the formal causes” (but
               a<lb/>more litteral translation would be: “the formal reasons”) respectively, of
               the<lb/>triangle and of the trilateral; and they are “different aspects of one and
               the<lb/>same thing”. In this case too, Leibniz seems to agree that “triangle”
               and<lb/>“trilateral” correspond to two different <hi rend="italic">formalities.</hi>
               What is the same is not<lb/>the idea, but the “thing” - presumably all the concrete
               instances of trian-<lb/>gles, which are, simultaneously, triangular and
               trilateral.</p>
            <p>As Benson Mates observes, the “triangle-trilateral” issue is also of
               some<lb/>importance for understanding Leibniz’s attitude towards the
               substitutivity<lb/>principle <hi rend="italic">salva veritate</hi> (and, hence,
               Leibniz’s attitude towards identity).<note place="foot" xml:id="ftn13" n="13"> Cf. B.
                  Mates, <hi rend="italic">The Philosophy of Leibniz. Metaphysics and Language</hi>,
                  Oxford-New York,<lb/>Oxford University Press, 1986, pp. 123 ff. </note>
               <lb/>The expressions “triangle” and “trilateral” are Leibniz’s favourite
               examples<lb/>for showing that, in some peculiar contexts, the substitutivity
               principle fails.<lb/>And, if we look at the way in which Leibniz characterizes these
               contexts,<lb/>we can gain precious hints concerning <hi rend="italic"
                  >formalities.</hi> Consider, for example,<lb/>the following passage from a text
               written about 1686:</p>
         <p>A = B means that A and B are the<hi rend="italic"> same</hi> - i. e. that one can be substi-<lb/>tuted for
               the other anywhere. (Unless it is precluded, as happens in those<lb/>contexts in
               which one states that a given term is considered in a peculiar re-<lb/>spect. For
               instance: even though the triangle and the trilateral are identical,<lb/>if one
            states<hi rend="italic"> The triangle insofar as it is a triangle has 180 degrees</hi>, one can-<lb/>not
            substitute<hi rend="italic"> trilateral</hi> for <hi rend="italic">triangle</hi>. Here there is something material
               which<lb/>precludes the substitution). <note place="foot" xml:id="ftn14" n="14"> VE
                  8, p. 1935 (C, p. 261). </note>
            </p>
         <pb n="264" facs="UNITA/UNITA_264.jpg"/>
            <p>What is interesting here is, first of all, Leibniz’s use of the expression
               “ma-<lb/>terial”, which has, in this circumstance, a meaning very near to that of
               the<lb/>medieval tradition, as opposed to the meaning considered above.
               Second,<lb/>we may observe that preventing the substitutivity in the given context is
               the<lb/>fact that a term <hi rend="italic">is considered</hi> in a peculiar respect.
               Once more it is the <hi rend="italic">way<lb/>of considering</hi> which puts in
               evidence the <hi rend="italic">formal</hi> difference between two<lb/>terms, which
               are (the formai difference notwithstanding) identical. In <hi rend="italic"
                  >Gen-<lb/>eral Inquiries</hi>, propositions which do not allow the substitutivity
               of equiva-<lb/>lents are called “formal” and “reflexive”: they «assume one of the
               coinci-<lb/>dentials in such a way that it is distinguished from the others» and they
               «do<lb/>not so much speak about a thing, as about <hi rend="italic">our way of
                  conceiving it»</hi>
               <note place="foot" xml:id="ftn15" n="15"> G. W. Leibniz, <hi rend="italic">Allgemeine
                     Untersuchungen...,</hi> p. 34 (C, pp. 366-367). </note> (em-<lb/>phasis mine).
               Similarly, in a text on geometry written about the same time<lb/>as <hi rend="italic"
                  >General Inquiries</hi>, Leibniz writes a short remark with the aim of
               empha-<lb/>sizing that the substitutivity principle for terms holds in all the
               propositions<lb/>which are “direct” (<hi rend="italic">directae) </hi> - i. e. in all
               those propositions «which do not<lb/>take into account the way of considering <hi
                  rend="italic"> (nec in ipsum considerandi modum<lb/>reflectuntur</hi>)».<note
                  place="foot" xml:id="ftn16" n="16"> Cf. M. Mugnai, <hi rend="italic">Leibniz’s
                     Theory of Relations</hi>, Stuttgart, F. Steiner Verlag, 1992, p. 147. </note>
               Thus, Leibniz seems to be claiming that the concept corre-<lb/>sponding to “triangle”
               is different from that corresponding to “trilateral”<lb/>(their <hi rend="italic"
                  >formalities</hi> are different); and, consequently, he seems to assume:
               1)<lb/>that it is the <hi rend="italic">way of considering</hi> the denotation of a
               term that gives rise to a<lb/>
               <hi rend="italic">formality; </hi> 2) that his <hi rend="italic">salva veritate</hi>
               principle provides a criterion for the<lb/>identity of <hi rend="italic">things</hi>
               not of <hi rend="italic">concepts.</hi>
            </p>
            <p>Therefore, the problem arises of how to evaluate the position as-<lb/>sumed by
               Leibniz in our text. If the “third way” of considering a term is<lb/>explicitly
               mentioned only here, then one may argue that it is something<lb/>which does not
               correspond to Leibniz’s prevailing view. And hence it can<lb/>be assumed to be a
               position held on this occasion and suddenly dismissed.<lb/>But it seems to me that
               this is not so. I have the impression that what we<lb/>see here is Leibniz’s
               ambiguous attitude towards nominalism. The ambigu-<lb/>ity consists mainly of
               Leibniz’s acceptance of the existence of ideas and<lb/>essences of things, on the one
               hand, and of his simultaneous adhesion, on<lb/>the other, to a kind of cautious
               nominalism which undermines the impor-<lb/>tance of abstract terms and ideas, on the
                  other.<note place="foot" xml:id="ftn17" n="17"> Cf. M. Mugnai, <hi rend="italic"
                     >Leibniz’s Theory...,</hi> pp. 18 ff. </note>
            </p>
            <p>At the beginning of <hi rend="italic">General Inquiries</hi>, for example, Leibniz
               claims that<lb/>the fact that “triangle” and “trilateral” coincide can easily be
               shown by</p>
         <pb n="265" facs="UNITA/UNITA_265.jpg"/>
            <p>means of analysis, and states in a odd way the following general principle<lb/>about
               logical coincidence: if terms «are analysed until it appears <hi rend="italic">a
                  priori<lb/>
               </hi>that they are possible, and if the same terms appear formally, then
               different<lb/>terms are the same».<note place="foot" xml:id="ftn18" n="18"> G. W.
                  Leibniz, <hi rend="italic">Allgemeine Untersuchungen</hi>..., p. 20 (C, p. 362);
                  english transl. in: Leibniz,<lb/>
                  <hi rend="italic">Logicai Papers,</hi> ed. by G. H. R. Parkinson, Oxford,
                  Clarendon Press, 1966, p. 53. </note> Given two terms <hi rend="italic">A</hi> and
                  <hi rend="italic">B,</hi> they coincide «if the one<lb/>can be substituted in
               place of the other without loss of truth, or if, on<lb/>analysing each of the two by
               substitution of their values (i. e. of their defini-<lb/>tions) in place of the
               terms, the same terms appear on both sides».<note place="foot" xml:id="ftn19" n="19">
                  <hi rend="italic">Ibid.</hi>
               </note> As<lb/>Leibniz specifies, “the same terms” means “the same <hi rend="italic"
                  >formally</hi>” <hi rend="italic">-</hi> i. e. at the<lb/>end of the analysis it
               is necessary that on both sides of the copula the same<lb/>
               <hi rend="italic">formalities</hi> or ideas appear as component parts of <hi
                  rend="italic">A</hi> and <hi rend="italic">B:</hi> «the same I<lb/>mean, formally
               - for example, if L, <hi rend="italic">M</hi> and N appear on both sides».<note
                  place="foot" xml:id="ftn20" n="20">
                  <hi rend="italic">Ibid.</hi>
               </note> Here<lb/>Leibniz seems to hold a point of view quite similar to that of our
               list of<lb/>definitions: terms like “triangle” and “trilateral” have the same ideal
               or<lb/>conceptual content. Presumably, then, they are not different <hi rend="italic"
                  >formalities</hi>, but<lb/>simply different ways of expressing the same <hi
                  rend="italic">formalities</hi> or the same finite<lb/>set of ideas.</p>
            <p>That “triangle” and “trilateral” are different espressions of the same<lb/>idea can
               be inferred as well from the following short passage from <hi rend="italic"
                  >New<lb/>Essays</hi>:</p>
         <p>... I have remarked earlier that there are redundant ways of expressing<lb/><hi rend="italic"
            > ideas</hi>,
            which add nothing to <hi rend="italic"
               >things</hi>. It is as though someone were to say “By<lb/> <hi rend="italic"
                  >Triquetrum</hi>
               I mean a trilateral triangle” and to infer from that that some tri-<lb/>laterals are
               not triangular. <note place="foot" xml:id="ftn21" n="21"> A Vi, vi, p. 423 (english
                  transl. by P. Remnant and J. Bennett). </note>
            </p>
            <p>To Des Bosses (about 1712), however, Leibniz clearly writes that «Triangle<lb/>and
               Trilateral are the same being, but different terms: they differ formally,<lb/>not
                  materially».<note place="foot" xml:id="ftn22" n="22"> GP II, pp. 470-71. </note>
            </p>
            <p>Thus, in some texts Leibniz writes that terms like “triangle” and “trilat-<lb/>eral”
               differ formally, whereas in other texts, written at about the same pe-<lb/>riod, he
               plainly states that “triangle” and “trilateral” are only different ex-<lb/>pressions
               of the same formality or of an identical set of ideas. Analogously,<lb/>whereas in
               most of his writings Leibniz holds that the <hi rend="italic">salva veritate</hi>
               princi-<lb/>ple provides a criterion for the identity of things, there are texts (<hi
                  rend="italic">General</hi>
            </p>
         <pb n="266" facs="UNITA/UNITA_266.jpg"/>
            <p>
               <hi rend="italic">Inquiries,</hi> for instance, as we have seen) in which he claims
               that terms like<lb/>“triangle” and “trilateral” are interchangeable <hi rend="italic"
                  >salva veritate</hi>, because they<lb/>express the same idea (or ideas). All this
               can be easily explained if we take<lb/>into account that for Leibniz the existing
               world splits up, as it were, into<lb/>two different realms: that of (existing) things
               and that of the ideas, or<lb/>essences, corresponding to things. Both realms are
               “real” and each existing<lb/>thing has an ideal counterpart - an essence which, like
               the thing itself, is<lb/>one in number. The same essence, however, may be expressed
               by several<lb/>definitions:</p>
            <p>To reinforce the distinction between essence and definition, bear in<lb/>mind that
               although a thing has only one essence, this can be expressed by<lb/>several
               definitions, just as the same town can be represented by different<lb/>drawings in
               perspective depending on the direction from which it is<lb/>viewed. <note
                  place="foot" xml:id="ftn23" n="23"> A VI vi, p. 294 (english transl. by P. Remnant
                  and J. Bennett). </note>
            </p>
            <p>In the same vein, as we have seen, Leibniz observes that «there are redun-<lb/>dant
               ways of expressing <hi rend="italic">ideas</hi>, which add nothing to things».<note
                  place="foot" xml:id="ftn24" n="24"> Cf. A VI, vi, p. 423. </note> Thus, if,
               on<lb/>the one hand, different definitions of the same essence are not
               different<lb/>essences, but simply different ways of presenting one and the same
               essence,<lb/>on the other, a conceptual distinction does not correspond to any
               distinc-<lb/>tion among things.</p>
            <p>In Leibniz’s logical ontology, terms considered as linguistic items ex-<lb/>press
               ideas or concepts, and denote things. Concepts and ideas are objec-<lb/>tive,
               non-linguistic in nature and independent of their expressions: they do<lb/>not exist
               in space and time, but they have a kind of existence in what Leib-<lb/>niz names “the
               realm of ideas” <hi rend="italic"> (regio idearum</hi>).<note place="foot"
                  xml:id="ftn25" n="25"> On Leibniz’s use of the expression <hi rend="italic">regio
                     idearum</hi> cf. VE 7, p. 1461: «Ego soleo dicere esse<lb/>aliquid in regione
                  idearum itaque spatium numericum videtur esse tantum consideratione
                  quadam<lb/>existentiam in regione idearum, quasi in spatio aut tempore»). </note>
               Hence, insofar as Leibniz<lb/>tries to be coherent with his “provisional
                  nominalism”,<note place="foot" xml:id="ftn26" n="26"> On Leibniz’s nominalism <hi
                     rend="italic">per provisionem</hi> cf. Grua, p. 547 and M. Mugnai, <hi
                     rend="italic">Astrazione e<lb/>realtà. Saggio su Leibniz</hi>, Milano,
                  Feltrinelli, 1976, pp. 133 ff.; B. Mates, <hi rend="italic">The Philosophy of
                     Leibniz,<lb/>
                  </hi>p. 170 ff. </note> it is quite natural<lb/>that he should look at terms as
               directly denoting things; whereas if he em-<lb/>phasizes the “realistic” side of his
               ontology, he considers terms as repre-<lb/>senting first concepts or ideas. That the
               nominalistic mood is prevalent,<lb/>however, is easily shown by our text, where
               Leibniz, after having distin-<lb/>guished the <hi rend="italic">formalities</hi> from
               the ways of considering them, states that, in<lb/>the last analysis, what really
               matters is “staying with things”.</p>
         <pb n="267" facs="UNITA/UNITA_267.jpg"/>
            <p>Another element of interest in our text can be found in the second<lb/>passage
               between brackets, of which I offer a translation:</p>
            <p>Thus rime and place are neither subjects nor attributes. And the same<lb/>holds for
               numbers and orders. Thus “ten” is not an attribute of any thing.<lb/>In fact the
               number ten cannot be said either of the aggregate or of the sin-<lb/>gle numbered
               things. The same holds for relations which are common to<lb/>the related things - for
               instance, the similarity of two things. Hence, there<lb/>are attributes which are at
               the same time in several subjects: of this kind<lb/>are, therefore, orders and even
               time and place.</p>
            <p>The most relevant points of the entire passage are two: 1) Leibniz, after<lb/>having
               asked himself if there is a third kind of “things” besides subjects<lb/>and
               attributes, answers the question positively; 2) numbers and relations are<lb/>said to
               share some common feature, which determines their belonging to<lb/>this third kind of
               entities.</p>
            <p>That numbers and relations have something in common is a claim<lb/>widely held by
               scholastic authors. Consider, for example, what Ockham<lb/>says in the following
               passage: «Similarity stays for (<hi rend="italic">supponit) </hi> two
               similar<lb/>things considered simultaneously, as in the case of the name of a
               number<lb/>[...] Thus Socrates is not two, even though Socrates and Plato are
                  two».<note place="foot" xml:id="ftn27" n="27"> Ockham, <hi rend="italic"
                     >Quodlibeta septem,</hi> ed. by J. C. Wey in <hi rend="italic">Opera
                     Philosophica et Theologica,</hi> New<lb/>York, The Franciscan Institute, St.
                  Bonaventure, vol. IX, 1980, p. 616. </note>
               <lb/>Relying on this scholastic tradition, Leibniz thinks that numbers are
               neither<lb/>predicates of single things nor predicates of the aggregate (or set) of
               num-<lb/>bered things: they are attributes of a special kind, which, like relations
               «are<lb/>simultaneously in several subjects». What is striking here is that, at
               first<lb/>glance, Leibniz seems to recognize that relations (and numbers) belong
               to<lb/>an autonomous “third kind” of beings besides subjects (substances) and
               at-<lb/>tributes, contrary to his “official” doctrine that relations <hi
                  rend="italic">are not beings</hi> (<hi rend="italic">en-<lb/>tia</hi>).<note
                  place="foot" xml:id="ftn28" n="28"> Cf. M. Mugnai, <hi rend="italic">Leibniz’s
                     Theory of Relations</hi>..., pp. 18 ff. </note> On a more careful reading,
               however, we may attempt to maintain<lb/>some coherence on Leibniz’s part on this
               point. In fact, Leibniz does not<lb/>explicitly state that numbers and relations are
                  <hi rend="italic">beings</hi> (<hi rend="italic">entia) </hi> - a
               conclusion<lb/>which plainly contrasts with his main ontological views. He writes
               that all<lb/>beings are substances or attributes; he asks himself if there is
               something<lb/>“third” besides substances and attributes and answers that numbers and
               re-<lb/>lations are this third kind of “thing”. Clearly, if the realm of all beings
               di-<lb/>vides itself into substances and attributes and if relations and numbers
               are<lb/>something “third”, in addition to substances and attributes, then
               relations<lb/>and numbers cannot be beings <hi rend="italic"> (entia) </hi> in the
               same sense in which sub-</p>
         <pb n="268" facs="UNITA/UNITA_268.jpg"/>
            <p>stances and attributes are. And if they are attributes, they must be
               attributes<lb/>of a special kind.</p>
            <p>In this case we are faced with a typical Leibnizian attitude about rela-<lb/>tions:
               on the one hand, Leibniz emphasizes that relations <hi rend="italic">inhere
                  simultane-<lb/>ously in several subjects; </hi> on the other, he claims that <hi
                  rend="italic">because</hi> of that very<lb/>fact, they cannot be real attributes
               of things. As is well known, Leibniz<lb/>agrees with the scholastic thesis that the
               same accident cannot inhere in<lb/>things numerically distinct. Therefore, relations
               are not real accidents or<lb/>real properties, but “mere ideal things”, as we read in
               the fifth letter to<lb/>Clarke.<note place="foot" xml:id="ftn29" n="29"> Cf. GP VII,
                  p. 401. </note> Otherwise stated: relations do not inhere properly in things -
               their<lb/>kind of inherence is merely mental, a product of our thinking-together
               sev-<lb/>eral things at once. Thus, if Leibniz plainly recognizes the true logical
               na-<lb/>ture of relations - insofar as he recognizes their “multiple inherence” -
               this<lb/>happens at the expense, as it were, of their ontological reality.
               Relations<lb/>have a “diminished being” and are not real, as substances and
                  accidents<lb/>are.<note place="foot" xml:id="ftn30" n="30"> Cf. M. Mugnai, <hi
                     rend="italic">Leibniz’s Theory of Relations</hi>..., pp. 18 ff. </note>
            </p>
            <p>A major element of interest in our text lies in the fact that here
               Leibniz<lb/>mentions, as a characteristic property of relations, “multiple inherence”
               and<lb/>states that <hi rend="italic">numbers</hi> share with relations this same
               property. Clearly, insofar<lb/>as numbers inhere simultaneously in several subjects,
               they have a mere<lb/>mental nature as well. And, properly speaking, their “inherence”
               is not real,<lb/>but is due to our capacity to think-together a plurality of things
               in the same<lb/>act of thought. Chauvin’s <hi rend="italic">Lexicon
                  philosophicum</hi> helps us better understand<lb/>of what Leibniz says about
                  numbers.<note place="foot" xml:id="ftn31" n="31"> Cf. Stephanus Chauvin, <hi
                     rend="italic">Lexicon Philosophicum</hi>, Leeuwarden, 1713 (second edition) -
                  Re-<lb/>printed: Duesseldorf, Stern Verlag, Janssen &amp; Co, 1967, p. 444.
               </note> Under the entry <hi rend="italic">Numerus</hi>, Chauvin<lb/>writes that a
               number «says a plurality of unities» and that it implies a «si-<lb/>multaneity of the
               same unities» - i. e. that a number expresses a <hi rend="italic">relation</hi>
               be-<lb/>tween our minds and a plurality of things. Therefore a number is made
               up<lb/>of matter and form: matter is constituted of «things which can be
               num-<lb/>bered, as, for instance, coins»; form is «the idea by means of which a
               plural-<lb/>ity of things is reduced to unity» by means of an act of the
               understanding.<lb/>Hence, whereas form is dependent on mind, matter is not. In this
               case too,<lb/>the analogy with relations is very strong: whereas for a (real)
               relation it is<lb/>necessary that some individuate exist and that the mind grasp them
               in an<lb/>act of thought, for the existence of numbers it is necessary that a
               plurality<lb/>of things (objectively) exist and that the mind have the possibility of
               grasp-<lb/>ing them in a single act of thought. As Chauvin observes - and surely
               Leib-</p>
         <pb n="269" facs="UNITA/UNITA_269.jpg"/>
            <p>niz would agree - the <hi rend="italic">form</hi> of a number is not something which
               adds itself<lb/>really to the numbered things. The idea which constitutes the <hi
                  rend="italic">form</hi> of a<lb/>given number is a modification of the mind of the
               person who is thinking -<lb/>it is not a modification of the numbered things. This
                  <hi rend="italic">form</hi> «may be named<lb/>an “extrinsic denomination”».<note
                  place="foot" xml:id="ftn32" n="32">
                  <hi rend="italic">Ibid.</hi>
               </note>
            </p>
         <p> This shows once more how deep the influence of scholastic (and late-<lb/>scholastic)
               doctrines is on Leibniz’s ontology and philosophy of logic.</p>
      </body>
   </text>
</TEI>
