Edmund Husserl: Logische Untersuchungen 9783050050133, 9783050043913

Citation preview

Edmund Husserl

Logische Untersuchungen

Klassiker Auslegen Herausgegeben von Otfried Höffe Band 35

Otfried Höffe ist o. Professor für Philosophie an der Universität Tübingen

Edmund Husserl

Logische Untersuchungen Herausgegeben von Verena Mayer unter Mitwirkung von Christopher Erhard

Akademie Verlag

Titelabbildung: Edmund Husserl, © Husserl-Archives Leuven

Bibliografische Information der Deutschen Nationalbibliothek Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliografie; detaillierte bibliografische Daten sind im Internet über http://dnb.d-nb.de abrufbar.

ISBN: 978-3-05-004391-3 © Akademie Verlag GmbH, Berlin 2008 Das eingesetzte Papier ist alterungsbeständig nach DIN/ISO 9706. Alle Rechte, insbesondere die der Übersetzung in andere Sprachen, vorbehalten. Kein Teil dieses Buches darf ohne schriftliche Genehmigung des Verlages in irgendeiner Form – durch Photokopie, Mikroverfilmung oder irgendein anderes Verfahren – reproduziert oder in eine von Maschinen, insbesondere von Datenverarbeitungsmaschinen, verwendbare Sprache übertragen oder übersetzt werden. Gesamtgestaltung: K. Groß, J. Metze, Chamäleon Design Agentur Berlin Satz: Veit Friemert, Berlin Druck und Bindung: MB Medienhaus, Berlin Printed in the Federal Republic of Germany



Inhalt Zitierweise und Siglen . . . . . . . . . . . . . . . . . . . . VII Vorwort . . . . . . . . . . . . . . . . . . . . . . . . . . . IX 1. Einleitung Verena Mayer . . . . . . . . . . . . . . . . . . . . . . . . .

1

2. Husserl’s concept of Pure Logic (Prolegomena, §§ 1–16, 62–72) Richard Tieszen . . . . . . . . . . . . . . . . . . . . . . . . . .

9

3. Husserl’s Arguments against Logical Psychologism (Prolegomena, §§ 17–61) Robert Hanna . . . . . . . . . . . . . . . . . . . . . . . . . .

27

4. Husserls phänomenologische Semiotik (I. Logische Untersuchung, §§ 1–23) Vittorio De Palma . . . . . . . . . . . . . . . . . . . . . . .

43

5. Die Objektivität der Bedeutung (I. Logische Untersuchung, §§ 24–35) Gianfranco Soldati . . . . . . . . . . . . . . . . . . . . . . . .

61

6. Zugang zum Idealen: Spezies und Abstraktion (II. Logische Untersuchung, §§ 1–12) Peter Simons . . . . . . . . . . . . . . . . . . . . . . . . . .

77

7. The Critique of Empiricist Accounts of Abstraction (II. Logische Untersuchung, §§ 13–42) A. D. Smith . . . . . . . . . . . . . . . . . . . . . . . . . .

93

VI

Inhalt

8. Wholes, Parts, and Phenomenological Methodology (III. Logische Untersuchung) John J. Drummond . . . . . . . . . . . . . . . . . . . . . . . 105 9. Grammatik und Intentionalität (IV. Logische Untersuchung) Jocelyn Benoist . . . . . . . . . . . . . . . . . . . . . . . . . . 123 10. Intentionalität und Bewusstsein (V. Logische Untersuchung, §§ 1–21, Beilage der VI. Untersuchung ) Dan Zahavi . . . . . . . . . . . . . . . . . . . . . . . . . . . 139 11. Die Bedeutung objektivierender Akte (V. Logische Untersuchung, §§ 22–45) Verena Mayer/Christopher Erhard . . . . . . . . . . . . . . . . . 159 12. Intention und Erfüllung, Evidenz und Wahrheit (VI. Logische Untersuchung, §§ 1–39, 67–70) Rudolf Bernet . . . . . . . . . . . . . . . . . . . . . . . . . . 189 13. Kategoriale Anschauung (VI. Logische Untersuchung, §§ 40–66) Dieter Lohmar . . . . . . . . . . . . . . . . . . . . . . . . . . 209 Auswahlbibliographie . . . . . . . . . . . . . . . . . . . . . . 238 Personenregister . . . . . . . . . . . . . . . . . . . . . . . . 243 Sachregister . . . . . . . . . . . . . . . . . . . . . . . . . . 245 Hinweise zu den Autoren . . . . . . . . . . . . . . . . . . . . 248

Zitierweise und Siglen Husserls Texte werden nach der Husserliana-Ausgabe zitiert: Edmund Husserl 1950 ff.: Husserliana. Gesammelte Werke. Den Haag/Dordrecht. Als Abkürzung wird die Sigle „Hua“ mitsamt Bandangabe in römischen und Seitenangabe in arabischen Ziffern verwendet. Bei englischsprachigen Beiträgen sind die von den Autoren individuell angegebenen Textausgaben und Zitationsweisen maßgeblich. Die Materialien­bände werden mit „Hua Mat“ mitsamt Band- und Seitennummer zitiert. Folgende Ausgaben der Husserliana werden verwendet: Hua I

Cartesianische Meditationen und Pariser Vorträge. Hrsg. und eingeleitet von Stephan Strasser. Nachdruck der 2. verb. Auflage. 1991 Hua II Die Idee der Phänomenologie. Fünf Vorlesungen. Hrsg. und eingeleitet von Walter Biemel. Nachdruck der 2. erg. Auflage. 1973 Hua III Ideen zu einer reinen Phänomenologie und phänomenologischen Philosophie. Erstes Buch: Allgemeine Einführung in die reine Phänomenologie. (= Ideen I) In zwei Bänden. 1. Halbband: Text der 1.–3. Auflage; 2. Halbband: Ergänzende Texte (1912–1929). Neu hrsg. von Karl Schuhmann. Nachdruck. 1976 Hua IV Ideen zu einer reinen Phänomenologie und phänomenologischen Philosophie. Zweites Buch: Phänomenologische Untersuchungen zur Konstitution. Hrsg. von Marly Biemel. 1952 (=Ideen II) Hua VI Die Krisis der europäischen Wissenschaften und die transzendentale Phänomenologie. Eine Einleitung in die phänomenologische Philosophie. Hrsg. von Walter Biemel. Nachdruck der 2. verb. Auflage. 1976 Hua IX Phänomenologische Psychologie. Vorlesungen Sommersemester 1925. Hrsg. von Walter Biemel. 2. verb. Auflage. 1968 Hua X Zur Phänomenologie des inneren Zeitbewußtseins (1893–1917). Hrsg. von Rudolf Boehm. Nachdruck der 2. verb. Auflage. 1969 Hua XI Analysen zur passiven Synthesis. Aus Vorlesungs- und Forschungsmanuskripten (1918–1926). Hrsg. von Margot Fleischer. 1966 Hua XII Philosophie der Arithmetik. Mit ergänzenden Texten (1890–1901). Hrsg. von Lothar Eley. 1970 Hua XIII Zur Phänomenologie der Intersubjektivität. Texte aus dem Nachlass. Erster Teil: 905–1920. Hrsg. von Iso Kern. 1973 Hua XVI Ding und Raum. Vorlesungen 1907. Hrsg. von Ulrich Claesges. 1973 Hua XVII Formale und transzendentale Logik. Versuch einer Kritik der logischen Vernunft. Mit ergänzenden Texten. Hrsg. von Paul Janssen. 1974 Hua XVIII Logische Untersuchungen. Erster Band: Prolegomena zur reinen Logik. Text der 1. und 2. Auflage. Hrsg. von Elmar Holenstein. 1975 Hua XIX/1 Logische Untersuchungen. Zweiter Band. Erster Teil. Untersuchungen zur Phänomenologie und Theorie der Erkenntnis. Hrsg. von Ursula Panzer. 1984 Hua XIX/2 Logische Untersuchungen. Zweiter Band. Zweiter Teil. Untersuchungen zur Phänomenologie und Theorie der Erkenntnis. Hrsg. von Ursula Panzer. 1984

VIII

Zitierweise und Siglen

Hua XX/1 Logische Untersuchungen. Ergänzungsband. Erster Teil. Entwürfe zur Umarbeitung der VI. Untersuchung und zur Vorrede für die Neuauflage der Logischen Untersuchungen (Sommer 1913). Hrsg. von Ullrich Melle. 2002 Hua XX/2 Logische Untersuchungen. Ergänzungsband. Zweiter Teil. Texte für die Neufassung der VI. Untersuchung: Zur Phänomenologie des Ausdrucks und der Erkenntnis (1893/94–1921). Hrsg. von Ullrich Melle. 2005 Hua XXII Aufsätze und Rezensionen (1890–1910). Hrsg. von Bernhard Rang. 1979 Hua XXIII Phantasie, Bildbewußtsein, Erinnerung. Zur Phänomenologie der anschaulichen Vergegenwärtigungen. Texte aus dem Nachlass (1898–1925). Hrsg. von Eduard Marbach. 1980 Hua XXIV Einleitung in die Logik und Erkenntnistheorie. Vorlesungen 1906/07. Hrsg. von Ullrich Melle. 1984 Hua XXVI Vorlesungen über Bedeutungslehre. Sommersemester 1908. Hrsg. von Ursula Panzer. 1987 Hua XXVIII Vorlesungen über Ethik und Wertlehre (1908–1914). Hrsg. von Ullrich Melle. 1988 Hua XXXI Aktive Synthesen. Aus der Vorlesung „Transzendentale Logik“ 1920/21. Ergänzungsband zu „Analysen zur passiven Synthesis“. Hrsg. von Roland Breeur. 2000 Hua XXXVI Transzendentaler Idealismus. Texte aus dem Nachlass (1908–1921). Hrsg. von Robin D. Rollinger in Verbindung mit Rochus Sowa. 2003 Hua Mat II Logik. Vorlesung 1902/03. Hrsg. von Elisabeth Schumann. 2001 Hua Mat III Allgemeine Erkenntnistheorie. Vorlesung 1902/03. Hrsg. von Elisabeth Schuhmann. 2001 Hua Mat V Urteilstheorie. Vorlesung 1905. Hrsg. von Elisabeth Schuhmann. 2002

Zitierte Manuskripte aus dem Nachlass: Ms. L Ms. A

Bernauer Manuskripte Mundane Phänomenologie

Vorwort Der vorliegende Kommentarband zu den Logischen Untersuchungen erscheint zu Husserls 150. Geburtstag im Jahr 2009. Damit ein solches Gemeinschaftswerk rechtzeitig fertig gestellt werden konnte, waren vielfache Anstrengungen notwendig. Ich bedanke mich bei Rudolf Bernet und dem Husserl-Archiv Leuwen für die freundliche Überlassung des Titelfotos; bei Dieter Lohmar für die bereitwillige Unterstützung bei der Autorensuche; und nicht zuletzt bei den Autoren, die ihre Manuskripte mit bewundernswerter Pünktlichkeit eingereicht haben. Mein besonderer Dank gilt Christopher Erhard, der den Band von der Konzeption bis zur Fertigstellung sachkundig und engagiert begleitet hat. Verena Mayer

1 Verena Mayer

Einleitung

Husserls Logische Untersuchungen sind eines der detailliertesten Bücher, die je über die Phänomenologie des Bewusstseins geschrieben wurden. Die beiden 1900 (Prolegomena) und 1901 (Logische Untersuchungen) erschienenen Bände waren das Ergebnis zehnjähriger Arbeit und sind unter mannigfachen Geburtswehen publiziert worden. Insbesondere der erste Band, die Prolegomena zu einer reinen Logik, wurde sogleich von der Fachwelt viel beachtet; dieses Buch scheint dem damals vorherrschenden Psychologismus den Todesstoß versetzt zu haben. Auch der zweite Band hat eine breite Wirkung entfaltet. Er enthält nicht nur Beiträge zu den in den philosophischen Schriften der Zeit vielfach diskutierten Themen Bedeutung, Begriff, Urteil, Erkenntnis und Wahrheit, sondern weitet diese Analysen auch auf damit verbundene Bewusstseinsakte, wie Wahrnehmungen, Imaginationen und die Intentionalität schlechthin aus. Indem Husserl immer wieder Erkenntnisse früherer Kapitel aufnimmt, schafft er schließlich ein umfassendes Gesamtbild der Struktur von objektiven Erkenntnisleistungen. Mit einer Trennschärfe, die erst in der zweiten Auflage ganz sichtbar wird, scheidet Husserl dabei das Psychologische aus der Analyse des Bewusstseins-überhaupt aus und argumentiert für die Notwendigkeit einer Einsicht in Wesenszusammenhänge, welche die von uns immer schon beanspruchte Objektivität der Erkenntnis überhaupt erst möglich macht. Der genaue Blick auf die Strukturen, die dem Bewusstsein beim Urteilen und Erkennen zugrundeliegen, die Unterscheidung der verschiedenen Arten von Gegenständen und intentionalen Bezugnahmen, überhaupt das Insistieren auf der Vielschichtigkeit unserer Bewusstseinsphänomene machen den bis heute bei weitem nicht ausgeschöpften Reichtum der Untersuchungen aus. Das Werk profitiert immer noch davon, dass



Verena Mayer

sein Verfasser sich nicht von den populären Dichotomien seiner Zeit – hier Psychologismus, dort Logizismus – vereinnahmen ließ, sondern konsequent seinen dritten Weg verfolgte, der in der systematischen Rückbindung des Objektiven an das urteilende und erkennende Subjekt besteht. Diese Rückbindung kann heute als die Haupttugend der Logischen Untersuchungen gelten. In ihren Grundstrukturen hat Husserl die hier formulierten Erkenntnisse mit wenigen Ausnahmen beibehalten und oft stillschweigend in der späteren Entwicklung seiner Gedanken vorausgesetzt. So legt er etwa mit der Analyse intentionaler Akte bereits die Spur für das spätere Konzept eines tranzendentalen Subjekts, auch wenn er in den Untersuchungen noch schreibt, er könne das „primitive [reine] Ich als notwendiges Beziehungszentrum [intentionaler Akte] schlechterdings nicht […] finden.“ (Hua XIX/1, 374) Die Unterscheidung zwischen abhängigen und selbstständigen Teilen, die er von Brentano übernimmt und in der dritten Untersuchung ausbaut, wird später kaum noch thematisiert, ist aber in der durch das ganze spätere Werk wiederkehrenden Rede von Konstitution, Fundierung und Moment ständig präsent. Eine genaue Kenntnis der Logischen Untersuchungen ist deshalb für das Verständnis der übrigen Schriften Husserls unabdingbar, sie verlangt aber auch vom modernen Leser nicht wenig Anstrengung. Denn es ist nicht nur so, dass Husserl eine eigene Terminologie mit manchmal formalem Exaktheitsanspruch entwickelt, die er zumeist (aber doch nicht immer) konsequent über das Werk hinweg einsetzt. Wer etwa den spezifischen Gebrauch, den Husserl von dem Wort „Vorstellung“ macht, nicht beachtet, wird leicht in interpretatorische Untiefen geraten. Darüber hinaus aber verlangt der systematische Blick auf die essentiellen Strukturen von Bewusstseinsakten eine ganz eigenartige philosophische Einstellung, die Husserl später mit dem Terminus epoché gekennzeichnet hat. Husserl argumentiert in der Regel nicht, sondern beschreibt, was er vorfindet, er analysiert, klassifiziert und rekonstruiert die elementaren Einheiten des Bewustseins, die er Akte nennt, und die weit mehr umfassen als die Urteile, Wahrnehmungen und Empfindungen, von denen die klassische Erkenntnistheorie ausgeht. Husserl betätigt sich so gesehen am Bewusstsein etwa wie ein Botaniker an der Pflanzenwelt. Er ordnet und zeigt Unterschiede und Gemeinsamkeiten auf, er weist philosophische Behauptungen, die nicht durch Bewusstseinstatsachen evident begründet sind, zurück, und überhaupt interessieren ihn metaphysische Annahmen, etwa über die Struktur von Wahrnehmungen oder Urteilen, ebenso wenig wie entsprechende empirisch-psychologische Untersuchungen. Nicht umsonst steht der phänomenologische Wahlspruch „zu den Sachen

Einleitung



selbst“ schon im Vorwort der Untersuchungen (Hua XVIII, VI): also weg von theoretischen Annahmen und hin zu dem, was in der Analyse unserer Bewusstseinsakte unmittelbar ersichtlich wird. Diese Beschreibung der Logischen Untersuchungen soll nicht nahelegen, dass das Werk ein bloßes Sammelsurium von Beobachtungen und Analysen ohne inneren Zusammenhang darstelle. Tatsächlich ist es oft so gelesen worden. Schon zeitgenössische Rezipienten haben etwa das Fehlen einer inneren Verbindung zwischen dem ersten und dem zweiten Band beklagt und den zweiten als einen Rückfall in den vom ersten zurückgewiesenen Psychologismus betrachtet. Husserl, der diese Kritik als schmerzhaftes Missverständnis betrachtete, spricht in einem Zusatz zur Einleitung von einer „natürlichen Reihenfolge“ der Themen, macht aber gleichzeitig darauf aufmerksam, dass die Untersuchung sich „gleichsam im Zickzack“ bewege, da sie immer wieder Unklarheiten beseitigen müsse, bevor sie auf ihrem Weg fortschreiten könne. Ausgangspunkt ist dabei die Zurückweisung des Psychologismus in der Logik und Mathematik des 19. Jahrhunderts, die Husserl in den Prolegomena vornimmt. Die scharfsinnigen und überaus einflussreichen Argumente dieses ersten Bandes der Logischen Untersuchungen zeigen, dass die Logik nicht in psychischen Ereignissen oder Gesetzen und ebensowenig in kulturellen Faktoren begründet sein kann. Husserl bleibt aber in diesem Band eine eingehende eigene Begründung der Logik schuldig. Die Erkenntnis, dass die Logik ideale Gesetze darstellt, die jeder Wissenschaft vorhergehen, läuft jedenfalls nicht auf die Fregesche Behauptung hinaus, dass solche Gesetze in einem „dritten Reich“ der Gedanken lokalisiert seien, von wo sie auf rätselhafte Weise ins faktische Denken hinüberwanderten. Vielmehr fragt Husserl am Ende der Prolegomena explizit, was nun für den Philosophen zu tun bleibe, da gezeigt wurde, dass Logik und Mathematik grundlegender als jede empirische Wissenschaft sind, und beantwortet diese Frage so: „Dem Philosophen ist es nicht genug, daß wir uns in der Welt zurechtfinden, daß wir Gesetze als Formeln haben, nach denen wir den künftigen Verlauf der Dinge voraussagen, den vergangenen rekonstruieren können; sondern was das Wesen von ‚Ding‘, ‚Vorgang‘, ‚Ursache‘, ‚Wirkung‘, ‚Raum‘, ‚Zeit‘ u.dgl. ist, will er zur Klarheit bringen; und weiter, was dieses Wesen für wunderbare Affinität zu dem Wesen des Denkens hat, daß es gedacht, des Erkennens, daß es erkannt, der Bedeutungen, daß es bedeutet sein kann usf.“ (Hua XVIII, 254) Es geht dem Philosophen mit anderen Worten um das, was Husserl das „Korrelationsapriori“ nennt: die notwendige Entsprechung und Beziehung zwischen den Akten des Bewusstseins und seinen Gegenständen, und damit auch um die Frage, wie sich die allem wissenschaft-



Verena Mayer

lichen Denken grundliegenden logisch-mathematischen Strukturen im Bewusstsein konstituieren. Diesen Fragen geht der zweite Band der Logischen Untersuchungen nach. Für den Leser des ersten Bandes ist dabei nicht ohne weiteres ersichtlich, weshalb die folgenden Untersuchungen mit einer Analyse der Bedeutung einsetzen, oder was die vielfältigen Themen, die darauf folgen, die etwa über logische Grammatik, Teil-Ganzes-Beziehungen, Intentionalität oder Evidenz handeln., mit der zuvor erwiesenen Idealität von Logik und Mathematik zu tun haben sollen. In den Ideen zu einer einen Phänomenologie aus dem Jahr 1913, die sich in vieler Hinsicht als weiterführender Kommentar zu den Logischen Untersuchungen lesen lassen, schildert Husserl den Zusammenhang kurz und gedrängt so: „Da jede Wissenschaft nach ihrem theoretischen Gehalt, nach allem, was in ihr ‚Lehre‘ ist (Lehrsatz, Beweis, Theorie), sich im spezifisch ‚logischen‘ Medium, in dem des Ausdrucks objektiviert, so sind die Probleme von Ausdruck und Bedeutung für den von allgemein logischen Interessen geleiteten Philosophen und Psychologen die Nächsten, und sie sind dann auch die ersten, welche überhaupt, sobald man ihnen ernstlich auf den Grund zu kommen sucht, zu phänomenologischen Wesenserforschungen drängen. Man wird von da aus zu den Fragen geführt, wie das ‚Ausdrücken‘ von ‚Ausgedrücktem‘ zu verstehen sei, wie ausdrückliche Erlebnisse zu nicht ausdrücklichen stehen, und was die letzteren im hinzutretenden Ausdrücken erfahren: man wird sich auf deren ‚Intentionalität‘ verwiesen finden, auf den ihnen ‚immanenten Sinn‘, auf ‚Materie‘ und ‚Qualität‘ […].“ (Hua III/1, 258) Während also Husserl in den Prolegomena den Begriff der reinen Logik, letztlich als eine formale Wissenschaftstheorie, entwickelt (siehe dazu in diesem Band den Beitrag von Richard Tieszen) und diesen gegen die psychologistische Vereinnahmung verteidigt (Robert Hanna), untersucht er im zweiten Band die apriorischen Beziehungen der reinen Logik zum Bewusstsein, insbesondere zum sprachlichen Ausdruck, zum Denken und schließlich zum Erkennen. Seinem Motto „zu den Sachen selbst!“ folgend, rekurriert er dabei nicht auf theoretische Konstrukte und überlieferte Begriffsbestimmungen, sondern erarbeitet sich jeden Aspekt des Bewusstseins, der dabei auf die eine oder andere Weise ins Spiel kommt, deskriptiv und von Grund auf neu. Die Grundlegung der reinen Logik gerät damit zu einer umfassenden Strukturbeschreibung des Bewusstseins und seiner Akte überhaupt.

Einleitung



Es ist einleuchtend, dass Husserl, wenn er die „Affinität“ der Logik zum Bewusstsein klären will, von den sprachlichen Ausdrücken ausgeht, und also zunächst eine Beschreibung der Beziehungen von Ausdruck und Bedeutung, und damit eine Semantik, vorlegt. (Vittorio De Palma) Die Untersuchung der Relation zwischen Ausdruck und Ausgedrücktem führt ihn dabei sogleich in das Problem der idealen Einheit der Bedeutung gegenüber der Mannigfaltigkeit der Bedeutungsakte, das unter dem Titel der okkasionellen oder indexikalischen Ausdrücke auch die analytische Philosophie des 20. Jahrhunderts vielfach beschäftigt hat. Für die reine Logik der Prolegomena und ihre Psychologismuskritik ist dabei wesentlich, dass die Bedeutungen – die „logischen“ Inhalte der Bedeutungsakte – von den psychologischen Erlebnissen scharf geschieden werden. (Gianfranco Soldati) Da Husserl hier die von ihm selbst später zurückgewiesene Auffassung vertritt, die Bedeutung eines Ausdrucks sei die „Spezies“ der entsprechenden Bedeutungsakte und werde durch einen Prozess der Abstraktion gewonnen, folgt in „natürlicher Reihenfolge“ die II. Untersuchung über „die ideale Einheit der Spezies und die neueren Abstraktionstheorien“. (Peter Simons, Arthur Smith) Die III. Untersuchung nimmt nun den Begriff der Abstraktion zum Anlass, den Unterschied zwischen „abstrakten“ und „konkreten“ Inhalten von Akten zu analysieren. (John Drummond) Genauer geht Husserl nun von der weitgehend historisch-philosophischen Erörterung der Begriffe in der II. Untersuchung zu eigentlich formalen Unterschieden zwischen Ganzen und Teilen über. Die von Brentano übernommene „Mereologie“ erlaubt ihm dabei eine gegenüber der Tradition durchaus neue Definition der Termini „abstrakt“ und „konkret“, welche dann die Basis für eine phänomenologische Theorie der sog. „ideierenden Abstraktion“ und die formale Grundlage für viele weitere phänomenologische Deskriptionen bildet. Erst in der VI. Untersuchung über die Idee der reinen Grammatik (Jocelyn Benoist) kehrt Husserl dann wieder ausführlich zur Ebene der Ausdrücke und ihrer Bedeutungen zurück, und zwar indem er die zuvor entwickelte Unterscheidung der selbstständigen und der unselbstständigen Teile auf die sprachlichen Zeichen anwendet. Keineswegs hat er hier die eigentliche und ursprüngliche Aufgabe, die Grundlegung der reinen Logik, aus den Augen verloren. Vielmehr scheidet die reine Grammatik Sinn von Unsinn und gibt damit „der reinen Logik die möglichen Bedeutungsformen vor“, auch wenn sie noch nicht die eigentlichen logischen Gesetze formuliert (Hua XVIII, 295). Die V. Untersuchung erst widmet sich direkt der Frage, die als der eigentliche Zweck des zweiten Bandes bezeichnet worden war: wie wir



Verena Mayer

uns nämlich in Bedeutungs- und anderen Akten auf die Gegenstände dieser Akte beziehen, und in welcher Weise die Inhalte der Akte mit ihren Gegenständen in Beziehung stehen. Das übergreifende Thema ist der Begriff der Intentionalität (Dan Zahavi), den Husserl von Brentano übernimmt, aber in seiner eigenen Weise ausformt. Intentionalität stellt den entscheidenden Bezug zwischen „deskriptiver Psychologie“ und Logik her, der in den Prolegomena noch eher ein Rätsel geblieben war; die Beschreibung ihrer Struktur und ihrer Dynamik erklärt uns, wie wir zu den Sachen selbst kommen. Akte, in denen wir dies versuchen, nennt Husserl nun objektivierende Akte, und widmet der Abgrenzung und Beschreibung ihrer Struktur einen großen Teil der V. Untersuchung. (Verena Mayer/Christopher Erhard) Die VI. Untersuchung handelt dann von der Frage, wie wir die Sachen nicht nur „treffen“, sondern wie wir zu Erkenntnis von ihnen gelangen können, und wodurch sich solche Erkenntnisakte auszeichnen. Der bis dahin oft in Anspruch genommene Begriff der Evidenz findet hier seine Aufklärung. (Rudolf Bernet) Aber da das Ziel der Untersuchungen die Frage der Erkenntnis der reinen Logik ist, kann es nicht nur um die Erkenntnis von empirischen Tatsachen gehen. Vielmehr muss Husserl nun auch klären, wie wir zu wahrer Einsicht in logische Gesetze und Grundbegriffe gelangen können. Die VI. Untersuchung enthält daher auch eine Analyse der Struktur der kategorialen Erkenntnis. (Dieter Lohmar) Damit ist das Projekt, das in den Prolegomena umrissen worden war, in groben Zügen vollendet, allerdings, wie Husserl bald bemerkt, mit vielen Unklarheiten und Sprüngen. Schon bald macht er sich an eine Umarbeitung, von der die Bände XX/1 und XX/2 der Husserliana beredtes Zeugnis geben. Dabei und währenddessen aber wandeln sich seine Auffassung in manchen Hinsichten so entscheidend, dass er schließlich das Projekt der Umarbeitung fallen lässt. Nur wenig korrigierte Neuauflagen erscheinen, die allerdings an vielen Stellen sowohl die Kontinuität, als auch die Differenz deutlich erkennen lassen. Seine Umgestaltungen betrachtete Husserl dabei niemals als einen Wechsel in der „Theorie“, vielmehr waren es stets die Sachen selbst, die ihn bei genauerer Analyse zu Änderungen zwangen. Wie die Kommentare dieses Bandes zeigen, lässt sich diese Dynamik vielfach in den Logischen Untersuchungen selbst schon beobachten. Überhaupt ist die Phänomenologie, wie sie sich in den Logischen Untersuchungen erstmals en détail darstellt, ein offenes Projekt, eine, wie Husserl einmal in den Ideen I schreibt, „anfangende Wissenschaft“, die zur Weiterentwicklung, Präzisierung und Umgestaltung einlädt. Als Motto dieses Bandes, der zur lebendigen Weiterführung dessen anregen soll, was man eine analytische

Einleitung



Phänomenologie nennen könnte, mag daher das folgende Bekenntnis aus den Ideen I dienen, wo Husserl schreibt: „Unser Verfahren ist das eines Forschungsreisenden in einem unbekannten Weltteile, der sorgsam beschreibt, was sich ihm auf seinen ungebahnten Wegen, die nicht immer die kürzesten sein werden, darbietet. Ihn darf das sichere Bewußtsein erfüllen, zur Aussage zu bringen, was nach Zeit und Umständen ausgesagt werden mußte und was, weil es treuer Ausdruck von Gesehenem ist, immerfort seinen Wert behält – wenn auch neue Forschungen neue Beschreibungen mit vielfachen Besserungen erfordern werden. In gleicher Gesinnung wollen wir im weiteren getreue Darsteller der phänomenologischen Gestaltungen sein und uns im übrigen den Habitus innerer Freiheit auch gegen unsere Beschreibungen wahren.“ (Hua III/1, 201)

2 Richard Tieszen

Husserl’s concept of Pure Logic (Prolegomena, §§ 1–16, 62–72)

Husserl’s conception of logic in the Logische Untersuchungen is quite different from the view of logic he held in his first book, Philosophie der Arithmetik (PA). The Logische Untersuchungen open with the Prolegomena to Pure Logic which contain an extended critique of the view that logic and mathematics have their foundations in psychology. Underwriting this critique of psychologism is Husserl’s sharp new distinction between real and ideal objects and truths. The comments on ideal objects and ideal truths in the Logische Untersuchungen suggest a newfound platonism about logic and mathematics. In the text of the Prolegomena Husserl credits a number of figures with helping him to arrive at his new conception of pure logic. He mentions in particular Kant, Herbart, Lotze, Leibniz, Lange and Bolzano. Leibniz and Bolzano are held in high regard and it is Bolzano and Lotze who are especially praised as the anti-psychologistic logicians. Lotze is given credit for showing how a form of platonism about logic can be defended. In his characterization of the positive tasks of pure logic, as we will see below, Husserl also points to ideas in the work of Riemann, Grassman, Lie, Cantor, and some other figures.

1. Pure Logic as the Science of All Possible Sciences At the outset of the Prolegomena Husserl asks whether logic is a theoretical or practical discipline, whether it is a formal discipline or not, whether is has the character of an a priori, demonstrative discipline or of an empirical, inductive discipline, and whether it is independent of the other sciences, especially of psychology and metaphysics (Hua XVIII, § 3). Husserl’s goal

10

Richard Tieszen

is to demarcate the domain of pure logic as an independent science. Logic is endangered, he says, through confusions with other sciences, especially psychology and some other empirical sciences. In the Prolegomena he will seek to sharpen the conception of pure logic as an a priori, theoretical discipline that is formal and demonstrative in nature. Several different conceptions of logic are considered along the way – logic as algorithmic, as normative, and as technical practice – but it is argued that they fail to do justice to pure logic as theoretical science. In Logische Untersuchungen we are presented with a conception of logic that is quite different from the one in his earlier work: pure theoretical logic is now said to be concerned with ‚ideal‘ meanings. Moreover, it is also concerned with the ontological correlates of ideal meanings, ideal objects and ideal states of affairs. In Logische Untersuchungen pure mathematics is taken to be part of pure logic in this broad sense. Logic is described in the Prolegomena as the „science of all possible sciences“ or as the „theory of science“. Husserl says (in § 12) that excellent thoughts towards the circumscription of logic are to be found in Bolzano’s Wissenschaftslehre. The essence of science consists of the unity of knowledge in a whole system of grounded validations. Husserl examines at some length the nature of the unity of science, i. e., the interconnections of things and the interconnections of truths in science (§§ 62–64). The realm of truth is not a disordered chaos but is dominated and unified by law. The investigation and presentation of truths must therefore also be systematic. Connections of validation are governed by reason and order, by regulative laws, not by caprice or chance (§§ 6–8). In mathematics, for example, we can find many examples of reasoning about different kinds of objects (e. g., triangles, numbers) that have the form „Every A is B, a is an A, and therefore a is a B“. Here we are carrying out a valid piece of reasoning that is precisely not a function of chance or caprice. Arguments that take us validly from given pieces of knowledge to new knowledge always have a form that applies to countlessly many examples. Validating procedures do not stand in isolation. With them a definite type is always brought out. Forms of reasoning of the kind just cited are free of all essential relation to some limited field of knowledge. Logic is what makes possible the existence of sciences. All testing, invention and discovery rests on regularities of form. It is the wide degree of independence of form from a field of knowledge that makes possible a „theory of science“. Were there no such independence there would be only coordinated logics corresponding separately to the different sciences. There would not be a general logic. Husserl says that in fact both are needed. There should be investigations into the theory of science as it

Husserl’s concept of PURE LOGIC

11

concerns all sciences and, supplementary to these, particular investigations concerning the theory and method of the separate specific sciences. Some of the methods used in science are validation procedures but some are simply auxiliary devices for validation procedures (§ 9). The theory of science needs to take both of these into account. In passages that are interesting to read in connection with PA, Husserl says that some scientific procedures are abbreviations or substitutions for validating arguments and are used to economize thought. These are methods or procedures that originally received their sense and value from such validations but are now used without cognitive insight. Among these auxiliary devices Husserl includes algorithmic methods. Their function is „to save as as much genuine deductive mental work as is possible by artificially arranged mechanical operations on sensible signs.“ They may be executed blindly. Husserl says that whatever marvels these methods may achieve, their sense and justification depends on thought that can be validated. All „mechanical methods“, including those of calculating machines, are included among these auxiliary methods. Logic, as theory of science, is in a sense a normative discipline. It seeks that which pertains to genuine, valid science as such, so as to use this Idea of science to measure whether the empirically given sciences are in agreement with this Idea, to what degree they approach it, and where they offend against it. With this norm as its end, logic readily gives rise to a technology. Logic will have many practical uses in this role. Husserl asks whether the definition of logic as a technology, as an applied or practical logic, really captures the essential character of logic. His answer is that it does not. Furthermore, even if logic has a normative function the logical laws are not themselves normative prescriptions. They do not, as part of their content, tell us how one should judge. They can be employed for normative purposes but they are not therefore norms. Anyone who judges both that every A is a B and every B is a C ought to also judge that every A is a C. What we are told in logic itself, however, is only that if every A is a B and every B is a C then it is also true that every A is a C. There are no normative terms in the latter case (§ 16). A different thought-content is involved. Purely theoretical statements admit of normative transformations but that does not make them normative statements. Logic has a normative function but every normative discipline presupposes one or more theoretical disciplines as its foundation. In the sense just suggested, it must have a theoretical content that is not itself normative. Every normative discipline demands that we know certain non-normative truths and the latter are taken from certain theoretical sciences. We

12

Richard Tieszen

are therefore led to the question: which theoretical sciences provide the essential foundations of the theory of science? In particular, does logic have its place in sciences that have already been marked off and independently developed?

2. Pure Logic, Psychologism, Empiricism, and Relativism One of the central answers to these questions in Husserl’s time was that it is psychology that provides the foundations of logic. The philosophers Mill, Lipps, Sigwart, Wundt, and others had made such claims. In thinking of logic as normative, psychologistic thinkers point out that what is regulated by logic is the mental activities or products of those who reason. What logic talks about are concepts, judgments, deductions, and so on. These are all taken to be psychological activities or products. Husserl claims that psychologistic arguments show only that normative logic may be helped by psychology, not that psychology provides the essential foundation of normative logic. The possibility remains open that „pure logic“ is the foundation of normative logic. Earlier thinkers may not have succeeded in making clear what pure logic is but this should give us all the more reason to seek such clarification. Husserl proceeds to critically examine the view that psychology is the foundation of logic. The critique of psychologism is similar in a number of ways to Frege’s (Frege 1893 and 1894), except that Husserl goes into much more detail. As the critique unfolds we learn about many of the features of pure logic. Husserl’s arguments continue to be of interest today since many of the same issues are still in play in efforts to naturalize logic and mathematics. Husserl starts by pointing out that psychology is supposed to be a factual or empirical science (§ 21). It has so far lacked genuine and exact laws. The propositions of psychology are merely vague generalizations from experience. They are propositions about approximate regularities. If this is the case then there are serious consequences for the psychologistic logicians. If psychological laws lack exactness then the same must be true for the laws of logic. Laws of logic and mathematics, however, are exact. Even if one had exact natural laws in psychology it is still true that natural laws are not a priori. They are instead established by induction from singular facts of sense experience. They are probabilities. Thus, laws of logic would have to be probabilities. But this seems patently false. Laws of logic have an a priori validity. Husserl says that we know about basic a priori laws on

Husserl’s concept of PURE LOGIC

13

the basis of direct insight. Laws of logic are not causal laws. Psychologistic logicians confuse the contents of judgment with judgments as psychological processes or entities. The latter are real events having causes and effects. A law of logic, however, ought not to be confused with the judging or with knowledge of the law. The ideal ought not to be confused with the real. There is a fundamental and essential gulf between ideal and real laws, between normative and causal regulation, logical and real necessity, logical and real grounds. There is no conceivable gradation that could mediate between the ideal and the real. Another consequence of psychologism is that logical laws must themselves be psychological in content. This, according to Husserl, is palpably false. No logical law implies a matter of fact. Laws of logic presuppose nothing mental. They presuppose no facts of psychic life. They do so no more than the laws of pure mathematics do. One should not confuse the psychological „presuppositions“ and „bases“ of the knowledge of a law with the logical presuppositions, the grounds and premises, of that law. Psychological dependence, or dependence of origin, is distinct from logical demonstration and justification. In comments that indicate his newfound platonism, Husserl says that the truth of laws of logic is raised above time. One cannot attach temporal being to it. It does not arise or perish (§ 24). Husserl considers Mill’s psychologistic account of some particular laws of logic, e. g., the law of contradiction. For Mill, this law states nothing more than the real incompatibility of two acts of judgment. Mill’s interpretation, Husserl argues, yields a wholly vague, scientifically unproven empirical proposition, not a law of logic. What the law of contradiction is about is the ideal impossibility that the two propositions could both be true. Husserl is led to a broader consideration of basic errors of empiricism by his examination of psychologism (§ 26). He says that extreme empiricism is as absurd as extreme skepticism. It destroys the possibility of the rational justification of mediate knowledge (in the form of deductive inference) and so destroys its own possibility as a scientifically proven theory. Since it puts full trust only in the singular judgments of sense experience it abandons all hope of rationally justifying mediate knowledge. It will not acknowledge as immediate insights and as given truths the ultimate principles upon which the justification of mediate knowledge depends. Instead, it tries to derive them from sense experience and induction. Empiricism appeals to a naive, uncritical everyday experience to found logical laws, instead of to immediately evident universal principles. Husserl considers Hume to be a moderate empiricist since he distinguished matters of fact from relations of ideas and surrendered only the former to sense experi-

14

Richard Tieszen

ence. Nonetheless, Hume’s position is also untenable. For Hume, mediate judgments of fact never permit of rational justification but only of psychological explanation. This must apply to Hume’s theory itself. Husserl says that our capacity to ideate universals in singulars, to „see“ a concept in an empirical presentation and to be assured of the identity of our conceptual intentions in repeated presentations, is presupposed by the possibility of knowledge. Just as we can intuit one concept in an act of ideation as the single species whose unity against various instances is given with insight, so can we apprehend with evidence the logical laws relating to concepts (see also Investigations II and VI). These concepts are ideal unities. Wherever we can carry out conceptual presentations in this sense we can also apply logical laws. The validity of these laws, however, is absolutely unrestricted. It does not depend on our power nor on anyone’s power to achieve acts of conceptual presentation, nor to sustain or repeat such acts. Investigation II of the Logische Untersuchungen supplements these arguments on the nature of pure logic. It is filled with objections to nominalism, conceptualism, and the empiricist theories of abstraction of Locke, Berkeley, Hume, and Mill. It is argued that psychologism and, more generally, empiricism, is a skeptical relativism. Empiricism about logic undermines itself. One can distinguish individual from specific relativism. In the former case one makes truth relative to each person, while in the latter case one makes it relative to humans as a species. Husserl’s term for species relativism is „anthropologism“. Husserl of course argues that both forms of relativism about logic are absurd. Sigwart in particular is singled out for his anthropologism. Sigwart resolves truth into conscious experiences. Experiences are real particulars, temporally determinate, that come into being and pass away. According to Husserl, however, truth is an „Idea“ that is beyond time. It makes no sense to give truth a date in time nor a duration that extends through time. Truth can of course be apprehended or grasped but this is not like apprehending some empirical content that comes into being and then vanishes at some later stage. The experience of truth is experience of a universal, an Idea. Other beings could not have logical principles different from ours. To think that this is a possibility is to confuse psychological or anthropological possibility with logical possibility. All change affects sensory individuals. It makes no sense in regard to concepts. Real possibilities involve sensory individuals but ideal possibilities do not. If we think of truth as ideal in this sense then the empirical sciences as a whole only approximate truth, just as real objects only approximate ideal objects. These kinds of remarks, which occur throughout the Prolegomena, are the basis for the claim that Husserl has adopted a kind of

Husserl’s concept of PURE LOGIC

15

rationalistic platonism about logic and mathematics in the Logische Untersuchungen. A number of later philosophers and logicians who were inspired by Husserl’s writings, such as Kurt Gödel (Gödel 1961, and Tieszen 2005, Part II), would no doubt have found this rationalistic platonism attractive. Husserl says that we are conscious of truth as we are conscious of a species, e. g., the color red, as an ideal object. A red object may stand before us but this object is not the (ideal) species red. The concrete object here does not contain the species as a psychological or metaphysical part. The non-independent moment of red (as opposed to a piece – see Investigation III) that is given to us is itself something individual, something here and now that arises and vanishes with the concrete whole object. It is similar but is not identical in different red objects. Redness, however, is an ideal unity that does not come into being or pass away. The part (moment) red is not Redness itself but is an instance of redness. Just as universal objects differ from singular ones, so too do our acts in apprehending them. Reference to an individual in consciousness is different from reference to its species, or its Idea (see also Investigations II and VI). In several acts of ideation we come to be aware of the identity of the ideal unities that are meant in our single acts. This is a strict identity. There is awareness of an identical species. Truth is likewise an Idea. We are aware of the unity and identity of truth over against the dispersed multitude of concrete compared cases. Husserl says that the statements „It is the truth that P“ and „There could have been thinking beings having insight into judgments to the effect that P“ are equivalent. If there are not intelligent beings or if they are in a real sense impossible then the ideal possibilities remain without actual fulfillment. The apprehension of truth is simply nowhere realized. Each truth, however, retains its ideal being and remains what it is. It is a case of validity in the timeless realm of Ideas (§ 39). This idea of truth could not be merely relative to the human species. That would be to miss its sense. The relativization of truth presupposes the objective being of the fixed point with respect to which things are relative. In not seeing this, relativism is caught in a contradiction. Logic might seem to be about mental phenomena and processes since it speaks of judgments, proofs, and so on. Husserl says that if we compare logic with mathematics we will see that logic could not be about mental phenomena. In the nineteenth century psychologism had not taken hold in the foundations of mathematics as it had in logic. Psychologism, Husserl notes, would also make mathematics a branch of psychology. Husserl thinks that there is a theoretical unity of logic and mathematics. Like pure mathematics, the territory to be investigated by pure logic is an ideal territory

16

Richard Tieszen

(§ 46). Mathematics, Husserl says, no longer needs to fight for its independent existence. (Husserl would perhaps be surprised at the extent to which philosophers since his time have attempted in various ways to naturalize mathematics.) We would have no numbers without counting, no sums without addition, and so on, and yet no one regards the theories of pure mathematics as parts of psychology. In a section of Logische Untersuchungen that should be read in connection with the earlier view of PA, Husserl says that counting and arithmetical operations as facts, as mental acts proceeding in time, are certainly of concern to psychology since psychology just is the empirical science of mental facts. Arithmetic, however, is different. Numbers, sums, products and so on are not causal acts of counting, adding and multiplying that are carried out here and there. They are not the same as the presentations in which they are given. The number 5 is not my own or any other person’s counting of 5. In the counting of 5 we have 5 as a possible object of acts of presentation whereas the number 5 itself is the ideal species of a form whose concrete instances are found in what becomes objective in certain acts of counting, in the collective whole constituted thereby. The number cannot be regarded as a part or side of a mental experience. Therefore, it is not something real. If we are to conceive of 5 correctly we will first have an articulate, collective presentation of this or that set of five objects. In this act a collection is intuitively given in a certain formal articulation as an instance of the number species in question. On the basis of this intuited individual we perform an abstraction. That is, we not only isolate the non-independent moment of collective form in what is before us but we apprehend the Idea in it. The number 5, as the species of the form, is the reference of this conscious act. We can see how Husserl is now grafting his ontology and epistemology of ideal objects onto his earlier PA account of number. What we are now meaning, he says, is not this individual instance, not the intuited object as a whole, not the form immanent in it but still inseparable from it. What we mean is rather the ideal form-species that is identical in whatever mental act it may be individuated in as an intuitively constituted collection. It is a species that is untouched by the contingency, temporality and transience of our mental acts. Acts of counting arise and pass away. Arithmetical propositions tell us nothing about what is real, neither about the real things counted nor the real acts in which they are counted. The propositions of arithmetic are laws rooted in the ideal essence of the genus Number. The singulars that come within the range of these laws are ideal singulars, the determinate numbers that are the lowest specific differences of the genus number. What has been said here about pure arithmetic likewise carries over at all points to pure logic.

Husserl’s concept of PURE LOGIC

17

Terms such as „judgment“, „concept“, „proof“, and so on are equivocal. On the one hand they stand for mental states that belong to psychology but, on the other hand, they stand for ideal entities. One must always be careful to separate these two meanings. It is yet another prejudice of psychologism to hold that the recognition of the truth of a judgment can be adequately understood in terms of the psychology of inner evidence. Pure laws of logic, however, say nothing about the psychology of inner evidence or its conditions. Psychological possibilities about inner evidence are real possibilities but what is psychologically impossible may very well be ideally possible. There are, for example, decimal numbers with trillions of places and there are truths relating to them. No one can actually imagine such numbers nor do the additions, multiplications and so on relating to them. Inner evidence here is psychologically impossible yet, ideally speaking, it represents a possible state of mind. Moreover, inner evidence is often taken to be a special feeling that attends some judgments. This view must be categorically rejected. The correct view of „inner evidence“ in the case of logic and mathematics is that inner evidence is the experience of truth as ideal. Truth is an Idea whose particular case is an actual experience in an inwardly evident judgment. A judgment that is not self-evident stands to a self-evident judgment much as an arbitrary positing of an object in imagination stands to its adequate perception. A thing adequately perceived is not a thing merely meant in some matter or other. It is a thing given in our act as what we mean. As in the realm of perception, the unseen does not coincide with the non-existent. Lack of inner evidence does not amount to untruth. Evidence is the experience of agreement between meaning and what is itself present, the meant. It is the experience of agreement between the sense of an assertion and the self-given state of affairs. (These ideas are treated in more detail in Investigations I and VI, and also in other works of Husserl.) The Idea of this agreement is truth, whose ideality is also its objectivity. To have evidence in this sense is also to be aware that no other person could have evidence that is at variance with our own. If evidence were merely a matter of subjective or even intersubjective feeling then it would not be possible to escape skepticism about claims to evidence. The problems of psychologism, empiricism, relativism, and naturalism about logic result, generally speaking, from failing to clearly grasp the distinction between the real and the ideal.

18

Richard Tieszen

3. Pure Logic, Economy of Thought, and Evolutionary Biology Yet another empiricist attempt to find a basis for logic and epistemology can be found in the efforts to provide a biological basis for these disciplines (§§ 52–56). Husserl’s discussion of this kind of „biologism“ is especially relevant to recent developments in philosophy and the sciences. Husserl has in mind the work of Avenarius and Mach, who were thinking at a relatively early stage about logic in connection with evolutionary biology. Avenarius’ doctrine of least action and Mach’s doctrine of an economy of thought are examined in the Prolegomena. Here one thinks of science in terms of evolution or adaptation. One conceives of science as the most purposive (economical, power-saving) adaptation of thought to the varied fields of phenomena. In particular, a creature will be better adapted the more rapidly or efficiently it can perform the acts needed for its survival and success. This leads to the notion of an „economy of thought“. Our intellectual powers are severely limited, as Husserl himself already noted in PA. There is a fairly narrow sphere within which complex, abstract notions can be fully understood. A significant effort has to be made to understand complexities of this sort. When these facts are considered it is all the more amazing that the more comprehensive rational theories and sciences should have been developed at all. How could sciences such as mathematics, with their towering structures of thought, be possible? Art and method make it possible to overcome the defects of our cognitive constitution. They permit an indirect achievement by way of symbolic processes from which intuition, all true understanding and inner evidence are absent. One has some sense of security, however, in using such arts and methods to economize thought. There are certain natural processes of thought economy that are then perfected and developed. Once the methods have been developed and justified they can be used without insight. They can be used mechanically. The reduction of insight to mechanism in our thought processes leads to an indirect mastery over the complexities of thought that admit of no direct mastery. Here Husserl gives many examples from mathematics (§ 54). The surrogative, operational concepts that are developed on this basis turn signs into „counters“ and make possible extensive fields of mathematical thought and research. They take these areas down from the exhausting heights of abstraction to comfortable intuitive ways where imagination, guided by insight, can move within the limits of rules, as in regulated games.

Husserl’s concept of PURE LOGIC

19

A vast thought economy is present in recent purely mathematical disciplines. Genuine thought is replaced by surrogative, signitive thinking. This economy leads to formal generalizations of our original trains of thought. In this manner, the horizon of deductive disciplines is greatly enlarged. Out of elementary arithmetic, for example, arises a more generalized form of arithmetic in which numbers and magnitudes no longer count as basic concepts but merely as chance objects of application. Fully conscious reflection now takes place and the pure theory of manifolds (see below) emerges as a further extension. In its form it covers all possible deductive systems. The form-system of formal arithmetic is merely one of its special instances. Husserl thinks all of this contains important insights and ought to be investigated in great detail. Its relation to logic as a practical technology is immediately understandable. It yields an important foundation for such a technology. Here Husserl especially praises the work of Mach. We need to keep in mind, however, that Avenarius and Mach base their ideas about an economy of thought on certain biological facts. Ultimately, we are dealing with a branch of the theory of evolution. For just this reason, these thinkers are able to throw light on practical epistemology and the methodology of scientific research but not at all on pure epistemology and the ideal laws of pure logic. Husserl says that all of the arguments against psychologism and relativism mentioned above can be brought to bear on the effort to found logic and pure epistemology on a biological economy of thought. The principle of economy can be thought of either as something factually given or as logically ideal. People like Mach and Avenarius tacitly substitute the former for the latter. We see that it is a supreme goal of science to arrange facts under laws that are as general as possible and in this manner bring them together with the maximum possible rationality. This maximization is an ideal of a pervasive, all-embracing rationality. The „basic laws“ would be laws of supreme coverage and efficacy, whose knowledge yields the maximum of insight and explanation in some field. The axioms of elementary geometry are examples of this. If we idealize this, we have the notion that there are no limits to our power to deduce and subsume. The goal or principle of maximal rationality in this sense is the supreme goal of the rational sciences. It is self-evident that it would be better for us to know laws more general than those that we already possess at a given time. Such laws would lead us back to grounds that are deeper and more embracing. This principle, however, is no mere empirical, biological principle. It is a purely ideal principle, and an eminently normative one. To identify the movement toward maximum possible rationality with a drift

20

Richard Tieszen

towards biological adaptation, or to derive the former from the latter, amounts to confusion. It parallels the psychologistic misreadings of laws of logic and their misconception as laws of nature. The ideal movement of logical thinking is towards maximal rationality. The thought economist who engages in biologism turns this into a real, empirical drift of human thought, bases it on a vague principle of power-saving, and ultimately on adaptation. One is certainly justified in speaking of an economy of thought but only in that one compares one’s actual thought with an ideal norm. The ideal validity of this norm is presupposed by all talk of an economy of thinking. It is therefore not a possible explanatory outcome of a theory of such economy. We measure our empirical thinking against our ideal thinking and we then say that the former to some extent runs as if guided by the latter. Before all economizing of thought we must already know our ideal. We must know what science ideally aims at. Pure logic is prior to biological thought economy. It is absurd to base the former on the latter.

4. Pure Logic When Husserl turns toward his own positive view of pure logic (§§ 62– 72) he refers, as we noted above, to the views of Herbart, Lotze, Leibniz, Lange and Bolzano. His view emerges out of an examination of the work of these thinkers but cannot be wholly identified with any one of them. His suggestions about the nature of pure logic begin with reflections on what constitutes the unity of science. Given Husserl’s very broad conception of logic, this becomes a question about the conditions of the possibility of theory in general. He is of course concerned with ideal conditions, not real conditions of knowledge. Truths of science are what they are whether we have insight into them or not. Since they do not hold insofar as we have insight into them, but we can only have insight into them insofar as they hold, they must be regarded as objective or ideal conditions for the possibility of our knowledge of them. Logical justification of a given theory, i. e., justification in virtue of its pure form, demands that we go back to the essence of its form, to the concepts and laws that are ideal constituents of theory in general that regulate in an a priori, deductive fashion all specializations of the Idea of theory in its possible kinds. Thus, we are dealing with an a priori, theoretical, nomological science that concerns the ideal essence of science as such. As noted earlier, we are concerned with the theory of theory, or the science of the sciences. Echoing Leibniz and

Husserl’s concept of PURE LOGIC

21

Bolzano, Husserl says that pure logic in this broad sense will even include the pure theory of probability. Husserl describes three tasks that should be assigned to pure logic in this sense. The description of these tasks in the Prolegomena is not nearly as clear as it could be. It is possible, however, to form a clearer conception of the tasks by consulting Husserl’s Formale und transzendentale Logik (Hua XVII, FTL) and Einleitung in die Logik und Erkenntnistheorie (Hua XXIV, see also Tieszen 2004). My comments below will be informed, in part, by these sources. The first task of pure logic is to lay down and clarify the primitive concepts that make possible the interconnected web of theory. Husserl divides the primitive concepts into two groups: concepts that involve categories of meaning and concepts that involve objective categories. Concepts that involve the meaning of expressions are, among others, concept, proposition, and truth. The elementary connective forms of logic are also involved here: conjunction, disjunction, conditional linkage of propositions, and so on. The correlative ontological concepts such as object, state of affairs, unity, number, relation, and connection fall on the side of the pure formal objective categories. Husserl contributes to the development of this first task of pure logic in Investigation IV of the Logische Untersuchungen, which is on the idea of pure grammar. In Investigation III on parts and wholes Husserl distinguished independent from non-independent parts, and in Investigation IV he applies this distinction to ideal meanings. The distinction between independent and non-independent meanings is at the foundation of essential categories of meaning on which many a priori laws of meaning rest. These laws abstract from the objective validity or truth of propositions. They precede such matters. They are laws of grammar that provide pure logic with possible meaning-forms. These are the a priori forms of complex meanings that are significant as wholes. Such laws guard against senselessness (Unsinn), which occurs when we combine meaningful parts to form a whole that is not meaningful, as in „a round or“. Laws that guard against senselessness are to be distinguished from laws that guard against formal or analytic nonsense (Widersinn), i. e., absurdity or formal contradiction. In the latter case we have meaningful wholes that are nonetheless contradictory, such as „a round square“. The laws of grammar merely tell us what is required in the case of complex meanings if we are to have a significant semantic unity. As such, they are a priori patterns in which meanings belonging to different semantic categories can be united to form one meaning instead of producing chaos. Husserl’s view of grammar includes not only formal syntax but also

22

Richard Tieszen

categorial grammar. Within pure logic, we need to distinguish the pure theory of semantic forms from the pure theory of validity which presupposes it. Building on these ideas, Husserl wishes to promote the old idea, conceived by the rationalists in the seventeenth and eighteenth centuries, of an a priori or universal grammar. As is to be expected, he contrasts the idea of universal grammar with the idea of founding grammar on psychology or other empirical sciences. The second task of pure logic is to find the laws concerning (i) the possible truth or falsity of meanings (propositions) as such, purely on the basis of their categorial formal structure, and (ii) the laws concerning the objective correlates of propositional meanings, that is, of the being and non-being of objects as such and of states of affairs as such, on the basis of their pure categorial form. The laws in the one case concern meanings and in the other case concern objects as such. On the side of meaning we have, for example, theories of inference and laws that guard against formal or analytic nonsense (Widersinn), i. e., absurdity or formal contradiction. On the side of the objective correlates, of formal ontology, we have, for example, the pure theory of multiplicities or the pure theory of numbers. We should try to find the laws, which in their formal universality, span all possible meanings and objects, under which every particular theory or science is ranged, and which it must obey if it is to be valid. Husserl says that not every theory presupposes every such law as the ground of its possibility and validity. Rather, the ideal completeness of the theories and laws in question will yield a comprehensive fund from which each particular valid theory derives the ideal grounds of essential being appropriate to its form. In FTL the grammar of pure logic is included in what Husserl calls the first level of formal logic. The first level is concerned, as is the first task of logic in Logische Untersuchungen, with the mere possibility of judgments as judgments, without inquiring into whether they are true of false or even whether they are merely consistent or contradictory. Husserl’s second task of pure logic in Logische Untersuchungen is in some ways a precursor of what is described as the second level of logic in FTL The second level of logic in FTL is concerned with the possible forms of true judgments. Husserl calls this „consequence-logic“ (Konsequenzlogik) or the „logic of non-contradiction“ Here we are to focus on the question of whether a given form is included in or excluded by the forms of judgments in a premise set. In the former case we have an analytic consequence relation while in the later case we have an analytic anti-consequence, or an analytic contradiction. Husserl says that this second level of logic concerns only the (non)-contradictoriness of judgments. It is not yet concerned with the truth of judgments. Judgments

Husserl’s concept of PURE LOGIC

23

may be formally consistent with one another or not. If not, then they have no possibility of all being true, and this is purely a matter of form. Laws at this level would guard against Widersinn. Non-contradiction is a condition for possible truth. The ontological correlate of a consistent formal theory is a formal ontology or „manifold“. In FTL Husserl goes on to distinguish a third level of logic, which he calls truth-logic (Wahrheitslogik). This is an inquiry into the formal laws of possible truth. The idea in FTL is that grammaticality is a condition for the possibility of consistency of judgments, and consistency of judgments is a condition for the possibility of truth of judgments. Some additional ideas about the second and third levels of logic in FTL are anticipated to some extent in Logische Untersuchungen in Investigation VI (see, e. g., Chps. 4 and 8). Husserl says in Logische Untersuchungen that with the completion of the first two tasks of pure logic we will have done justice to the Idea of a science of the conditions of the possibility of a theory in general. Once we understand these conditions, however, we might ask whether, at the highest level of abstraction, there could be a theory that covers all possible deductive theories. What is needed, according to Husserl’s third task of pure logic, is a theory of the possible forms of theories or a „pure theory of manifolds“. The forms of theories are not mutually unrelated. Husserl says that there will be a definite, ordered procedure which will enable us to construct the possible forms of theories, to survey their legal connections, and to pass from one to the other by varying their basic determining factors. There will be universal propositions that will govern the legal connection and the transformation and mutual interchange of these forms, if not for the forms of theory generally then at least for the forms of theory belonging to defined classes. Husserl says that he is not claiming that mathematicians themselves have as yet correctly discerned the ideal essence of such a new discipline or have risen to the height of abstraction of an all-comprehensive theory. Mathematicians, however, have used the term „manifold“ for the objective correlates of possible formal theories. The term „manifold“ covers possible fields of knowledge over which theories of various forms will preside. The objects in a manifold are not determined directly as individual or specific singulars, nor indirectly by way of their material species or genera, but solely by the form of the connections attributed to them. These connections are as little determined in content as are their objects. Only their form is determined through the forms of the laws that are assumed to hold of them. The most general idea of a theory of manifolds is that it is to be a science that works out the form of the essential types of possible theories and investigates their legal relations with one another. All actual

24

Richard Tieszen

theories are then specializations or singularizations of corresponding forms of theory, just as all fields of knowledge are individual manifolds. It is of the highest methodological importance for mathematics to understand the ranging of forms in more comprehensive forms or classes of forms. As examples of what he has in mind, Husserl points to the theories of manifolds that arose from generalizations of geometric theory and its forms and to extensions of the formal theory of real numbers into the formal two-dimensional field of ordinary complex numbers. Regarding geometry, Husserl is referring to the theory of n-dimensional manifolds, whether Euclidean or non-Euclidean. He mentions especially the work of Riemann. He also mentions Grassman’s theory of extensions, Lie’s theory of transformation groups, some work of Helmholtz, and Cantor’s investigations into numbers and manifolds. Husserl says it is senseless to speak of different geometries if „geometry“ names the science of the space of everyday phenomena. If we mean by „space“ the categorial form of worldspace, however, and by „geometry“ the categorial theoretic form of geometry, then we can extend our conceptions of these fields. From this point of view, the theory of a Euclidean manifold of three dimensions is an ultimate ideal singular in a connected range of a priori, purely categorial-theoretic forms (formal deductive systems). Several other examples are mentioned by Husserl (§ 70). Although Husserl does not explicitly refer to Felix Klein in the Prolegomena, Klein’s Erlanger program, inaugurated in 1872, can be used to illustrate the kinds of points Husserl is making here (Tieszen 2005, Chp. 3 ). Klein proposed that we think of different geometries as characterized by groups of transformations. Each geometry investigates the properties left invariant under groups of transformations. What is invariant in one geometry is not necessarily invariant in another. The different geometries can, however, be systematically related to one another to a remarkable extent. Euclidean geometry can be viewed as the study of those properties left invariant under so-called rigid motions: translations, rotations and reflections. Projective geometry is concerned with the smaller class of invariants that are a function of the rigid motions plus projections. Topology is concerned with the still smaller class of invariants that we obtain if our variations are even more radical, including the most extreme stretchings and twistings. Thus, length and angle are Euclidean invariants but they are not invariant under projective variations. Linearity and triangularity are projective invariants but they are not invariant under the more radical topological variations. Connectedness and number of holes (to be more precise, „genus“), for example, are topological invariants. One might say that topological invariants are

Husserl’s concept of PURE LOGIC

25

quite abstract or deep, relative to other geometric invariants. They quite literally result from greater levels of abstraction. We can relate this idea of greater levels of abstraction directly to the types of variations to which one subjects objects. In topology we do not, for example, lose the property of dimensionality of a geometric figure or object, or the property of having a boundary or not, but we do lose things like size and shape as these are understood in Euclidean geometry. Topological equivalence is much more abstract than equivalence in, say, Euclidean geometry. We can thus think of topology as a generalization of Euclidean geometry, and indeed of projective geometry. The group of topological transformations has the group of projective transformations as a subgroup, and therefore has the group of Euclidean transformations as a subgroup. The generalization/specification relations can in fact be made very precise here. We can say exactly what is involved in making a geometry more „general“ or more „specific“ by pointing to groups of transformations (variations). Some variations are more radical than others. We cannot perform them without changing a property that was invariant for a range of variations into a property that is no longer invariant. The idea is then to map all of this out, characterizing in each case the geometry obtained. This classification and organization extends to Euclidean and non-Euclidean geometries. One can start with Euclidean geometry and abstract until one obtains projective geometry or topological invariants or one can specialize from the top down to Euclidean geometry. In the latter case the number of invariant properties increases by specializing the transformations under consideration until we have the invariant properties of Euclidean geometry. As we said, this is all made very precise in modern geometry by considering subgroups of a group of transformations. This is closely related to Husserl’s view that we can investigate different a priori ontologies and their logical relations to one another as part of a systematic, unified conception of the ideal sciences in which forms are ranged under more comprehensive forms or classes of forms. Modern geometry can be seen as a model for at least part of what he has in mind. We think of each geometry as giving us a different spatial ontology. These spatial ontologies or „manifolds“ are then all related to one another in a systematic, interconnected whole. Each ontology is governed by a set of laws that express properties that remain invariant under particular groups of transformations. Such unification is, as it were, an ideal of reason. The theory of manifolds that Husserl envisions is intended to be much more comprehensive than these examples. It should truly be a mathesis universalis. It would evidently be a realization of the goal of maximal ratio-

26

Richard Tieszen

nality. The ideal is to include all deductive theories. This might put one in the mind of modern developments such as category theory. Husserl’s ideas about manifold theory in the third task of pure logic have in fact been likened to ideas in universal algebra or the Bourbaki program, except that one has to also be careful about how manifold theory might differ from these and from category theory (see Rosado-Haddock 2006). In the developments of logic and mathematics since Husserl’s time we can see that various elements of Husserl’s three tasks for pure logic have in fact been realized or at least partially realized. In philosophy, however, the effort to naturalize logic and mathematics in one way or another has not abated.

Literature Da Silva, J. 1998: „Husserl’s Conception of Logic“, in: Manuscrito XXII/2, 367–387. Da Silva, J. 2000: „Husserl’s Two Notions of Completeness“, in: Synthese 125, 417–438. Frege, G. 1893: Grundgesetze der Arithmetik I, Jena. Frege, G. 1894: „Rezension von E. Husserl: Philosophie der Arithmetik“, in: Zeitschrift für Philosophie und philosophische Kritik 103, 313–332. Gödel, K. 1961: „The Modern Development of the Foundations of Mathematics in the Light of Philosophy“, in: Gödel, K. 1995: Collected Works, Vol. III, S. Feferman et al (eds.), Oxford, 374–387. Hill, C. 2000: „Husserl’s Mannigfaltigkeitslehre“, in: Hill, C./Rosado-Haddock, G.: Husserl or Frege?, Chicago and La Salle, Ill, 161–178. Husserl, E. 1975: Introduction to the Logical Investigations, The Hague. Rosado-Haddock, G. 2006: „Husserl’s Philosophy of Mathematics: Its Origin and Relevance“, in: Husserl Studies 22, 193–222. Smith, B. 1989: „Logic and Formal Ontology“, in: Mohanty, J. N./McKenna, W. (eds.): Husserl’s Phenomenology: A Textbook, Lanham, 29–68. Smith, D. 2003: „Pure Logic, Ontology, and Phenomenology“, in: Revue Internationale de Philosophie 57, 133–156. Tieszen, R. 2004: „Husserl’s Logic“, in: Gabbay, D./Woods, J. (eds.): Handbook of the History of Logic, Vol. III. The Rise of Modern Logic: From Leibniz to Frege Amsterdam, 207– 321. Tieszen, R. 2005: Phenomenology, Logic, and the Philosophy of Mathematics, Cambridge. Willard, D. 1984: Logic and the Objectivity of Knowledge, Athens, Ohio.

3 Robert Hanna

Husserl’s Arguments against Logical Psychologism (Prolegomena, §§ 17–61)

3.1. Introduction According to Edmund Husserl in the Prolegomena to Pure Logic, which constitutes the preliminary rational foundation for – and also the entire first volume of – his Logical Investigations, pure logic is the a priori theoretical, nomological science of „demonstration“ (LI 1, 57; Hua XVIII, 23). For him, demonstration includes both consequence and provability. Consequence is the defining property of all and only formally valid arguments, i. e., arguments that cannot lead from true premises to false conclusions. And provability (a. k. a. „completeness“) is the property of a logical system such that, for every truth of logic in that system, there is, at least in principle, a rigorous step-by-step logically valid procedure demonstrating its validity according to strictly universal, ideal, and necessary logical laws. In this way, the laws of pure logic completely determine its internal structure. Moreover, these laws and these proofs are all knowable a priori, with selfevident insight (LI 1, 196; Hua XVIII, 185–195). So not only is pure logic independent of any other theoretical science, in that it requires no other science in order to ground its core notion of demonstration, it also provides both epistemic and semantic foundations for every other theoretical science, as well as every practical discipline or „technology.“ To the extent that pure logic is the foundation of every other  Citations of Husserl include an abbreviation of the English title, volume number, and page number, followed by the corresponding volume number of the Husserliana, and corresponding page number. The English edition used is Findlay’s translation of the Logical Investigations (1970, = LI). I generally follow the English translation, but have occasionally modified it where appropriate.

28

Robert Hanna

theoretical science, it is the „theory of science“ (Wissenschaftslehre) in Bolzano’s sense of that term (LI 1, 60; Hua XVIII, 27), the „science which deals with the ideal essence of science as such“ (LI 1, 236; Hua XVIII, 244), and thus the science of science. Logical Psychologism, or LP, is a particularly strong version of the denial that pure logic is an independent and absolutely foundational science. LP was a widely held view in the second half of the 19th century, grew out of the neo-Kantian and neo-Hegelian traditions alike, and is closely associated with the origins of empirical psychology as an autonomous discipline (Kusch 1995). Husserl’s arguments against LP in chapters 1–8 of the Prologemena, often referred to simply as Husserl’s „refutation“ of LP, constitute one of the most famous and broadly influential critical set pieces in 20th century philosophy, comparable in these respects to W. V. O. Quine’s attack on the analytic-synthetic distinction in Two Dogmas of Empiricism published almost exactly fifty years after the Prolegomena. In this connection, it is surely by no means a historical or philosophical accident that the original working title of another one of Quine’s famous and closely-related essays from the same period was Epistemology Naturalized: Or, the Case for Psychologism (Kusch 1995, 11). By the 1950s, psychologism was making a serious comeback in epistemology, if not in the philosophy of logic. But radically unlike Quine’s seminal papers (Quine 1961, Quine 1969, Quine 1976a, Quine 1976b), which are still widely read, studied, and taught in contemporary North American and European departments of philosophy, Husserl’s Prolegomena nowadays is rarely read or studied, and even more rarely taught. To the extent that the debate between LP and anti-psychologism is still an issue, it is the logico-philosophical writings of Gottlob Frege that are taken as the seminal texts on anti-psycholgism. It is obvious that Husserl’s conception of pure logic shares much with Frege’s conception of pure logic in his 1879 Begriffsschrift and other manuscripts he was working on in the 1880s and 90s (Frege 1979), even allowing for differences in the formal details of their logical theories. It is also obvious that Husserl’s critique of LP shares much with Frege’s critique of LP in his 1884 Foundations of Arithmetic and the Foreword of his 1893 Basic Laws of Arithmetic, and that there is a direct, important, influential relationship between Frege’s devastating 1894 review of Husserl’s Philosophy of Arithmetic (Frege 1984) and Husserl’s lengthy and passionate defense of his conception of pure logic against LP. Indeed, this is all explicitly conceded by Husserl in the second half of an unintentionally ironic footnote buried away almost exactly in the middle of the Prolegomena (LI 1, 179, n. 2; Hua XVIII, 172, n. 2).

Husserl’s Arguments against Logical Psychologism

29

But whatever the precise nature of Frege’s influence on Husserl himself, and whatever the contemporary status of Frege’s anti-psychologistic writings, Husserl’s arguments against LP in chapters 3–8 of the Prolegomena are independently philosophically interesting, and in fact they had a massively greater intellectual and professional impact on the development of German and European philosophy in the first half of the 20th century, than Frege’s arguments did (Kusch 1995, chs. 1, 3, 4). Moreover, and perhaps most importantly, as we shall see in section IV, one of the deepest problems in the philosophy of logic arises directly from Husserl’s arguments against LP. Husserl’s two-part response to this deep problem offers a prima facie compelling line of argument to which contemporary philosophers of logic and philosophical logicians should pay close attention.

3.2. What LP is, and its Three Cardinal Sins According to Husserl, LP is the thesis that „the essential theoretical foundations of logic lie in psychology, in whose field those propositions belong – as far as their theoretical content is concerned – which give logic its specific character (Gepräge).“ (LI 1, 90; Hua XVIII, 63) In this way, LP is the thesis that logic is explanatorily reducible to empirical psychology (Hanna 2006, ch. 1), in the strong, dual sense that (i) a complete knowledge of the empirical, natural facts and causal laws with which empirical psychology deals would yield a complete a priori knowledge of the existence and specific character of logic, and (ii) the empirical, natural facts and causal laws with which empirical psychology deals strictly determine the existence and specific character of logic. Or in other words, according to LP, logic is nothing over and above empirical psychology. This does not entail that empirical psychologists of logic are themselves logicians, but instead only that whatever it is that logicians know about logic, can in principle be known by empirical psychologists wholly and solely by virtue of their knowing all the empirical, natural facts and causal laws that are relevant to logical thinking. Husserl’s presentation of LP proceeds by means of a lengthy and sometimes repetitive critical exposition of the views of the leading recent and exponents of LP, including Mill, Bain, Spencer, Wundt, Sigwart, Erdmann,

30

Robert Hanna

Lange, Lipps, Mach, and Avenarius. Against the „psychologicists,“ Husserl explicitly aligns himself with Leibniz, Kant, Herbart, Bolzano, Lotze, and (somewhat more covertly, as I noted above) Frege. In the crucial case of Kant, however, there is some apparent equivocation, when in a footnote Husserl asserts that „even transcendental psychology also is psychology“ (LI 1, 122, n.1; Hua XVIII, 102, n. 3). This apparent equivocation on Husserl’s part can perhaps be explained away by distinguishing between Kant’s theory of logic, which is explicitly and strongly anti-psychologistic (Hanna 2001, 71–76), and neo-Kantian theories of logic, which are arguably psychologistic. If this is correct, then Husserl is not really equivocating; instead, he is attributing psychologism to the mere followers (a. k. a. „epigones“) of Kant, but not to Kant himself, who would on the contrary be historically and rhetorically aligned with Husserl’s own anti-psychologism. Quite apart from the historical and rhetorical vehicle of Husserl’s critique of LP, however, its underlying content and structure involve, first, a pair-wise contrastive characterization of LP’s conception of logic over and against Husserl’s own conception of pure logic, and then second, a set of critical arguments showing how LP either fails by external rational standards or internally refutes itself. The pair-wise contrastive characterization of logic according to LP versus pure logic according to Husserl can be summarized as follows: Logic according to LP is: Pure Logic according to Husserl is: contingent based on particulars based on empirical facts concretely real governed by causal laws conditional belief-based based on relativized, subjective truth known by sense experience a posteriori empirical instrumentally normative

necessary based on real universals based on non-empirical essences abstractly ideal governed by strictly universal laws unconditional truth-based based on absolute, objective truth known by self-evident insight a priori non-empirical categorically normative

It should be especially noticed that the items on the left-hand side all differ from the corresponding items on the right hand side not in degree but rather in kind. In each pairing, some extra non-natural or ideal property has been added by Husserl to the right-hand item of that pair in order

Husserl’s Arguments against Logical Psychologism

31

to distinguish it in kind from the corresponding item on the left-hand side. The extra properties attributed by Husserl to pure logic are „nonnatural“ or „ideal“ in two senses. First, none of the extra properties is to be found in the physical, spatiotemporal world. Second, none of the extra properties is knowable by experiential, experimental methods. So according to Husserl, pure logic is uniquely characterizable in terms of a set of special non-natural or ideal kinds to which LP has no ontological access (since LP has access only to the physical, spatiotemporal world) or explanatory access (since LP has access only to concepts and beliefs that are generated by experiential, experimental methods). This catalogue of sharply opposed conceptions of logic is then strategically exploited by Husserl in his three basic charges against LP – as it were, the three „cardinal sins“ of LP. Husserl’s first basic charge against LP is that LP is committed to what I will call Modal Reductionism about Logic or MRL, which says logical laws and logical truths are explanatorily reducible to merely causal laws and merely contingent, probabilistic truths: „The task of psychology is to investigate the laws governing the real connections of mental events with one another, as well as with related mental dispositions and corresponding events in the bodily organism […]. Such connections are causal. The task of logic is quite different. It does not inquire into the causal origins or consequences of intellectual activities, but into their truth-content.“ (LI 1, 93–94; Hua XVIII, 67) „Laws of thought, as causal laws governing acts of knowledge in their mental interweaving, could only be stated in the form of probabilities.“ (LI 1, 101; Hua XVIII, 76) Logical laws according to Husserl are necessary rules, and logical truth according to Husserl is necessary truth. On the classical Leibnizian account, a rule or proposition is logically necessary if and only if it is true in every „possible world,“ i. e., in every total set of „compossible“ or essentially mutually consistent substances, insofar as this compossibility is completely envisioned by God. Sometimes this Leibnizian, or theocentric, type of logical necessity is also called metaphysical necessity. By contrast, on the classical Kantian account, a rule or proposition is logically necessary if and only if it is „strictly universal“ and also „analytic,“ i. e., it is true in a complete class of humanly conceivable variants on the actual experienced world, there is no humanly conceivable variant on the actual experienced world which is an admissible counterexample to it, and its denial would entail

32

Robert Hanna

a contradiction (Hanna 2001, chs. 3 and 5). Sometimes this Kantian, or anthropocentric, type of logical necessity is also called conceptual necessity. Otherwise put now , and regardless of whether the necessity is construed as metaphysical necessity (Leibnizian or theocentric logical necessity) or as conceptual necessity (Kantian or anthropocentric logical necessity), logical laws and logical truths, as necessary, are always absolutely or unrestrictedly true. By sharp contrast, merely causal laws and merely probabilistic laws are inherently restricted by brute facts about the actual world. As Hume pointed out, there is no absolute guarantee that any causal law, no matter how generally it holds in the actual world of sensory experiences, will always hold. And mere probabilities, no matter how probable, are always less than 1. So Husserl’s first basic charge against LP, or MRL, says that by explanatorily reducing logical laws and logical truths to merely causal laws and merely contingent, probabilistic truths, LP radically restricts the scope of pure logical truth. Husserl’s second basic charge against LP is that LP is committed to what I will call Epistemic Empiricism about Logic or EEL, which says that logical knowledge is explanatorily reducible to merely a posteriori knowledge: „[According to LP] no natural laws can be known a priori, nor established by sheer insight. The only way in which a natural law can be established and justified, is by induction from the singular facts of experience.“ (LI 1, 99; Hua XVIII, 73 f.) „On this basis [of LP], no assertion could be certainly judged correct, since probabilities, taken as the standard of all certainty, must impress a merely probabilistic stamp on all knowledge.“ (LI 1, 101; Hua XVIII, 76) Logical knowledge according to Husserl is a priori knowledge and also certain knowledge. A priori knowledge, in turn, is belief that is sufficiently justified by evidence which is underdetermined by all sets and sorts of sensory experiences, possibly also including evidence that includes no sensory experience whatsoever and is rationally „pure.“ Certain knowledge is indubitable belief, i. e., belief that is not open to refutation by actual or possible counterexamples, and more particularly not open to refutation by sensory experiences or factual statistics. So Husserl’s second basic charge against LP, or EEL, says that LP radically underestimates the epistemic force of pure logical knowledge. Husserl’s third basic charge against LP is that it is committed to what I will call Skeptical Relativism about Logic, or SRL, which says that logical laws, logical necessary truth, and logical knowedge are explanatorily reducible

Husserl’s Arguments against Logical Psychologism

33

to either individually-held beliefs (individual relativism) or species-specific beliefs (specific relativism): „In order to criticize psychologism we have […] to discuss the concept of subjectivism or relativism, which is also part of the abovementioned [skeptical] theory. One of its original forms is caught in the Protagorean formula: ‚man is the measure of all things,‘ provided this last is interpreted as saying ‚The individual man is the measure of all truth.‘ For each man that is true which seems to him true, one thing to one man and the opposite to another, if that is how he sees it. We can therefore opt for the formula ‚All truth (and knowledge) is relative‘ – relative to the contingently judging subject. If, however, instead of such a subject, we make some contingent species of judging beings the pivot of our relations, we achieve a new form of relativism. Man as such is then the measure of all truth. Every judgment whose roots are to be found in what is specific to man, in the constitutive laws of man as species – is a true judgment, for us human beings. To the extent that such judgments belong to the form of common human subjectivity, the term ‚subjectivism‘ is in place here too (in talk of the subject as the ultimate source of knowledge, etc.). It is best to employ the term ‚relativism‘, and to distinguish individual from specific relativism. The restriction of the latter to the human species, stamps it as anthropologism.“ (LI 1, 138; Hua XVIII, 122) Relativism – or more precisely, cognitive relativism, which is about theoretical beliefs and truth, as opposed to moral relativism, which is about ethical beliefs and principles of conduct – says that truth is determined by belief or opinion. There are two distinct types of cognitive relativism. On the one hand, individual cognitive relativism says that truth is determined by individual beliefs or opinions (= subjective truth); and on the other hand, specific cognitive relativism or anthropologism says that truth is determined by beliefs or opinions that are either the result of human agreement (= truth by mutual contract, or truth by general convention) or are innately biologically specified in all human beings (= truth by instinct). According to Husserl, logical truth is objective truth, hence mind-independent truth, hence truth that is inherently resistant to determination by any merely subjective, contractual, conventional, or biological facts. So Husserl’s third basic charge against LP, or SRL, says that LP implies a mistaken and indeed ultimately skeptical theory of the determination of truth.

34

Robert Hanna

3.3. Husserl’s Three Basic Arguments against LP Corresponding to the three „cardinal sins“ of LP, Husserl develops three basic arguments against it. It is possible to spell out Husserl’s arguments in step-by-step detail (Hanna 1993; Kusch 1995, ch. 3). But for our purposes here, it is necessary only to cite Husserl’s formulations of the arguments, describe their general form, and then offer a brief exposition of Husserl’s underlying rationale for each argument.

3.3.1. Husserl’s Argument against LP from its Modal Reductionism about Logic (MRL) Here is what Husserl says about MRL: „[According to LP] logical laws, must accordingly, without exception rank as mere probabilities. Nothing, however, seems plainer than that the laws of ‚pure logic‘ all have a priori validity.“ (LI 1, 99; Hua XVIII, 74) „The psychologistic logicians ignore the fundamental, essential, never-to-be bridged gulf between ideal and real laws, between normative and causal regulation, between logical and real necessity, between logical and real grounds. No conceivable gradation could mediate between the ideal and the real.“ (LI 1, 104; Hua XVIII, 80) Here is the general form of Husserl’s anti-MRL argument: (1) LP entails MRL. (2) MRL is inconsistent with the existence and specifically modal character of pure logic – in particular, MRL is inconsistent with the absolute necessity of pure logical laws and pure logical truths. (3) Therefore, LP is false. And here is the underlying rationale for Husserl’s anti-MRL argument. Given Husserl’s characterization of the modal character of pure logic, it follows that pure logical laws and pure logical truths are absolutely or unrestrictedly true, regardless of whether this absolute truth is construed, Leibniz-wise, as metaphysical necessity, or else construed, Kant-wise, as conceptual necessity. Now if LP is correct, then MRL is correct, and then logical laws and logical truths are non-absolutely or restrictedly true precisely because they are restricted to the actual world. But logical laws and logical truths are absolutely or unrestrictedly true. So LP must be false.

Husserl’s Arguments against Logical Psychologism

35

3.3.2. Husserl’s Argument against LP from its Epistemic Empiricism about Logic (EEL) Here is what Husserl says about EEL: „[The laws of pure logic] are established and justified, not by induction, but by apodeictic inner self-evidence. Insight justifies no mere probabilities of their holding, but their holding or truth itself.“ (LI 1, 99; Hua XVIII, 74) „The justified possibility of [the exact factual sciences] becomes the absurdity of [pure logic]. We have insight into, not merely the probability, but the truth of logical laws. Against the truth that is itself grasped with insight, the strongest psychologistic argument cannot prevail; probability cannot wrestle with truth, nor surmise with insight.“ (LI 1, 100; Hua XVIII, 75) „How plausible the ready suggestions of psychologistic reflection sound. Logical laws are laws for validation, proofs. What are validations but perculiar human trains of thought, in which, in normal circumstances, the finally emergent judgments seem endowed with a necessarily consequential character. This character is itself a mental one, a peculiar mode of mindedness and no more […]. How could anything beyond empirical generalities result in such circumstances? Where has psychology yielded more? We reply: Psychology certainly does not yield more, and cannot for this reason yield the apodeictically evident and so metempirical and absolutely exact laws which form the core of all logic.“ (LI 1,100–101; Hua XVIII, 75 f.) Here is the general form of Husserl’s anti-EEL argument: (1) LP entails EEL. (2) EEL is inconsistent with the existence and specifically epistemic character of pure logic – in particular, EEL is inconsistent with the self-evident insights of pure logical knowledge, which are both a priori and certain. (3) Therefore, LP is false. And here is the underlying rationale for Husserl’s anti-EEL argument. Given Husserl’s characterization of the epistemic character of pure logic, it follows that logical beliefs are sufficiently justified by self-evident insights, i. e., rational intuitions. Self-evident insights, or rational intuitions, are a priori or non-empirical, or if not strictly infallible, then at least certain and indubitable. Now if LP is correct, then EEL is correct, and then even suffi-

36

Robert Hanna

ciently justified logical beliefs are all a posteriori or empirical, fallible, and dubitable. But sufficiently justified logical beliefs are a priori and certain or indubitable. So LP must be false.

3.3.3. Husserl’s Argument against LP from its Skeptical Relativism about Logic (SRL) Here is what Husserl says about SRL: „[The individual relativist] will naturally reply: My theory expresses my standpoint, what is true for me, and need be true for no one else. Even the subjective fact of his thinking he will treat as true for himself and not as true in itself […] The content of such assertions rejects what is part of the sense or content of every assertion and what accordingly cannot be significantly separated from any assertion.“ (LI, 1, 139; Hua XVIII, 123) „Specific relativism makes the assertion: Anything is true for a given species of judging beings that, by their constitution and laws of thought, must count as true. This doctrine is absurd. For it is part of its sense that the same proposition or content of judgment can be true for a subject of the same species […], but may be false for another subject of a differently constituted species. The same content of judgment cannot, however, be both true and false: this follows from the mere sense of ‚true‘ and ‚false‘. If the relativist gives these words their appropriate meaning, this thesis is in conflict with its own sense […]. ‚Truth for this or that species,‘ e. g., for the human species, is, as here meant, an absurd mode of speech. It can no doubt be used in good sense, but then it means something wholly different, i. e., the circle of truths to which man as such has access. What is true absolutely, intrinsically true: truth is one and the same, whether men or non-men, angels or gods apprehend it. Logical laws speak of this ideal unity, set over against the real multiplicity of races, individuals, and experiences, and it is of this ideal unity that we all speak when we are not confused by relativism.“ (LI 1, 140; Hua XVIII, 125) Here is the general form of Husserl’s anti-SRL argument: (1) LP entails SRL. (2) SRL is self-refuting, given the fact of the existence and specifically alethic (i. e., truth-based) character of pure logic – in particular, SRL is inconsistent with the objectivity of the truths of pure logic.

Husserl’s Arguments against Logical Psychologism

37

(3) Therefore, LP is false. And here is the underlying rational for Husserl’s anti-SRL argument. Given Husserl’s characterization of the alethic character of pure logic, it follows that logical truth is objective, or mind-independent, and inherently resistant to determination by merely subjective, contractual, conventional, or biological facts. Now if LP is correct, then SRL is correct, and then truth is either individually relativized or specifically relativized. Suppose that truth is individually relativized. Then whatever anyone believes or opines is true, is true. This includes the person who believes or opines that LP is false. So if truth is individually relativized, then LP is both true (relative to the defender of LP) and false (relative to the critic of LP) and thus self-contradictory. Suppose, alternatively, that truth is specifically relativized. Then there can be other communities, or other species, that say radically different and opposing things about the nature of truth. This is the possibility of conceptual, semantic, and theoretical incommensurability. But given the possibility of conceptual, semantic, and theoretical incommensurability, it follows that these other communities or other species are really talking about something other than what we mean by „truth“ – instead, they are really talking about schmuth, or whatever. But truth, after all, is objective or mind-independent. So if truth is specifically relativized, then these other communities or other species are not actually disagreeing with us about truth, since they are talking about something other than truth. To summarize: If LP is correct, then SRL is correct, and if SRL is correct, then it is either self-contradictory or talking about something other than truth. So LP must be false.

3.3.4. Has Husserl Begged the Question against LP? The Logocentric Predicament, and a Husserlian Way Out It should be very clear from the previous section that Husserl’s three basic arguments against LP all have the same general form, and that they all directly invoke non-natural or ideal facts about the specific character of pure logic, whether modal, epistemic, or alethic. But it can be objected that Husserl only ever asserts that pure logic exists and also has the several non-natural or ideal specific characters he attributes to it, and that he never actually justifies this assertion. In this way, on the face of it, Husserl seems to have merely begged the question against LP.  The question-begging objection was first made in 1901 by Paul Natorp. See Natorp 1977, 57. See also Hanna 2006, ch. 1; and Kusch 1995, ch. 4.

38

Robert Hanna

But has he? It is equally clear that Husserl would reply to this charge by saying that he has not begged the question against LP. Instead, and on the contrary, what he has done is to show that and also precisely how the existence and specific character of pure logic is covertly presupposed and used, even by the defenders of LP: „Logic […] can as little rest on psychology as on any other science; since each science is only a science in virtue of its harmony with logical rules, it presupposes the validity of these rules. It would therefore be circular to try to give logic a first foundation in psychology.“ (LI 1, 95; Hua XVIII, 69)

In other words, since LP is a theory, it fall under logical constraints, e. g., laws of logical consistency, laws of logical consequence, and the inferential justification of its theses and beliefs. So LP covertly invokes pure logic, just as every other theory and every science explicitly or implicitly invokes pure logic. But given this line of argument, as Husserl himself anticipates, the defenders of LP have one last arrow in their quiver, and it is a very sharp one indeed: „The opposition will reply: That this argument cannot be right, is shown by the fact that it would prove the impossibility of all logic. Since logic itself must proceed logically, it would itself commit the same circle, and would itself have to establish the validity of rules that it presupposes.“ (LI 1, 95; Hua XVIII, 69) In other words, the defenders of LP will retreat to the charge that in his showing pure logic to be what is covertly presupposed and used by the defenders of LP, Husserl has himself run up against one of the deepest problems in the philosophy of logic, namely, the explanatory and justificatory circularity of logic – or what the Harvard logician Harry Sheffer later very aptly called the „logocentric predicament“: „The attempt to formulate the foundations of logic is rendered arduous by a […] ‚logocentric‘ predicament. In order to give an account of logic, we must presuppose and employ logic.“(Sheffer 1926, 228) A specific version of the Logocentric Predicament is Lewis Carroll’s famous skeptical argument, published in Mind in 1895 – and which Husserl may well have read, or at least have read about – which says that that any attempt to generate the total list of premises required to deduce the conclusion

Husserl’s Arguments against Logical Psychologism

39

of a valid argument leads to a vicious regress (Carroll 1895). But for our purposes here, the Logocentric Predicament is just this: How can pure logic in Husserl’s sense ever be explained or justified, if every explanation or justification whatsoever both presupposes and uses pure logic in Husserl’s sense? How will Husserl respond to the Logocentric Predicament? One possible way out of the Logocentric Predicament would be for Husserl just to concede that pure logic is explanatorily and justificationally groundless, in the manner of the imaginary mock-logician invented by Carroll, Tweedledee: „If it was so, it might be; and if it were so, it would be; but as it isn’t, it ain’t. That’s logic.“ (Carroll 1988) But then Husserl would have no rational defense against LP and no rational response to the Logocentric Predicament. And it would clearly be selfstultifying for Husserl to defend anti-psychologism and to respond to the Logocentric Predicament by lapsing into a non-rational, or as it were fideist, approach to the foundations of pure logic, which by Husserl’s own reckoning – not to mention by an historical and rhetorical appeal to the authority of Kant’s theory of logic – is supposed to provide categorically normative laws of rationality. It made good sense for Kant to claim in the Preface to the Critique of Pure Reason that in order to make room for moral faith in freedom of the will, he had to „deny“ or limit our scientific knowledge of universal natural determinism; but it would make no sense for Husserl to say that in order to make room for pure logic, he had to deny rationality. Husserl’s actual strategy of response to the Logocentric Predicament has two parts. First, he distinguishes carefully between reasoning according to logical rules, and reasoning from logical rules: „Let us, however, consider more closely what such a circle would consist in. Could it mean that psychology presupposes the validity of logical laws? Here one must notice the equivocation in the notion of ‚presupposing‘. That a science presupposes the validity of certain rules may mean that they serve as premises in its proofs: it may also mean that they are rules in accordance with which the science must proceed in order to be a science at all. Both are confounded in our argument for which reasoning according to logical rules, and reasoning from logical rules, count as identical. There would be a circle only if the reasoning were from such rules. But, as many an artist works without the slightest knowledge of aesthetics, so an investi-

40

Robert Hanna gation may construct proofs without ever having recourse to logic. Logical laws cannot therefore have been premises in such proofs. And what is true of single proofs is likewise true of whole sciences.“ (LI 1, 95; Hua XVIII, 69 f.)

What Husserl is saying is that it is only if one mistakenly confuses reasoning according to logical rules and reasoning from logical rules that one will also cite those logical rules as axiomatic premises in one’s argument, and thereby encounter the circularity problem. But logical rules can be perfectly legitimately used in proofs without also citing or mentioning them as premises in those very proofs. Indeed, the very idea of natural deduction systems, later discovered by Gerhard Gentzen, is based on this fact (Gentzen 1969). Furthermore, that Husserlian observation seems to be precisely the right reply to make to Carroll’s vicious regress version of the Predicament (Hanna 2006). But I think that Husserl is also making an even deeper point than this one. His deeper point is that it is not only possible but necessary, given our commitment to human rationality, to conceive of the laws of pure logic as supreme constructive categorically normative logical meta-principles, telling us how we unconditionally ought to go about constructing all possible lowerorder logical principles or rules, all possible lower-order logical proofs, all possible lower-order logical systems, all possible lower-order exact scientific principles or rules, all possible lower-order exact scientific proofs, and all possible lower-order exact sciences themselves. It is to be particularly emphasized that this does not mean that the lower-order sciences are supposed to be deduced from these supreme meta-principles, construed as axiomatic premises. Instead and on the contrary, the lower-order sciences are all simply constructed and operated according to these supreme constructive categorically normative meta-principles. This deeper point, in turn, leads directly to the second step of Husserl’s response to the Predicament. Second, then, Husserl explicitly addresses the issue of how to characterize the explanatory and justificatory status of pure logic, when we assume we must always reason according to (i. e., not from) the laws of pure logic conceived as supreme constructive (i. e., not deductive) categorically normative (i. e., not instrumental, causal, or merely descriptive) meta-principles (i. e., not lower-order principles) that tell us how we unconditionally ought to construct first-order exact sciences, including all first-order logical systems. Here is what he says: „[The unifying aim or purpose of pure logic] is the ideal of a pervasive, all-embracing rationality. If all matters of fact obey laws, there

Husserl’s Arguments against Logical Psychologism

41

must be some minimum set of laws, of the highest generality […]. These ‚basic laws‘ are, accordingly, laws of supreme coverage and efficacy, whose knowledge yields the maximum of insight in some field, which permits the explanation of all that is in any way explicable in that field. […] This goal or principle of maximum rationality we recognize with insight to be the supreme goal of the rational sciences. It is self-evident that we would be better for us to know laws more general than those which, at a given time, we already possess, for such laws would lead us back to grounds deeper and more embracing. Plainly, however, our principle is no mere biological principle, or principle of thought-economy: it is a purely ideal principle, an eminently normative one […]. The ideal drift of logical thinking is as such towards rationality.“ (LI 1, 208; Hua XVIII, 209 f.) In other words, Husserl is arguing that insofar as we must always reason according to pure logic, and insofar as the laws of pure logic are conceived as supreme constructive categorically normative meta-principles for constructing all lower-order exact sciences, then it follows that pure logic is the necessary a priori condition of the possibility of any explanation or justification whatsoever, in the sense that it is innately constitutive of human rationality. This argument assumes, as a „transcendental fact,“ that we are rational human animals, and that as a consequence our manifest capacity for generating and using pure logic in the cognitive or practical construction of any explanation or justification whatsoever belongs innately to our cognitive and practical rational human nature. Therefore pure logic exists and also has the specific character attributed to it by Husserl. In turn, from this „transcendental argument from rationality“ it would also directly follow that Husserl’s arguments against LP are sound. Whether or not one ultimately accepts a Husserl-style transcendental rationalist solution to the Logocentric Predicament (Hanna 2006, chs 3,7), and whether or not one ultimately accepts Husserl’s correspondingly robust reinforcement of his arguments against LP, which might otherwise seem to be question-begging, nevertheless Husserl’s response to the Logocentric Predicament is at least prima facie compelling. It therefore provides an  In the elided passage, Husserl seems to be asserting precisely what he himself had earlier rejected in his response to the circularity objection – namely, that the laws of pure logic are themselves axiomatic premises in deductive proofs. But charitably interpreted, this must be a mere slip. Even Husserl nods.

42

Robert Hanna

independently sufficient reason for contemporary philosophers of logic and philosophical logicians to re-read and seriously reconsider Husserl’s Prolegomena to Pure Logic. Husserl’s Prolegomena §§ 17–61 provides a classic, and arguably independently defensible, defense of anti-psychologism. One can foresee a day when every History of Twentieth Century Philosophy course everywhere will begin its list of Required Readings with selections from the Prolegomena, alongside the familiar selections from Frege, and when Quine’s so-called „refutation“ of the analytic-synthetic distinction will also be compelled to face up to the Logocentric Predicament.

Literature Carroll, L. 1895: „What the Tortoise Said to Achilles,“ in: Mind 4, 278–280. Carroll, L. 1988: Through the Looking-Glass, New York. Frege, G. 1979: „Logic [1897]“, in: Frege, G: Posthumous Writings, trans. P. Long et al., Chicago, 127–151. Frege, G. 1984: „Review of E. G. Husserl, Philosophie der Arithmetik I,“ in: Frege, G: Collected Papers in Mathematics, Logic, and Philosophy, trans. M. Black et al., Oxford, 195–209. Gentzen, G. 1969: „Investigations into Logical Deduction,“ in: Gentzen, G.: The Collected Papers of Gerhard Gentzen, trans. M. Szabo, Amsterdam, 68–131. Hanna, R. 1993: „Logical Cognition: Husserl’s Prolegomena and the Truth in Psychologism,“ in: Philosophy and Phenomenological Research 53, 251–275. Hanna, R. 2001 Kant and the Foundations of Analytic Philosophy, Oxford. Hanna, R. 2006: Rationality and Logic, Cambridge, MA. Husserl, E. 1970: Logical Investigations, trans. J. N. Findlay, 2 vols., London. Kusch, M. 1995: Psychologism, London. Natorp, P. 1977: „On the Question of Logical Method in Relation to Edmund Husserl’s Prolegomena to Pure Logic,“ in: Mohanty, J. N. (ed.): Readings on Edmund Husserl’s Logical Investigations, The Hague, 55–66. Quine, W. V. O. 1961: „Two Dogmas of Empiricism,“ in: Quine, W. V. O.: From a Logical Point of View, 2nd edn., New York, 20–46. Quine, W. V. O. 1969: „Epistemology Naturalized,“ in: Quine, W. W. O.: Ontological Relativity, New York, 69–90. Quine, W. V. O. 1976a: „Truth by Convention,“ in: Quine, W. V. O.: The Ways of Paradox, 2nd edn., Cambridge, MA, 77–106. Quine, W. V. O. 1976b: „Carnap and Logical Truth,“ in: Quine, W. V. O. : The Ways of Paradox, 2nd edn., Cambridge, MA, 107–132. Sheffer, H. M. 1926: „Review of Principia Mathematica, Volume I, second edition,“ in: Isis 8, 226–231.

4 Vittorio De Palma

Husserls phänomenologische Semiotik* (I. Logische Untersuchung, §§ 1–23)

Ziel der Logischen Untersuchungen ist es, eine phänomenologische Aufklärung der Logik durch Rückgang zu den psychischen Akten zu liefern, in denen die idealen Bedeutungen gegeben werden, von denen sie aber unterschieden sind. Das Werk erwächst aus der Krise der „psychologischen Begründung“ des Logischen, die Husserl in der Philosophie der Arithmetik vertrat, und aus Reflexionen „über das Verhältnis zwischen der Subjektivität des Erkennens und der Objektivität des Erkenntnisgehaltes“ (Hua XVIII, 6 f.). Die Frage ist, „wie es […] zu verstehen sei, daß das ,an sich‘ der Objektivität zur ‚Vorstellung‘ komme, also am Ende doch wieder subjektiv werde“ (12 f.)1. Es geht für Husserl darum, die Einflüsse Lotzes und Bolzanos einerseits und den Ansatz seines Lehrers Brentano anderseits zu versöhnen bzw. die „unbegreiflich fremden Welten […] des rein Logischen und […] des aktuellen Bewußtseins, […] des Phänomenologischen und auch Psychologischen […] in eins zu setzen“ (Hua XXIV, 442 f.). Die Logik ist nach Husserl die „Wissenschaft von Bedeutungen als solchen“ (98), d. h. von den idealen Einheiten, die den theoretischen Gehalt jeder Wissenschaft bilden, in mannigfaltigen Ausdrücken formuliert und in mannigfaltigen Akten gedacht, aber von den zufälligen Ausdrücken und * Hinweise auf Hua XIX werden nur mit Seitenangaben gegeben. Mein Dank gebührt der Alexander von Humboldt Stiftung, die mir durch die Gewährung eines Stipendiums die Durchführung dieser Arbeit ermöglicht hat. Des Weiteren danke ich Klaus Sellge für hilfreiche Verbesserungsvorschläge zu einer früheren Version dieses Textes. 1 Später hat Husserl das Verhältnis zwischen dem subjektiv psychologischen Erlebnis und dem in ihm erfassten Objekt als das ursprüngliche Problem der Phänomenologie bezeichnet (Hua II, 75; Hua VI, 169 Anm. 1). Frege selbst (1969, 157 und Anm.) sagt, der „seelische[] Vorgang“ vom „Erfassen“ eines Gedankens bzw. eines idealen Gegenstandes sei vielleicht „der geheimnisvollste von allen“.

44

Vittorio De Palma

Akten unterschieden sind. Die Hauptaufgabe der I. Untersuchung ist es, eine Abgrenzung des Begriffs Bedeutung dadurch zu gewinnen, dass die Eigentümlichkeit und Unabhängigkeit der Bedeutung gegenüber dem hervorgehoben wird, was mit ihr verbunden, doch nicht mit ihr identisch ist, also nicht mit ihr verwechselt werden darf: den Akten, in denen die Bedeutung gegeben ist, dem Gegenstand, auf den man sich durch die Bedeutung bezieht, der Anschauung, in der sich die Bedeutung erfüllt. Was Bedeutung ist, lässt sich jedoch nach Husserl nicht definieren, da es sich um „ein deskriptiv Letztes“ handelt, aber es „kann uns so unmittelbar gegeben sein, wie uns gegeben ist, was Farbe und Ton ist“ (187). An dieser Stelle sei darauf aufmerksam gemacht, dass Husserl im Rahmen der Umarbeitung der VI. Untersuchung in den Jahren 1913/14 auch den Inhalt der I. Untersuchung, deren „Unausgereiftheit“ er betont (Hua XX/2, 23), einer gründlichen Revision unterzogen hat. Aus Platzgründen wird diese Revision in der folgenden Darstellung aber kaum berücksichtigt.

4.1. Ausdruck und Zeichen Wie Husserl in der Einleitung zum zweiten Band der Untersuchungen betont, wird Wahrheit nur in der Form von Aussagen zum bleibenden Besitz der Wissenschaft, und wissenschaftliche Urteile lassen sich kaum oder gar nicht ohne sprachlichen Ausdruck vollziehen (vgl. 7 f.). Die I. Untersuchung nimmt ihren Ausgang von der Abgrenzung der Ausdrücke gegen die anderen Arten von Zeichen. Denn jeder Ausdruck ist ein Zeichen, aber nicht jedes Zeichen ist ein Ausdruck. Husserl unterscheidet drei Arten von Zeichen: (1) natürliche Zeichen (z. B. fossile Knochen); (2) künstliche Zeichen, etwa Merkzeichen (das Stigma der Sklaven, die Flagge der Nation) oder Erinnerungszeichen (den „Knopf im Taschentuche“); (3) sprachliche Zeichen bzw. Ausdrücke, d. h. jede Rede und jeder Redeteil, unabhängig davon, ob sie wirklich geredet sind, also kommunikativer Funktion dienen. Zum Ausdruck gehört, eine Bedeutung zu haben bzw. etwas zu meinen. Das unterscheidet ihn von den sonstigen Zeichen (und auch von jenen mimischen und gestischen „Ausdrücken“, durch die man nichts mitteilt bzw. die nicht in mitteilender  Vgl. dazu Bernet 1988; Senigaglia 1998; Melle 1998/99.  Die Wichtigkeit der Zeichen für das Denken betont Husserl schon in der 1890 niedergeschriebenen Abhandlung „Zur Logik der Zeichen. Semiotik“ (Hua XII, 340–373), die in vieler Hinsicht der Thematik der I. Untersuchung vorgreift.

Husserls phänomenologische Semiotik

45

Absicht vollzogen werden), die bloße Anzeichen bzw. bedeutungslose Zeichen sind. Nur die „willkürlich und in anzeigender Absicht gebildeten Zeichen“ (31) bezeichnen etwas. Doch unabhängig davon ist Anzeichen alles, was „einem denkenden Wesen tatsächlich als Anzeige für irgendwas dient“ (ebd.), d. h. alles, dessen aktueller Bestand als nichteinsichtiges Motiv erlebt wird, um das Sein von etwas anderem zu setzen oder zu vermuten. Dieser Motivierungszusammenhang ist nicht mit dem logischen Notwendigkeitszusammenhang zu verwechseln, in dem der Bestand eines Sachverhaltes aus demjenigen anderer Sachverhalte einsichtig erschlossen wird. Denn ein Sachverhalt dient zur Anzeige für andere aus ihm zu folgernde Sachverhalte nicht als logischer Grund, d. h. vermöge des idealen Zusammenhangs der betreffenden Inhalte, sondern vermöge des empirischen Zusammenhangs von Überzeugungen als psychischen Erlebnissen. Der Begriff des Anzeichens hat seinen Ursprung eben in der Ideenassoziation. Sie bewirkt, dass ein Gegenstand nicht für sich allein gilt, sondern einen von ihm verschiedenen Gegenstand vorstellig macht und mit ihm eine phänomenale Einheit bildet. Dazu gehört die Anzeige, bei der ein Gegenstand bzw. Sachverhalt auf einen anderen hinweist und die Annahme seines Bestehens in unmittelbar fühlbarer Weise empfiehlt. Dagegen ist das Hinzeigen eines bedeutsamen Zeichens auf seine Bedeutung vom Anzeigen eines Anzeichens verschieden. Es handelt sich dabei um eine andere Weise des Über-sich-Hinausweisens, denn das Verhältnis des Ausdrucks zum Ausgedrückten ist kein induktives bzw. assoziatives Hinweisen eines Daseienden auf ein anderes. Das Wort „Tisch“ steht für bzw. ist Zeichen für Tische nicht in derselben Weise, wie die Wildspuren für das Wild stehen bzw. Zeichen sind: Mit dem bedeutsamen Zeichen meinen wir etwas. Später hat Husserl eingeräumt, dass auch das künstliche nicht-sprachliche Zeichen etwas meint, aber er hat zugleich betont, dass es – im Gegensatz zum sprachlichen Zeichen – etwas direkt und nicht  Später, im Zusammenhang mit der genetischen Phänomenologie, hat Husserl die Assoziation als eine in der Besonderheit der Inhalte gegründete apriorische Synthesis betrachtet und ihr eine grundlegende Rolle in der passiven Konstitution zugeschrieben. Die assoziative Induktion von Nichtgegebenem aus dem Gegebenen aufgrund der bisherigen Erfahrung wird dann zu einem der Erfahrungskonstitution zugrundeliegenden „ursprünglichen Vernunftakt“ (Hua XIII, 354 ff.). Bemerkenswert ist, dass der erste Text, in dem die Assoziation als wesensgenetisch apperzeptionstiftend aufgefasst wird, im Kontext der Umarbeitung der VI. Untersuchung entsteht und sich auf das Zeichenbewusstsein bezieht (Hua XX/2, 184 f.). Husserl (1939, 78) selbst betrachtet die in der I. Untersuchung vollzogene Aufweisung des Ursprungs der Anzeige aus der Assoziation als „den Keim der genetischen Phänomenologie“. Zu Husserls Assoziationslehre vgl. Holenstein 1972.

46

Vittorio De Palma

in seinem „kategorialen Wie“ (Hua XX/2, 129) bezeichnet, da es keine logische Form bzw. „keine Grammatik“ (Hua XX/2, 53) hat, die ihm ermöglichte, etwas als etwas, d. h. in einer bedeutungsmäßig bestimmten Weise, auszudrücken. Sprachliche Ausdrücke sind also gegenüber allen anderen Arten von Zeichen dadurch ausgezeichnet, dass sie in kategorial gegliederter Weise etwas meinen. Diese Kategorialität des Ausdrucks kommt in der I. Untersuchung immer wieder zur Geltung. Sie impliziert unter anderem, dass im Falle sprachlicher Bedeutungen die Erfüllung nicht von schlichten bzw. sinnlichen, sondern nur von fundierten bzw. kategorialen Anschauungsakten vollzogen werden kann (vgl. 55, 62, 77). In der mitteilenden Rede fungieren die Ausdrücke auch als Anzeichen. Was die Rede zur Rede macht, liegt in der vermittelten Korrelation zwischen den sinngebenden Akten der miteinander verkehrenden Personen bzw. zwischen „Kundgabe psychischer Erlebnisse im Sprechen und Kundnahme derselben im Hören“ (39). Die Ausdrücke dienen dem Hörenden als Anzeichen für die Gedanken des Redenden, sie haben eine kundgebende Funktion, deren Inhalt die kundgegebenen Erlebnisse bilden. Die Kundnahme bzw. das Verständnis der Kundgabe findet aber „ohne jede […] begriffliche Fassung“ statt, denn sie ist kein Urteil, sondern „eine bloße Wahrnehmung der Kundgabe“ bzw. des Anderen als Sprechenden und seiner Erlebnisse, obschon Letztere vom Hörenden nicht selbst erlebt werden können (41). Hier finden sich die ersten Ansätze einer Phänomenologie der Mitteilung, die Husserl später im Rahmen seiner Beschäftigung mit der Intersubjektivität entwickelt. In den Untersuchungen ist sein Interesse jedoch nicht auf die kommunikative Funktion der Sprache, sondern ausschließlich auf ihre logische Funktion gerichtet, die darin besteht, den Gedanken in der Weise einer Bedeutung auszudrücken. Es ist diese Funktion, die in ihrer Eigentümlichkeit zur Abhebung kommen soll. Husserl greift auf das einsame Seelenleben zurück, um die Bedeutungsfunktion von der anzeigenden Funktion abzusondern und zu beweisen, dass, „was die Ausdrücke zu Ausdrücken macht“ (41) bzw. „die Bedeutung des Ausdruckes […] nicht mit seiner kundgebenden Leistung zusammenfallen kann“ (42). In der einsamen Rede fungieren die Worte zwar als Zeichen, da sie – genauso wie in der Wechselrede – auf ihre Bedeutung hinzeigen, aber nicht als Anzeichen, da sie nichts anzeigen, indem sie nicht als daseiend gegeben sind  Nach Husserl sind die Erlebnisse eines Subjekts prinzipiell nur diesem selbst direkt zugänglich und den Anderen nur durch eine Indikation gegeben, die nie wirkliche Erfahrung des Indizierten werden kann. Trotzdem hält er die Fremderfahrung bzw. Einfühlung für eine Wahrnehmung und nicht für einen Denkakt oder einen Schluss.

Husserls phänomenologische Semiotik

47

und keine Daseinsüberzeugung motivieren. Denn im einsamen Seelenleben benötigen wir nur vorgestellte oder phantasierte Wortzeichen, aber die sinnliche bzw. physische Inexistenz der Worte trifft nicht die Bedeutungsfunktion bzw. die „Funktion des Ausdrucks als Ausdruck“, sondern einzig die kundgebende bzw. mitteilende Funktion, die nur „im wirklichen Sprechen und Hören“ stattfinden kann (43). In der monologischen Rede können die Worte nicht als „Anzeichen für das Dasein psychischer Akte dienen“, da Letztere „im selben Augenblick von uns selbst erlebt“ sind (43). Die Scheidung von anzeigender und bedeutender Funktion, die in der Isolierung der Letzteren in der einsamen Rede kulminiert, ist eine der umstrittensten Stellen der Untersuchungen. Von ihr geht Derrida (1967) aus, um zu zeigen, dass Husserls Phänomenologie eine Metaphysik der Anwesenheit ist. Der Vorrang der einsamen Rede, in der die Zeichen keine empirische Existenz haben, nimmt Derrida zufolge geradezu die Reduktion vorweg.

4.2. Bedeutung, Akt und Gegenstand Der Ausdruck ist sinnbelebter Ausdruck bzw. Zeichen (und nicht bloßer Wortlaut oder bloßes physisches Phänomen) nur vermöge der psychischen Akte der Bedeutungsintention, die ihm eine Bedeutung geben und die auch den Kern der Kundgabe bilden: „Er meint etwas, und indem er es meint, bezieht er sich auf Gegenständliches.“ (44) Husserl unterscheidet auf der Aktseite Ausdruckserscheinung, Bedeutungsintention und Bedeutungserfüllung, auf der Inhaltseite Ausdruck, intendierende bzw. erfüllende Bedeutung und Gegenstand. Die Akte sind real, subjektiv und verschieden. Der Inhalt ist ideal, objektiv und identisch, d. h. er kann unter verschiedenen Umständen und von verschiedenen Personen als derselbe erkannt werden. (Dabei kann der Gegenstand eines Ausdrucks sowohl real (wie „dieser Tisch“), als auch ideal sein (wie „die Zahl 2“), ist aber als immer wieder identifizierbar stets von einer gewissen Idealität.) Der Ausdruck besteht nicht in einem hic et nunc geäußerten Lautgebilde, das nie wieder identisch wiederkehren kann. Der Signifikant bzw. das sinnliche Zeichen ist einzig gegenüber seinem vielfachen Auftreten (vgl. Hua IX, 398; Hua XVII, 24 f.), denn „die im darstellenden sinnlichen Inhalt der Anschauung wirklich vorhandene Gestalt“ (619) muss ideal gleich bleiben,  Dazu vgl. Evans 1991; Costa 1998, 127 ff.; Mensch 2000 (mit weiteren Hinweisen).

48

Vittorio De Palma

damit der Ausdruck als derselbe Ausdruck in jedem faktischen Vorkommen wiedererkennbar ist. Dasselbe gilt für das Signifikat bzw. für das, was der Ausdruck sagen will. Die ideale Bedeutung ist nicht mit den realen bedeutungsverleihenden Akten zu verwechseln, in denen sie gegeben ist, da sie in verschiedenen realen Akten identisch vollzogen werden kann. Einen Ausdruck zu verstehen heißt eben, das zu verstehen, was er meint, und nicht, was der Sprecher erlebt. Wenn ich sage: „Die drei Höhen eines Dreiecks schneiden sich in einem Punkte“, so ist mein kundgegebener Urteilsakt (also meine subjektive Überzeugung, dass es so ist, aufgrund deren ich das Urteil ausspreche) nicht die Bedeutung des Aussagesatzes, die ein identisch Wiederholbares und nichts Subjektives ist. „Was die Aussage aussagt, ist dasselbe, wer immer sie behauptend aussprechen mag, und unter welchen Umständen und Zeiten er dies tun mag. […] Der Sachverhalt selbst ist, was er ist, ob wir seine Geltung behaupten oder nicht. […] Die Urteilsakte sind von Fall zu Fall verschieden. Aber, was sie urteilen, was die Aussage besagt, das ist überall dasselbe.“ (49 f.) Wir müssen also immer „von den flüchtigen Erlebnissen des Fürwahrhaltens und Aussagens ihren idealen Inhalt, die Bedeutung der Aussage als die Einheit in der Mannigfaltigkeit“ unterscheiden (50). Hier liegt der Kern der Widerlegung des Psychologismus in den Prolegomena. Husserl selbst hatte zuvor eine gewisse Vermengung von subjektiv-psychischem Akt und objektiv-idealem Inhalt in der Philosophie der Arithmetik vollzogen, wie Frege (1894) in der Rezension jenes Werkes bemerkte und er selbst selbstkritisch eingesteht: „Das Kollektivum ist keine sachliche, in den Inhalten der kolligierten Sachen gründende Einheit; nach der mir vorgegebenen Schablone, daß alles ‚Wirkliche‘ entweder ‚Physisches‘ oder ‚Psychisches‘ ist, konnte es also nichts Physisches sein: Also entspringt die Idee des Kollektivums durch ‚Reflexion‘, nämlich auf psychische Einheitsform. Aber ist denn der Begriff der Anzahl nicht etwas anderes als der Begriff des Kolligierens?“ (Hua XX/1, 295) Um es mit Freges (1918/19) Terminologie auszudrücken: Es gibt neben der Außenwelt der physischen Dinge und der Innenwelt der psychischen Vorstellungen also noch ein „drittes Reich“, das der idealen Gegenstände. Dieser Gedanke ist grundlegend für die Logischen Untersuchungen. Nach  Es war aber nicht Frege, der Husserl auf die Unterscheidung zwischen Vorstellung, Bedeutung und Gegenstand brachte, die sich bereits in einer Schrift findet, die Freges Besprechung

Husserls phänomenologische Semiotik

49

Husserl sind aber die eigentlichen Bedeutungsträger nicht Zeichen bzw. sprachliche Entitäten, sondern psychische Akte in specie. Zwischen Bedeutung und intentionalem Akt besteht ein innerer Zusammenhang, denn die Bedeutung ist „Identisches der Intention“ (50) bzw. der ideale Inhalt von Akten des Bedeutens, also zwar unabhängig von dem einzelnen Akt, als dessen Inhalt sie gegeben ist, nicht aber von dem Akt überhaupt. „Jeder Ausdruck besagt nicht nur etwas, sondern er sagt auch über Etwas; er hat nicht nur seine Bedeutung, sondern er bezieht sich auch auf irgendwelche Gegenstände. […] Niemals fällt […] der Gegenstand mit der Bedeutung zusammen.“ (52) Bedeutung und Gegenstand sind verschieden, insofern sie unabhängig voneinander wechseln können. Denn mehrere Ausdrücke können einerseits dieselbe Bedeutung, aber verschiedene Gegenstände, sowie andererseits verschiedene Bedeutungen, aber denselben Gegenstand haben. Beispiele des ersten Falles sind die universellen bzw. vielwertigen Namen, die einen Umfang haben und verschiedene Gegenstände bezeichnen: so hat der Ausdruck „ein Pferd“ immer dieselbe Bedeutung, da aber Letztere unbestimmt ist, kann er, je nach Redezusammenhang, sich auf verschiedene Einzeldinge beziehen. Beispiele des zweiten Falls sind die Namen „der Sieger von Jena“ und „der Besiegte von Waterloo“ bzw. „das gleichseitige Dreieck“ und „das gleichwinklige Dreieck“, oder Sätze wie „a ist größer als b“ und „b ist kleiner als a“. Hier ist nämlich dasselbe Gegenständliche in verschiedenen Weisen bzw. mit verschiedenen Bedeutungen gemeint. Wie der Name einen einfachen Gegenstand als Korrelat hat, so hat der Satz einen Sachverhalt als Korrelat. Wie Husserl (1939, 285 ff.) später präzisiert – ein Hinweis in diese Richtung ist allerdings bereits in der zweiten Auflage der Untersuchungen vorhanden, wo der betreffende Text entsprechend verändert wird –, sind a>b und b