There are many ways of subjecting myth to the axiomatic method and of answering the questions which determine a philosophy as a presumably consistent and systematic ontology. How are reality and appearance divided? What principles and what specific kind of demonstration are recognized as true and valid?

When the writings of a philosopher – or possibly some of his particular works – are analysed in order to make explicit the principles and methods of proof according to which he draws the boundary between reality and appearance, the historian is naturally led to comparisons. By reconstructing in their individuality some of the systems which constitute the object of his discipline, he meets open discord as well as seeming agreements with other systems, and is obliged to trace their cause to differences regarding a given premise or a given set of premises. Partial, occasional and a posteriori classifications result from such comparisons.

We have met some cases of reductiones ad absurdum that seemed to delineate an a priori framework for the possible solutions which philosophical systems could give to a particular question. Nothing, however, hinted at a possible generalization. Moreover, the formal convergences which are suggested when a contradiction is analysed away threaten to conceal or to prejudge the genuine philosophical issues.

We shall first ask whether a general and truly a priori classification of philosophical systems is possible and on just what conditions that would be.

THE MAIN OBJECT OF THE CHAPTER

The largest part of the present chapter deals with the question: what are the experienced relations which organize perception, and how do they expand so that in agreement with the evolution of the higher central nervous system they culminate both in simulating life-surroundings and in planning actions? The principles of organization will be shown to apply to the data so as to produce appearances of a given kind, and then to reapply to these appearances so as to yield appearances of yet another kind, until, no further innovation being yielded in this way, the process is put to an end and perceptual organization is considered complete. The diverse kinds of appearances that will be thus distinguished (qualities of position and of state, images and representations) are biologically useful only because they are given an objective interpretation. This description is then followed by two discussions. The first is about the kind of symbolic forms that span perceptual organization: have we to accept the claim of these forms to objectivity or must we rather interpret them in terms of subjective signals? The second discussion will concern what is wanting in perceptual organization for making group communication possible.

Though the following description of perceptual organization is systematic, it only aims at sketching the phenomenology of sensible appearances. It must be distinguished from psycho-physiology as well as from developmental psychology and philosophical constitution.

How is it possible for men to communicate their perceptual organization?

Firstly, an analysis of the conditions making this communication possible will reduce the question to the classification of the linguistic categories of individuals. Next a thread for the deduction of these categories will be sought. Linguistic categories will be shown to bypass perceptual organization and to allow its communication. Finally, the deduction itself will be made.

THE SENTENCE AS ULTIMATE CONSTITUENT OF LINGUISTIC COMMUNICATION AND HOW THE CLASSIFICATION OF INDIVIDUALS BY MEANS OF ELEMENTARY SINGULAR SENTENCES MANAGES TO COMMUNICATE THE ORGANIZATION OF PERCEPTION

The question if and how it is possible to communicate perception subdivides into two questions. There is first the general question of the possibility, answered by the general concept of a sentence. Then the particular question of finding a possible correspondence between perceptual organization and language will be shown to amount to classifying the categories of individuals through classifying the elementary singular sentences.

For individuals to be able to communicate their perceptual experiences, two conditions must be fulfilled. Their system or code of communication must be universal, and it must be open. A code is universal if no particular restriction is imposed on the content or on the nature of the messages, and it is open if nothing, in principle, limits their lengths. Natural language is such a code.

These conditions are necessary to build a complete sign, but a complete sign is not a complete message, i.e. a sentence.

In every discipline the order of exposition tends to reverse the order of inquiry.

I began my own inquiry about philosophical systems by reminding myself of what every historian of philosophy knows: namely, that philosophers are divided and that no part of the philosophical enterprise has ever been the object of common agreement. Neither Kant's critique of metaphysics nor the so-called ‘scientific method in philosophy’ has been successful in bringing peace, or even armistice, to the battlefield. Granting the truth of these statements, I had to seek the reason behind them by examining the nature and origin of philosophy. The systematic form of philosophy was elaborated to answer the ontological question posed by the advent of axiomatics when it jolted the unified world of myth, moulded as it was, by natural language. But all collective communication requires a formal recognition of the material that is perceived by means of the linguistic signs (what Saussure called the ‘signifier’). Thus, the question ‘How does language make the unity of the sensible world possible?’ came to be followed immediately by a question about the structure which is already given at the level of our animal perception.

In this essay my exposition reverses the order of my inquiry. The first chapter will deal with perceptual organization, and the second with language and the sensible world. In the third chapter I shall inquire into the nature and the systematic form of philosophy or rather of philosophies by analysing the relations that ontologies bear to axiomatics.

There are civilizations without philosophy, and there are civilizations where philosophy never could, and those where it never can, disentangle itself from a certain bondage to custom, law and social utility. Free philosophy, therefore, as opposed to Church- or state-organized ideology, must have arisen from a revolution in the use of the signs by means of which, for every civilization, language represents the sensible world. What was and still is the force which protects the old order of signs? What is the weak point in the net of the elementary sentences? Where can we expect innovations to subdue this conservative force? The first part of this chapter seeks to answer these questions by analysing the nature of myth: the development of free philosophy will appear as correlated with the development of the axiomatic method. The two go paired together like the Dioscuri. The second part of the chapter will examine the manner in which philosophy and axiomatics give signs a new interpretation.

FROM MYTH TO PHILOSOPHY; THE ROLE OF THE AXIOMATIC METHOD

According to Helmholtz's principle, the signals which constitute perceptual organization are spontaneously interpreted as objective things. An analogous principle is valid for language and the sensible world. The elementary sentences certainly admit of local distinctions between appearance and reality. But appearances are spread out as so many puddles which never conjoin into one and the same river.

The Qurʾān frequently says of sinners anfusahum yaẓlimūn or something similar. The usual interpretation of the expression is ‘They wrong themselves’, but this raises problems. Is it possible for anyone to wrong himself? If it is, are sinners likely candidates for wronging themselves? And does the Qurʾān assert or suggest that these people suffer any wrong? Although these problems have been perceived in varying degrees by medieval and modern commentators (most recently by Kāmil Ḥusayn in Egypt), I believe that a sharper understanding of them is possible when once we have understood what Aristotle had to say about injustice to onself in the Nicomachean Ethics, Book V, chapters 9 and 11. I shall therefore begin with a short account of Aristotle's reasoning and conclusions on this question, to the extent that they can lay the ground for a fresh interpretation of the Qurʾān.

Aristotle distinguishes two main senses of the Greek verb adikein (chs. 1–2). One is a general sense, ‘wronging’ (followed by a direct object for the person wronged, as in English ‘wronging someone’). In this sense we call wrong those acts which go against the aims of the laws, which are ‘to produce and preserve happiness and its components for the political society’ (ch. 1, 1129b 17–19): such acts as desertion and fleeing in war, adultery, assault and slander (1129b 20–3).

In the early centuries of Islam theology and philosophy were regarded as two distinct disciplines, following their own assumptions and intellectual traditions. ‘The science of dialectic’ (ʿilm al-kalām) meant Islamic theology, derived only from the revealed Quiʾān and the Traditions of the prophet Muhammad. Philosophy (al-falsafa) was a ‘foreign’ science based on natural knowledge and largely inspired by ancient Greek philosophy. As time passed theology and philosophy interacted increasingly, with varying relations in different parts of the Muslim world. But we must begin by describing them as separate, as they generally were until about the eleventh century a.d.

THEOLOGY

All teaching, discussion and writing on this subject were conducted in Arabic, the language of the Qurʾān and of other Islamic subjects such as exegesis and jurisprudence. The principal centres of theology were in Iraq and Iran.

After early discussions on conditions for salvation and the moral qualifications of caliphs the first systematic school, the Muʿtazilites, arose in Iraq in the eighth century, about a century after Muhammad's death (632). Their self-description as ‘the party of unity and justice’ announced their central doctrines.

The unity of God (Allāh) was understood very strictly by them, reflecting the emphasis of the Qurʾān. Polytheism and the Christian Trinity were rejected as a matter of course. The Qurʾān could not be eternal beside God but must have been created by the one eternal being.

The primary doctrine of consensus (ijmāʿ) in Sunnite Islam is simply this: that the unanimous opinion of the Sunnite community in any generation on a religious matter constitutes an authority (ḥujja), and ought to be accepted by all Muslims in later times. The importance of the doctrine in theory and practice was recognized particularly in classical Islam, and has been much emphasized, and in some respects exaggerated, by modern orientalists since the studies of Snouck Hurgronje (see § 17).

This being so, it was natural that serious thought should be given to the fundamental question: ‘What is the basis on which this authority of consensus rests?’ Or, what facts about Islam and the world legitimize it? This is the question known in Arabic as ḥujjīyat al-ijmāʿ, ‘the authoritativeness of consensus’. Much was written about it in classical Islam, as we shall see, and more variety has been added by modern Muslims. We should also expect to find some extensive historical study of the question by modern orientalists. But what has been written has generally been too brief, forming a part of a wider survey of Islamic consensus, or of Islamic law as a whole. The question has usually been thought to have been sufficiently dealt with by quoting the famous Tradition: ‘My Community does not agree on an error’, which was the major basis accepted for a thousand years.

These sixteen articles on various aspects of classical Islamic thought have an underlying unity of theme, aptly summed up in the book's title. They form a good part of their author's important contribution to the study of Islamic thought. Most of these articles have been published during the past three decades in different learned journals, in encyclopedias and books containing articles by other scholars. Their inclusion (with some revision) in one volume not only makes them more accessible, but also endows the collection with a historical perspective. The author has arranged the articles as closely as possible in the chronological order of their subjects.

The volume was already in the press when its author, George Fadlou Hourani, died of a heart ailment. It is the last gift of a scholar who had dedicated his life to research and teaching. Born in England in 1913 of Lebanese ancestry, he read philosophy and classics at Oxford, graduating in 1936. His graduate studies in the languages and history of the Near East were undertaken at Princeton University where he obtained his doctorate in 1939. His teaching career began in Jerusalem, Palestine, during the British mandate, where he taught classics, logic and the history of philosophy at the Government Arab College from 1939 until the end of the mandate in 1948. He then returned to England with his Egyptian wife, Celeste, where he spent two years in writing and research.

Normative ethics (akhlāq) in the sense of wise advice for a good and happy life, was written about by Ibn Ḥazm in two well-known books, Ṭawq al-ḥamāma ('The dove's neck ring’) and Mudāwāt an-nufūs ('Cures for souls’). The following article, however, is not concerned with his views on ethics in that sense, but with his answers to fundamental questions of modern philosophical ethics: the meanings of ethical concepts, the sources of our knowledge of them and of values in practice, the theory of moral motivation. In the religious tradition of medieval Islam, to which Ibn Ḥazm for all his individuality belonged, these questions were not marked off as a separate field of knowledge but fell somewhere between theology and law. More precisely, theology provided the framework of doctrines from which the principles of ethics could be derived, and these principles were applied by jurists in working out Islamic law.

Accordingly, we find Ibn Ḥazm's treatment of philosophical ethics mainly in his major work on theology, the Fiṣal, and his major work on jurisprudence, the Iḥkām. Both these works were written in the later years of his life, when his theological and legal position as a Zahirite was settled, and together with other works of the same period they supply a unified theory of ethics. His last work, Mudāwāt an-nufūs, although of a different literary genre, also throws light on this theory and is consistent with it. All these writings, then, will be used to reconstruct his ethics.

With all the breadth of his interests as a theologian, jurist, logician, educator, Ṣūfī, critic of philosophy and foe of Ismaʿilism, Ghazālī's central concern throughout his life (a.d. 1058–1111) may fairly be described as an ethical one: right conduct and the purification of the soul by the individual, as means to a harmonious relation with God and the attainment of everlasting joy. This is of course a religious view of ethics, and one believed to have been learned from God through prophetic revelation and associated divine sources accepted in classical Islam.

The present study will not attempt to treat the entire system of his ethics in its prolific details. We shall be concerned with some of its general aspects. We shall also limit the study to the sphere of conduct, and not deal with the sphere of character and improvement of the soul, important as that subject is in Ghazālī's total ethics.

In order to explain more precisely the object of study and its place in the system as a whole, it will be useful to begin with a summary account of three relevant religious sciences as Ghazālī conceived them: theology (ʿilm al-kalām), ‘law’ (ʿilm al-fiqh) and ethics of character (ʿilm al-akhlāq).

Kalām is defined by Ghazālī simply as the study of God, and it has four principal topics: the existence and fundamental nature of God, His attributes, His actions, and His prophets and revelation.