Idries Shah’s 1964 book The Sufis, which I read over the holidays, has several interesting passages on language, a couple of which I quote below. The first excerpt concerns the history and use of the protean word Sufism and some of the various terms used to refer to Sufis:
Exactly how old is the word “Sufism”? There were Sufis at all times and in all countries, says the tradition. Sufis existed as such and under this name before Islam. But, if there was a name for the practitioner, there was no name for the practice. The English word “Sufism” is anglicized from the Latin, Sufismus; it was a Teutonic scholar who, as recently as 1821, coined the Latinization which is now almost naturalized into English. Before him there was the word tasawwuf – the state, practice or condition of being a Sufi. This may not seem an important point, but to the Sufis it is. It is one reason why there is no static term in use among Sufis for their cult. They call it a science, an art, a knowledge, a Way, a tribe – even by a tenth-century portmanteau term, perhaps translatable as psychoanthropology (nafsaniyyatalinsaniyyat) – but they do not call it Sufism.
Tarika-sufiyya stands for the Sufi Way; and makes a very good English parallel because tarika stands for Path, as well as a way of doing something, and also conveys the notion of following a path, a line or streak – the Path of the Sufi. Sufism is referred to by different names in accordance with the sense in which it is being discussed. Thus, ilm-al-maarifat (the science of Knowing) may be found; or el-irfan (the gnosis); and the organized Orders or groups tend to be called the tarika. Similarly, the Sufi is known as the Seeker, the Drunken man, the enlightened one, the good, the Friend, the Near One, the dervish, a Fakir (humble, poor in spirit), or Kalandar, knower (gnostic), wise, lover, esoterist. Because there would be no Sufism without the Sufis, the word always applies to people, and cannot be considered as an abstract form, as, say, “philology” or “communism” could mean respectively the study of words or a theory of communalist action. Sufism, then, involves the body of the Sufis as well as the actual practice of their cult. It cannot really mean any theoretical presentation of the Way of the Sufis. There is no theoretical or intellectual Sufism; any more than there can be a Sufi movement, which latter is a redundancy, because all Sufic being is movement and a movement embracing all phenomena of a similar kind.
Later in the book Shah returns to the topic of definition, teasing out the subtleties in different perceptions and categorisations of Sufi and other words applied to the Sufis (albeit with some generalisation).
I’m reluctant to paraphrase much, because Shah’s prose is so deliberately structured and admirably plain:
A dervish is a Sufi. In North Africa, “dervish” is a term of respect, denoting something less than an arif (Knower, Wise One), while a Sufi is looked at askance as someone engaged in mysterious processes. In England, a Sufi is thought to be “a Mohammedan mystic of the pantheistic type,” while a dervish is something weird – what a North African might call a “Sufi.”
Although even kings may sign themselves “Fakir,” the label can be embarrassing in some places. A distinguished Indian academician said: “‘Fakir’ is confused with Hindu jugglers – and worse. I regard you not as a Fakir, but as a Man of the Path.”
Putting the word into a phrase helps to establish the usage. “He is a dervish,” means, “a good, simple person, devoted to truth.” “He is a Fakir,” means, “one who struggles to improve himself, with humility.” “He is a Sufi,” means, “one who follows the Sufi way,” also, “one who has attained progress in the Way.”
The confusion arises because of several factors. Not least of these is the fact that the Sufis do not use labels to denote fixed states or stages, since there is no such thing in Sufism. You can label a pound of butter “butter,” but a Sufi is never entirely a dervish or a Knower. His status changes in relation to the infinite gradations of truth and objectivity.
(Sufis’ aversion to labels is reminiscent of Alfred Korzybski’s warning, in his system of General Semantics, about what he called the is of identity. This philosophical approach to language gave rise to the variety of English known as E-Prime, which proscribes the verb be. Some writers adopted it for occasional use, and it remains an interesting experiment albeit one whose efficacy is disputed.)
Shah writes that the terms Sufi, dervish and Fakir are used in Sufi literature less often than Knower, lover, follower and traveller, because the former set “tend to be externalist labels”. He feels that dictionary definitions fall strikingly short, commenting on these entries in Chambers 1955:
DERVISH: “A member of one of numerous Mohammedan fraternities….”
SUFI: “A pantheistic Mohammedan mystic….”
FAKIR: “A religious (esp. Mohammedan) mendicant, ascetic….”
The meanings of the word “Mohammed” – or even “Mohammedan” – “fraternity,” “pantheism,” “mystic,” “religion,” “mendicant” and “ascetic” are different in Eastern usage, and especially in Sufi application, than in English.
A Persian dictionary, perhaps more poetically, if with less seeming precision, says: “What is a Sufi? A Sufi is a Sufi” – and succeeds in rhyming the entry: Sufi chist? – Sufi Sufi’st. This is actually a Sufi quotation. The compiler does not believe in trying to define the indefinable.
One reason definition can be elusive is the inconstancy of language, though dictionaries based on historical principles accommodate this to a degree. While going through my old Twitter favourites (which are like bookmarks) I rediscovered this wonderful expression of semantic change from historian and poet Matthew Battles:
Something tells me the Sufis would approve.