One of the most fundamental topics in the philosophy of language is signification (Arabic dalÄla), that is, the ability of terms to âreferâ (dalla). In antiquity, as in Platoâs Cratylus, there was a prominent debate as to whether signification occurs naturally or by imposition. If the former were the case, then we could perhaps âread offâ the nature of referents by looking at features of language, or even components of words. Thus Plato suggests that the sound represented by the Greek letter rho naturally signifies motion, which is why the word rhein (flowing) starts with this letter (Cratylus 426câd). But this sort of naturalism plays little or no role in post-Avicennan philosophy, since Avicenna is fully committed to Aristotelian conventionalism. Words, whether written or spoken [T8], have their meanings thanks to arbitrary imposition, apart from occasional exceptions like onomatopoeia, a point made by al-ḤillÄ« at [T28].
This does not mean, though, that all words signify in the same way. To the contrary, Avicenna recognizes three ways of signifying, which he calls congruence (muá¹Äbaqa), containment (taá¸ammun), and implication (luzÅ«m) [T1, T2], followed by BahmanyÄr [T6]. In the case of congruence, the expression exactly matches what is signified, as âhumanâ signifies humans or, more technically, whatever is picked out by the definition of âhuman,â namely rational animal. We speak of containment when an expression is taken to signify only part of that whole meaning, as âhumanâ signifies just animal. Finally, signification by implication involves, as BahmanyÄr puts it [T7], a âtransfer of the mindâ from the core meaning of the expression to an associated idea, as speaking of âroofâ calls to mind walls or a foundation, without which there could not be a roof. As usual, AbÅ« l-BarakÄt suggests a different way of thinking about it, since in his view containment is also a kind of implication, but of something internal to the core meaning as opposed to something external that follows from it [T11]. Broadly speaking, though, authors in our period accepted the threefold division and were happy to work with it, even applying it in fields outside of philosophy, as we shall see.
But that doesnât mean there were no problems for them to discuss. Several tweaks were introduced to the characterization of the three ways of signifying. Fakhr al-DÄ«n said that terms are signified by containment as parts of the overall meaning of the expression, but just insofar as they are parts [T12]. The worry here, also discussed at [T17], is that we need to exclude signification by homonymy, as for instance âpossible (mumkin)â can signify just the specifically contingent (what is neither necessary nor impossible), or the generally possible (whatever is not impossible) where the former is a part of the latter. Al-AbharÄ« and al-KÄtibÄ« expand this even to congruence [T23, T25]. To use the standard example, we should not say without qualification that the Arabic word Ê¿ayn signifies the eye by congruence, because it can also signify a source of water. A further warning is introduced by Ibn KammÅ«nÄ: we should not confuse signification by containment with the idea that a more general expression signifies something more specific [T26]. For instance, âhumanâ signifies animal because this is part of the nature of human, but the reverse is not the case. Thus every expression for a species signifies by containment both the genus and specific difference of that species [T14], whereas the expression for a genus does not signify the species that falls under that genus.
One reason Avicenna and our later authors are so interested in signification is that we need to keep careful track of it when doing science.1 This is something we already see in al-FÄrÄbÄ«, who warns against the mistake of using in demonstrations a term like âhuman animal.â This would be redundant, since âhumanâ already signifies animal by containment.2 A more fundamental methodological concern is raised by Avicenna at [T3, T4, T5], namely that signification by implication should not be employed in science. This is because whatever is implied by an expression lies outside its quiddity, and in science we are only interested in quiddities. Here it would be useful to note the ambiguity of the word used for âimplication,â namely luzÅ«m. In a narrow, technical sense, that which is lÄzim (âimpliedâ) can refer to necessary concomitants. For example âhaving 180 degrees of internal anglesâ necessarily belongs to triangle, so this property is among the lawÄzim of triangle. This is the sense of the word luzÅ«m that appears most frequently in the rest of the present volume. More broadly, and in the sense relevant in this chapter, an âimplicationâ can be anything that comes naturally to mind when an expression is heard. This could be because two things have an extramental causal relationship, like âwallâ and âroof,â or even just an accidental association, like âenjoys cheeseâ and âFrenchman.â It seems intuitive that these should have no place in science, whereas lawÄzim in the narrower sense certainly do. In [T24] al-ṬūsÄ« seems to recognize this ambiguity when he suggests that some implications might be retained in science.
Our authors duly consider the exact rationale that led Avicenna to exclude implications from science. One idea, proposed by al-SÄwÄ« [T9], is that immediately implied concepts would have further implications, which have then further implications, leading to a regress: roof implies wall, wall implies foundation, foundation implies earth, earth implies cold, cold implies ice cream, etc. But Fakhr al-DÄ«n rejects this explanation, on the grounds that we could stop the regress by simply limiting ourselves to the âobviousâ implications [T16]. Instead, he says that the problem is that this limitation would be vague. Some people might find it obvious that roof implies foundation, others not; presumably no one finds themselves thinking of ice cream when hearing âroof,â but you never know. Interesting here is his point that we need to consider what is uniformly communicated by an expression, in other words, what meanings would reliably be understood by anyone who hears the expression. This close link between signification and successful communication is also thematized at [T11, T27].
This is arguably also the intuition behind al-GhazÄlÄ«âs explanation for the exclusion of implications, namely that they simply were not intended by the âname giverâ who first imposed the expression for a given meaning [T10]. Similarly al-ÄmidÄ« [T18] and Ibn al-Ḥajib [T20] say that unlike congruence and containment, implication is not âverbalâ or âlinguisticâ signification. Even if the word âroofâ calls walls to mind, this is not part of what the word signifies. Rather, the signification results only from an operation of the mind, as also noted by al-KhÅ«najÄ« [T19]. Along the same lines, Fakhr al-DÄ«n distinguishes between signification that is âby impositionâ and that is âby reason (Ê¿aqlÄ«)â [T12, T15]. However, he classifies even signification by containment, not just signification by implication, as being âby reason.â This is because a mental operation is needed to get to the part of a meaning. For instance when animal is signified by âhuman,â this isnât part of the imposed meaning of the expression.
The influence of Avicennaâs scheme is shown by the fact that it was deployed beyond logic and philosophy of language. We see this even in Fakhr al-DÄ«n, who suggests that the three modes of signification belong to a grammatical theory that spans across all natural languages [T13]. But the most striking illustration comes in juridical authors, who use Avicennaâs distinction to discuss various kinds of legal reasoning. The argument e contrario (for example: from âtax is owed on free-grazing sheepâ we infer that no tax is owed on stall-fed sheep) is stated not to be any one of the three modes of signification [T21].3 By contrast, reasoning by analogy (as when, from âdate wine is forbiddenâ we infer that grape wine is forbidden) is a case of signification by implication [T22]. And, in an echo of Fakhr al-DÄ«nâs point about the varied effects of this sort of signification from person to person, we are told that this need not mean that the mind inevitably turns to what is implied, just that it can [T29].
Texts from: Avicenna, BahmanyÄr, al-SÄwÄ«, al-GhazÄlÄ«, AbÅ« l-BarakÄt al-BaghdÄdÄ«, Fakhr al-DÄ«n al-RÄzÄ«, al-ÄmidÄ«, al-KhÅ«najÄ«, Ibn al-ḤÄjib, TÄj al-DÄ«n al-UrmawÄ«, al-Bayá¸ÄwÄ«, al-AbharÄ«, al-ṬūsÄ«, al-KÄtibÄ«, Ibn KammÅ«na, al-QarÄfÄ«, al-ḤillÄ«, Shams al-DÄ«n al-Iá¹£fahÄnÄ«
[T1] Avicenna, ShifÄʾ, Maná¹iq, Madkhal, 42.15â43.15
[the three types of signification]
You know that whenever the expression âmovedâ signifies, there must be a mover; and whenever the expression âroofâ signifies, there must be a foundation. But we do not say that what is understood and signified by the expression âmovedâ is the mover, or that what is understood and signified by the expression âroofâ is the foundation. This is because the meaning of âsignification of an expressionâ is that the expression is a name (ism) for this meaning, according to [43] primary intention. And if there is a further meaning associated with this meaning in an extrinsic way, with the mind becoming aware of this other meaning simultaneously as it becomes aware of that first meaning, then the expression does not signify this [extrinsic meaning] primarily. [â¦] [43.5] The meaning given to âsignification (dalÄla)â is also twofold; one is primary, the other secondary. The primary one is like when we say âanimal,â for it signifies [primarily] the whole of âsensitive body having a soul.â The secondary one is like the way [âanimalâ] signifies body [alone], since the meaning of âbodyâ is necessarily included (muá¸ammanun) in the meaning of animality. For whatever signifies animality embraces the meaning of body, and not in such a way that it would indicate it extrinsically. [â¦] [43.12] If we wish to condense and summarize all this, we would assume the signification of expressions to occur in three ways: (a) signification by congruence (dalÄlat muá¹Äbaqa), as âanimalâ signifies the whole of âsensitive body having a soulâ; (b) signification by containment (dalÄlat taá¸ammun), like the expression âanimalâ signifies body; and (c) signification by implication (dalÄlat luzÅ«m), like âroofâ signifies the foundation.
[T2] Avicenna, Maná¹iq al-MashriqiyyÄ«n, 14.16â15.6
[the three types of signification]
The types of signification for an expression are three: signification by congruence, by containment, and by implication (iltizÄm), the latter being a transition by way of the meaning (wa-huwa al-naql min á¹arÄ«q al-maÊ¿nÄ). Signification by congruence is like the way the expression âhumanâ signifies rational animal. Signification by containment is like the way âhumanâ signifies animal and rational, for each of these two is a part of what âhumanâ signifies by congruence. Signification by implication is like the way âcreatedâ signifies creator, âfatherâ signifies son, âroofâ signifies wall, and âhumanâ signifies âcapable of laughter.â For [the expression] primarily (awwalan) signifies, with signification by congruence, the meaning that it [15] signifies in the first instance (awwalan). This meaning is then accompanied by another one, so that the mind also makes a transition to that second meaning, which is associated with the first meaning and accompanies it. Signification by congruence and signification by containment have in common that neither of them is a signification of something external to the thing [signified]. Signification by containment and signification by implication have in common that each of them depends on the primary signification.
[T3] Avicenna, IshÄrÄt, 53.5â10
[on signification by implication]
When we say âsuch-and-such an expression signifies such-and-such,â we mean by this the way of congruence or of containment, not the way of implication. How could it be otherwise, since what is signified by way of implication (bi-á¹arÄ«q al-iltizÄm) is without limit (ghayr maḥdÅ«d)? Moreover, if what is signified by way of implication were intended, then something non-constitutive would be suitable to signify the âwhat it is,â like âcapable of laughter,â which signifies rational animal by way of implication. But it is unanimously agreed that this sort of thing is unsuitable in reply to the question âwhat is it?â
[T4] Avicenna, ShifÄʾ, Maná¹iq, BurhÄn, 150.6â8 [trans. Strobino, mod.]
[implication]
Things are necessary in two ways: in implication (fÄ« l-luzÅ«m), without being necessary to one another in substance and natureâthese are extrinsic implicates (lawÄzim khÄrija), and we have explained earlier that they are not useful to obtain certain knowledgeâand necessary in substance and nature, that is, things that are essential (bi-dhÄtihÄ).
[T5] Avicenna, ShifÄʾ, IlÄhiyyÄt, 177.13â17
[implication]
We also say that the genus is predicated of the species in such a manner that it is a part of its quiddity, whereas it is predicated of the specific difference in such a manner that it is an implicate (lÄzim) of it, not in such a manner that it is a part of its quiddity. For example âanimalâ is predicated of âhumanâ in such a way that it is a part of its quiddity, but predicated of ârationalâ in such a manner that it is an implicate of it, not in such a manner that it is a part of its quiddity. By ârational (nÄá¹iq)â we mean something that has rationality (nuá¹q) and something that has a rational soul. But our saying ârationalâ does not by itself show whether that thing is a substance or not a substance; yet it is necessarily implied that this thing can be nothing but a substance, a body, and sensitive. These things are said of it [sc. ârationalâ] in the way that something implied (lÄzim) is said of that which necessarily implies [it]; for they [i.e. substance, body, and sensitive] are not intrinsic to that which is understood by ârational,â that is, something that has rationality.
[T6] BahmanyÄr, Taḥṣīl, 13.3â6
[threefold classification with relations between logical predicables]
There are three ways an expression signifies a meaning: the first is called congruence (muá¹Äbaqa), as âanimalâ signifies the species below it. The second one is by means of containment, as âhouseâ signifies wall and its definition, and as the species signifies the genus. The third is signification by implication, as âroofâ signifies wall and as the specific difference signifies the genus.
[T7] BahmanyÄr, Taḥṣīl, 16.9â15
[signification and implication]
We say: in reply to [the question] âwhat is it?â anything is suitable that expresses the state of commonality (sharika), but not the state of a single [shared element] (infirÄd), for example âanimalityâ for âhumanâ and âhorse,â but not âsensitiveâ (ḥassÄs) for âhumanâ and âhorse.â This is because âsensitiveâ does not signify the whole quiddity that human and horse have in common, given that, even if âsensitiveâ signifies some essential meaning, namely the fact that something has sensation (ḥiss), by way of congruence, it signifies the further thing that has sensation [only] by way of implication, not by way of containment. The difference between implication and containment is this: the roof implies the wall, and does not contain it, whereas the house both implies the wall and contains it. The meaning of âimplication (iltizÄm)â is that one gets to know something external by way of transferring the mind to it.
[T8] BahmanyÄr, Taḥṣīl, 39.1â5
[written, verbal, and mental significantion]
It is correct that the written and verbal expressions differ [from person to person], because they are natural for what is in the soul, but are due to imposition (waá¸Ê¿). As for the trace (athar) that is in the soul, it does refer naturally to things in extramental reality. Written [expressions] are posited significations (dalÄlat waá¸Ê¿iyya) that differ according to the impositions of the letters used in discourse (ḥurÅ«f al-kalÄm). Verbal expressions are posited significations that differ according to the impositions for the conceptions (taá¹£awwurÄt) of the soul. The conceptions of the soul are an instinctive (gharÄ«ziyya) signification for extramental things.
[T9] Al-SÄwÄ«, Baá¹£Äʾir, Maná¹iq, 33.5â8; 33.17â20
[studying meanings by studying expressions]
It has already been shown that the logician examines meanings. But if he confines himself to the investigation of expressions, their states and their divisionsâgiven that he really has to examine these, because of [the relation] between an expression and a meaningâthen this will spare him from having to start over with getting to know the states and divisions of meanings. For the expressions parallel the meanings. We say, then, that an expression signifies a meaning in three ways. [â¦]4
[implication is not used in science]
[33.17] It is as if [implication] were not a verbal signification; rather, the mind is transferred from one meaning, which the expression signifies by imposition, to another meaning which is associated with it and proximate to it. In the sciences, signification by congruence and by containment can be used, but not signification by implication, because it is unrestricted, since the implicates (lawÄzim) may have further implicates, and so on without end.
[T10] Al-GhazÄlÄ«, MiÊ¿yÄr al-Ê¿ilm, 72.6â19
[threefold classification of signification]
We say: expressions signify their meanings in three different ways. The first way is signification with regard to congruence, as a name is posited for the thing, for instance as the expression âwallâ for the wall. The second way is by way of containment, as the expression âhouseâ signifies the wall, and the expression âhumanâ signifies animal. And likewise as every maximally specific attribute (waá¹£f akhaṣṣ) signifies the most general substantial (jawharÄ«) attribute. The third way is the signification by way of implication and following, as the expression âroofâ signifies the wall, because it follows from it [â¦].5
[signification in definitions]
[72.15] In definitions (taÊ¿rÄ«fÄt) we [only] consider signification by congruence and by containment. Signification by implication is not considered, since it has not been laid down by the giver of language, unlike the other two. For what is signified by [implication] is unlimited and unrestricted, given that the implicates of things and the implicates of their implicates are unregulated and unrestricted. So this would lead to the result that the expression signifies an infinity of meanings, which is absurd.
[T11] AbÅ« l-BarakÄt, MuÊ¿tabar, vol. 1, 8.7â186
[the threefold classification in terms of communication]
(a) The expression may signify for the hearer the meaning intended by the speaker, as when he understands ârational animalâ from the expression âhumanâ; this is called âsignification by congruence.â (b) Or [the expression] may signify a meaning that is contained in it and [part] of its totality, as the expression âhumanâ signifies animal or rational; for in signifying [human] is signification of each of these two; this is called signification by containment. (c) And one may understood from [the expression] a meaning which is not the intended meaning, and which is not [part] of its totality, yet is an implicate of it and inseparably connected to it. This is called signification by implication, as the expression âmovedâ signifies the meaning of mover, and âroofâ signifies wall; for moved cannot be separated from mover, even though it is not itself the mover, and what is understood by âmoverâ is no part of what is understood by [âmovedâ]; nor can the roof be separated from the wall, even though it is neither the wall nor part of the wall.
[alternative way of classifying]
It would also be right to make there be just two kinds of signification: first congruence, and then the other two would both be placed under implication, one of them in a more proper sense, namely containment, the other signification by âimplicationâ in the sense of following. For the part is understood as implying understanding of the whole.
[T12] Al-RÄzÄ«, Maḥṣūl, Fiqh, vol. 1, 219.5â220.8
[threefold classification]
The signification of an expression may be considered either in relation to the entirety of what it names (tamÄm musammÄhu); or in relation to what falls within the meaning, insofar as it is such; or in relation to what falls outside the meaning, insofar as it is such. The first is congruence, the second containment, and the third implication.
Remarks. First: signification based on imposition (dalÄla waá¸Ê¿iyya) is signification by congruence. The other two [significations] are based on reason (Ê¿aqliyya), since when an expression is imposed for a notion (musammÄ), the mind turns from the meaning to what is impicated (lÄzim) by it. If its implicate falls within the meaning, it is a case of containment, if outside, then implication (iltizÄm). [220] Second: we said regarding containment that it is the signification of the expression of a part of the meaning, âinsofar as it is such,â as a precaution against the expression signifying a part of the meaning by congruence but only in a homonymous manner. And likewise for what we said about implication. Third: with regard to signification by implication, external implication (al-luzÅ«m al-khÄrijÄ«) is not considered. For substance and accident are mutually implicating, yet the expression used for one of them is not applied to the other. On the other hand, in the case of two opposites that exclude each other (á¸iddÄni mutanÄfiyÄni), an expression that signifies one may be used for the other, like Godâs statement: âlet harm be requited by an equal harmâ (QurʾÄn 42:40). But [in such a case] it is obviously mental implication that is being considered. Furthermore, this implication is [only] a necessary condition, not a sufficient condition (shará¹, lÄ mÅ«jib).
[T13] Al-RÄzÄ«, Sharḥ al-IshÄrÄt, vol. 1, 22.8â12
[connection to grammar]
[Avicenna] said: â[it is indispensable for the logician to consider the aspect of the absolute expression], insofar as it is not restricted to the language of a certain group of people, except in rare cases.â This means that inquiry into languages can be based on reason (Ê¿aqlÄ«), as when we say that significative expressions are name, verb, or particle, and that they signify by congruence, containment, or implication; for these are general inquiries into all languages. Or it can be a specific inquiry into the language of just one people, without considering another one. The first is what logician must deal with, the second falls outside logic.
[T14] Al-RÄzÄ«, Sharḥ al-IshÄrÄt, vol. 1, 88.14â16
[correlation of signification to logical notions]
It has been shown that the expression that signifies the true reality (ḥaqīqa) signifies the genus and specific difference by containment, whereas the expression that signifies the specific difference signifies the genus only by implication.
[T15] Al-RÄzÄ«, NihÄyat al-Ä«jÄz, 87.3â88.8
[signification by imposition and by reason, cf. T12]
[Signification] is either based on imposition (waá¸Ê¿iyya) or on reason (Ê¿aqliyya). That based on imposition is the signification of expressions for the meanings they were posited for, like the way âstone,â âwall,â âsky,â and âearthâ signify what they name. There can be no doubt that this is based on impositionâotherwise it would impossible for their significations to differ according to their different impositions. That based on reason is either when [what is signified] enters into what is understood (mafhÅ«m) by the expression, as âhouseâ signifies roof, which is a part of what is understood by âhouseââthere can be no doubt that this is based on reason, since it would be impossible for an expression to be posited for a composite true reality (ḥaqÄ«qa) without including [88] its partsâor when [what is signified] lies outside [what is understood by the expression], as the expression âroofâ signifies âwall.â Since it is usually impossible for the roof to be separate from wall, the expression that conveys the true reality of roof [also] conveys wall, by means of its signification of the former [i.e. roof]. Therefore this is also a signification based on reason.7 The master [Ê¿Abd al-QÄhir al-JurjÄnÄ«] states, concerning what we were just talking about: there is a brief explanation (Ê¿ibÄra mukhtaá¹£ara) here, namely to [distinguish between] the meaning (al-maÊ¿nÄ) and the meaning of the meaning (maÊ¿nÄ l-maÊ¿nÄ). The âmeaningâ is what is understood from the surface expression (áºÄhir al-lafáº), that is, is understood from it immediately. The âmeaning of the meaningâ is: once the meaning is understood from the expression, then this meaning conveys another meaning.8
[T16] Al-RÄzÄ«, Maná¹iq al-Mulakhkhaá¹£, 20.15â21.5
[why signification by implication should not be used in science]
The signification by implication is to be renounced in the sciences, not because it is said to be based on reason (Ê¿aqliyya); otherwise [21] this would be refuted by the case of containment [which is also said to be based on reason]. Nor because the implications are infinite, since the obvious ones (al-bayyina minhÄ) are finite. Rather it is because if the expressionâs signification of the implication is considered to be obvious, then this would be something that differs from one person to another. Accordingly what is signified would not be precise. If on the other hand it is not considered [obvious], this would be absurd (muḥÄl), since the purpose of pronouncing the expression is to let [other people] understand the meanings (ifhÄm al-maÊ¿ÄnÄ«). If this doesnât happen, then the expression is of no use (lam yakun al-lafẠmufÄ«dan).
[T17] Al-RÄzÄ«, Sharḥ al-IshÄrÄt, vol. 1, 32.12â33.3
[signification of part of a meaning]
You should know: it might be supposed that, whenever an expression signifies a meaning, then somehow, it signifies a part of that meaning by containment. But this is false. For it could be that the expression signifies [both] the meaning and its part by congruence, in a homonymous manner. For instance the case of âpossibleâ (al-mumkin), since it signifies both the specific possible (al-mumkin al-khÄṣṣ) and the general possible (al-mumkin al-Ê¿ÄmmÄ«), which is part of what is understood by the specific possible, so that it signifies both of them homonymously (ishtirÄk al-ism).9 Rather, the condition for signification by containment is that the expression signifies [32] a part of the meaning not primarily and in itself, but secondarily and accidentally. With [signification by containment] we mean that [the expression] signifies the part as a consequence (tabiÊ¿an) due to its signifying the whole (li-dalÄlatihi Ê¿alÄ l-kull), as the expression âtriangleâ signifies figure, since âtriangleâ does not signify figure in itself, but [figure] is signified because it is included under triangle (li-indirÄjihÄ« taḥta l-muthallath).10
[T18] Al-ÄmidÄ«, IḥkÄm, vol. 1, 32.4â16
[âverbalâ vs ânon-verbalâ signification]
About the parts of its signification: its signification is either (a) verbal (lafáºiyya) or (b) non-verbal. (a) âVerbalâ means that [signification] is considered in relation to (a1) the whole meaning, or (a2) part of the meaning, for which the expression has been imposed. (a1) The first case is signification by congruence, as the expression âhumanâ signifies its meaning. (a2) The second case signification by containment, as expression âhumanâ signifies what is in its meaning, like animal and rational. Congruence is more general than containment, since what is signified can be simple, with no parts. (b) âNon-verbalâ is signification by implication, where an expression has a meaning and that meaning has an external implicate. As soon as one understands from the expression what it signifies, the mind makes a transition from what is signified by the expression to its implicate. Without the ability to perform such a mental transition, this implicate would not be understood. Even though signification by containment shares with [implication] that both require rational inquiry (naáºar Ê¿aqlÄ«) in order to know the implicate, in the case of implication, and the part, in the case of signification by containment, there is still a difference. Namely that in the case of containment, one gets to know a part internal to what is signified by the expression, whereas in the case of implication one gets to know something external to what is signified by the expression. This is why signification by containment is verbal, unlike signification by implication.11
[T19] Al-KhÅ«najÄ«, Kashf Al-AsrÄr, 11.3â7; 12.13â14
[signification by implication is a mental process, and the outcome of imposition of meaning]
The [sort of] implying (luzÅ«m) that is considered to be an implication (iltizÄm) [in the strict sense] is mental implication (luzÅ«m dhihnÄ«). Otherwise, uttering the expression would convey no understanding of the external meaning at all, since there would be no imposition for it, and no transition in the understanding from its imposed [meaning] to [the further, external meaning]. So [the external meaning] would not be signified, since what is meant by saying âan expression signifies a meaningâ is that someone who knows the imposition will understand [the external meaning] in the moment of the utterance of [the expression]. External implication, by contrast, is not considered [as implication in the strict sense], because the expression might signify something that is not implied in external reality by what it names. [â¦]
[12.13] Implication is a signification based on imposition (waá¸Ê¿iyya), since the expression signifies the implicate of whatever it names, due to the fact that it has been imposed for something that implies [the implicate]. What is intended by âsignification based on impositionâ could be imposition either with or without an intermediary (ibtidÄʾan aw bi-wÄsiá¹a).
[T20] Ibn al-ḤÄjib, Mukhtaá¹£ar al-muntahÄ, 220.1â221.4
[linguistic vs. non-linguistic signification]
It belongs to the benevolence of God to have created the linguistic principles (iḥdÄthu l-mawá¸Å«Ê¿Ät al-lughawiyya). Let us speak about their definition and types. We begin with their imposition, and how they may be known. First, every expression has been posited for some meaning. Second, [expressions] are divided into two types: simple and compound. The simple one is [expressed] by a single word. It is said [by the logicians] that it is posited for a meaning, and none of its parts signifies.12 [â¦] [221.1] The simple [expression] is classified into noun, verb, and particle. Its verbal signification is, with regard to its complete meaning, signification by congruence; and with regard to a part of it, signification by containment. Non-verbal (ghayr lafáºiyya) [signification] is implication. And one may say: [it is an implication] if [the signification] is by means of the mind (dhihnÄ«).
[T21] TÄj al-DÄ«n al-UrmawÄ«, al-ḤÄá¹£il min al-Maḥṣūl fÄ« uṣūl al-fiqh, vol. 1, 440.4â813
[signification in law: reasoning from a contrary case]
If [the statement qualified by an attribute] signified that a ruling is not true for a statement with another attribute, it would signify this either by congruenceâbut this is wrong, since the denial that a legal ruling applies for one of two attributes, is not the same as affirming it for another attributeâor by implicationâwhich is also wrong, because the absence of an implied affirmation for one of two attributes does not amount to denying it for the other.
[T22] al-Bayá¸ÄwÄ«, MinhÄj al-wuṣūl ilÄ Ê¿ilm al-uṣūl, 69.6â8
[signification in law: reasoning by analogy]
It is said: if it did signify [i.e. if an explicit ruling signified a further ruling for an unmentioned case], it would do so either by congruence (muá¹Äbaqatan) or by implication (iltizÄman). We say that it signifies by implication, since it has been established that the assignment (tartÄ«b) signifies the causality (Ê¿illiyya). When the cause (Ê¿illa) is eliminated, the caused likewise.
[T23] Al-AbharÄ«, Kashf al-ḥaqÄʾiq, 19.5â12
[homonomy problem, cf. T12]
The ImÄm [Fakhr al-DÄ«n al-RÄzÄ«] said: congruence is for the expression completely to signify what it names; containment is for it to signify a part of what it names, just insofar it is a part. With the latter restriction [sc. âjust insofar as it is a partâ] we take precaution against expressions that signify part of what they name by congruence, but homonymously. Implication, finally, is for the expression to signify the implicate of what it names, just as such.
We say: this restriction is also necessary for congruence. So one should [instead] say: congruence is for the expression to signify what it names just insofar it has been posited for it. This as a precaution against the expression that is homonymous between whole and part signifying the part by containment.
[T24] Al-ṬūsÄ«, Sharḥ al-IshÄrÄt, 181.7â9
[against al-RÄzÄ« on signification by implication]
[Avicenna: When we say âsuch-and-such an expression signifies such-andsuch,â we mean by way of congruence or of containment, not by way of implication.] He means signifying the quiddity or the meaning of a name, not absolute signification (dalÄla muá¹laqa), as the commentator [i.e. al-RÄzÄ«] understood it, who went so far as to assume that signification by implication is to be renounced across the board.
[T25] Al-KÄtibÄ«, Asʾilat, 143.12â21
[modifications to al-RÄzÄ«âs definition of congruence in T12]
The ImÄm [Fakhr al-DÄ«n al-RÄzÄ«] said: the expression is considered either in relation to the entirety of what it names, which is congruence, etc.
We say: this calls for further inquiry. First: when the expression is considered in relation to the entirety of what it names, this is not congruence; and when in relation to its part, this is not containment; and when in relation to its external implicate, this is not implication; rather, these are the signification of the expression. This being so, he should have said âsignification of the expressionâ [not just âthe expressionâ]. Second: with regard to congruence, he ought to have added, âinsofar as it is what it names,â just as he said for containment âinsofar as it is a part of it,â and for implication, âinsofar as it is such.â [â¦] [143.20] [This qualification] has to be introduced for congruence in order to exclude signification of the part by containment, in the case of [an expression] imposed for both the whole and the part; and of the implicate by implication, in the case of [an expression] imposed for both the implicate and what implies it.
[T26] Ibn KammÅ«nÄ, Sharḥ al-TalwīḥÄt, vol. 1, 22.5â7
[the general does not signifiy the specific]
The general (Ê¿Ämm), for instance âanimal,â has no signification of the specific (khÄṣṣ), for instance a specific term like âhumanity,â because none of the three [kinds of] signification apply. The general [expression] has not been posited for the specific, so that it would be signified by congruence; the specific is not a part of it so that it would be signified by containment; and it is not an implicate of it so that it would be signified by implication.14
[T27] Al-QarÄfÄ«, Sharḥ Tanqīḥ, 26.15â19
[the classification in terms of communication]
[Signification] is of three kinds. Signification by congruence, where the hearer (sÄmiÊ¿) understands the totality of what is named by what the speaker has said. Signification by containment, where the hearer understands a part of what is named by what the speaker has said. Signification by implication, where the hearer understands the obvious implicate of what is named (lÄzim al-musammÄ al-bayyin) by what the speaker has said, and this is the implicate he had in mind. The first is for instance to understand the sum of two fives from the expression âten.â The second is for instance to understand five and its definition from the expression [âtenâ]. The third is for instance to understand evenness from the expression.
[T28] Al-ḤillÄ«, QawÄÊ¿id, 195.6â9
[the theory applies only to signification by imposition]
You should know: the expression may sometimes signify according to nature, as aḥ signifies trouble in the chest, or may do so according to reason, as âsound (á¹£awt)â signifies someone making a sound, or it may do so according to imposition. What is intended here is [only] the latter.
[T29] Al-Iá¹£fahÄnÄ«, Sharḥ al-MinhÄj, vol. 1, 293.1â5
[commentary on al-Bayá¸ÄwÄ« in T22, on legal reasoning by analogy]
The author [sc. al-Bayá¸ÄwÄ«] responds to the question at stake that [the legal ruling of the unmentioned case] is indeed signified by implication, since it is established that the assignment of the ruling to the attribute indicates the causality (Ê¿illiyya) of the attribute (waá¹£f) with regard to the ruling. And the absence of the cause assuredly implies the absence of the effect. As for what [the opponent] says, namely that the mind does not turn to it, this is rejected and not conceded. On the view of the legal theorists (uṣūliyÄ«n), for signification by implication it is sufficient that conviction (jazm) arises through what has the implication (malzÅ«m), by conceptualizing simultaneously both the implicate (lÄzim) and what implies it (malzÅ«m). It does not require that the mind [actually] turns from the latter to the former.
See further R. Strobino, âPer Se, Inseparability, Containment and Implication. Bridging the Gap between Avicennaâs Theory of Demonstration and Logic of the Predicables,â Oriens 44 (2016), 181â266.
See al-FÄrÄbÄ«, Commentary on On Interpretation, in M.T. DÄneshpažūh (ed.), al-Maná¹iqiyyÄt li-l-FÄrÄbÄ« (Qom: 1988), vol. 2, 175.15â21.
For this form of argument in relation to Avicennaâs account of signification see N. Kalbarczyk, âIn the Footsteps of Avicenna? The UṣūlÄ« Debate on the Argumentum e contrario,â in P. Adamson (ed.), Philosophy and Jurisprudence in the Islamic World (Berlin: 2019), 53â65, and for connections between the account and legal thought more generally, N. Kalbarczyk, Sprachphilosophie in der islamischen Rechtstheorie. Die avicennische Klassifikation der Bezeichnung bei Fakhr ad-dÄ«n ar-RÄzÄ« (Leiden: 2018).
There follows the standard list of three ways with the familiar examples of human and house.
Similar are al-GhazÄlÄ«, al-Mustaá¹£fÄ, vol. 1, 92.8â93.8 and al-GhazÄlÄ«, KitÄb Miḥakk an-naáºar, 73.6â74.1.
For the immediate context see the first paragraph of T11 in our chapter on the Semantic Triangle.
Similar at Fakhr al-DÄ«n al-RÄzÄ«, Sharḥ al-IshÄrÄt, vol. 1, 33.10â11. For other passages designating congruence as âby impositionâ and containment and implication as âby reason,â see al-ṬūsÄ«, Sharḥ al-IshÄrÄt, vol. 1, 139; and al-Iá¹£fahÄnÄ«, al-KÄshif Ê¿an al-Maḥṣūl, vol. 2, 7.9â15.
Equivalent (with minor differences) to Ê¿Abd al-QÄhir al-JurjÄnÄ«, DalÄʾil al-iÊ¿jÄz, 262, last lineâ263.3.
Generally possibility just means what is not impossible, while specific possibility means contingency, that is, what is neither impossible nor necessary. See Avicenna, ShifÄʾ, Maná¹iq, QiyÄs, 162.1â3.
Similar remark at al-ṬūsÄ«, Sharḥ al-IshÄrÄt, vol. 1, 139.3â5.
Similar passages at JamÄl ad-dÄ«n Ibn al-ḤÄjib, KitÄb MuntahÄ al-wuṣūl wa-l-amal fÄ« Ê¿ilmay al-uṣūl wa-l-jadal (Beirut: DÄr al-Kutub al-Ê¿ilmiyya, 1985 [1405]), 17.6â7 and Ibn al-ḤÄjib, Mukhtaá¹£ar al-muntahÄ fÄ« l-sÅ«l wa-l-amal fÄ« Ê¿ilmay al-uṣūl wa-l-jadal, ed. NaáºÄ«r ḤamÄdÅ« (Algiers: DÄr Ibn Ḥazm, 2006), 221.
Ibn al-ḤÄjib, MuntahÄ al-wuṣūl, 16, last line.
Note that this is not the same al-UrmawÄ« who is cited elsewhere in this volume, SirÄj al-DÄ«n al-UrmawÄ«.
Similar is Quá¹b al-DÄ«n al-ShirÄzÄ«, Sharḥ Ḥikmat al-ishrÄq, 36.14â¯f.