Dominicus Gundissalinus (c. 1115 – post 1190) is the author of over twenty translations from Arabic into Latin, which are of considerable importance to philosophy and science.Footnote 1 Working alone or with two collaboratorsFootnote 2 – i.e. Abraham ibn DaudFootnote 3 and Johannes HispanusFootnote 4 – Gundissalinus translated works – such as Avicenna's De anima and Metaphysica, al-Fārābī's De scientiis or Ibn Gabirol's Fons vitae – that were received with enthusiasm in the Latin world, facilitating and paving the way for the thirteenth century philosophical transition to Aristotelianism.
These translations were also used by Gundissalinus as the speculative basis for five personal philosophical works, where the author blends some of the principal results of Arabic, Jewish and Latin enquiries into a new doctrinal system quite autonomous from its sources. At first glance, these Gundissalinian treatises might be viewed as nothing more than a multiplicity of textual quotations deriving from the works that Gundissalinus translated and studied. However, upon closer inspection the author's personal theoretical perspectives become more perspicuous: whereas on the one hand the textual proximity of Gundissalinus’ treatises to their sources is undeniable, and this counts as a typical feature of his modus operandi, on the other the Toledan philosopher accomplishes a real synthesis of some speculative perspectives that often not only differ from one another but are even opposites, an example of this being Avicenna's ontology and universal hylomorphism.
Thus, the close relationship between the Gundissalinian text and its main sources allows us to understand some interesting nuances of Gundissalinus’ hermeneutics with regard to his Arabic sources, leading to deeper insight into both his philosophical thought and his translation method. In this contribution, I shall examine a prime example of these textual connections, namely the Gundissalinian quotation, in his De processione mundi,Footnote 5 of Avicenna's demonstrations of the necessary being.Footnote 6 I shall focus on the main philosophical issue directly connected to the textual modifications that Gundissalinus introduces in his cosmological treatise, setting aside issues concerning the Latin rendering of the Arabic and differences between the two Gundissalinian versions of this passage and Avicenna's original text.Footnote 7
AVICENNA'S DEMONSTRATIONS OF THE UNRELATED UNIQUENESS OF THE NECESSARY BEING
Avicenna dedicates Chapters 6 and 7 of his Liber de philosophia prima seu Metaphysica to the notorious discussion on the ontological difference between God and the creatures according to the theory of necessary and possible being.Footnote 8 The first book of the Metaphysica presents the first concepts and main elements of Avicenna's ontology,Footnote 9 amongst which there is the famous differentiation between essence and existence, so influential in later Latin speculations.Footnote 10
The two chapters we are going to examine play a crucial theoretical role within the philosophical system and within the Metaphysica itself. Here, Avicenna carries out an analysis of the necessary and possible beings and demonstrates through various important logical arguments the unrelated uniqueness and unity of the necessary being, totally uncaused, self-sufficient, and on the basis of which a distinction between essence and existence is declared impossible.Footnote 11 The opposite equivalent in this ontological bipartition is the possible being, insufficient to itself, as it always requires a cause that distinguishes its possibility of being from its possibility of not-being. Once this process is actuated, the possible being becomes a necessary being. However, it is not a necesse esse per se, that is, in itself, but a necessary being per aliud, i.e. depending on a cause that discriminates between his possible not-being and being.Footnote 12 Hence Avicenna presents to the reader an ontological system in which God and the creation are clearly and totally different as regards both their causality and their being. Only God is uncaused, while being the cause of all things that follow him in the fluxus of being.Footnote 13 Contrarily, the possible being, that is, the possible creation that becomes actual in the transition from its possibility to the status of necesse esse per aliud, always requires a cause of its being.
In these two chapters, Avicenna notably presents five arguments that demonstrate the true and unrelated uniqueness of the necessary being. As remarked by Lizzini,Footnote 14 these proofs are based on a common argument: given two beings that are identical in their definitions and existing simultaneously, their relation always involves a causal action. This action will cause either (1.) the existence of both of them, and therefore the cause is external to their relationship, or (2.) the existence of one of them, so that the first being is caused and the second is its cause. Hence, in every relation between two beings, we have only two possible situations: in the first case, they are both caused by a third cause; in the second case, one is the cause and the other is the caused. Avicenna's demonstration is preceded by a short paragraph on the characteristics of the necessary being, and followed by an ending on the qualities of the possible beings. In between, there are the five arguments, among which it is possible to distinguish two classes of demonstrations: the first two arguments state the unrelatedness of the necessary being (by means of the argument of homology and the argument of the two relatives), while the last three demonstrations testify to the absolute uniqueness of that being. In summary, the demonstrations are as follows:
First argument Footnote 15 (the “homology argumentFootnote 16”) – the premise of the argument is the hypothesis of the existence of two necessary beings homologous in their being and inseparable in existence, neither of which is the cause of the other. If we take one of these necessary beings (α), it will be either (a.) necessary in itself or (b.) not necessary in itself. Moreover, option (a.) entails that given that α is necessary per se, in relation to the existence of β we have the following options:
(aa.) β is necessary in itself as well and α may be necessary in itself and in virtue of the necessary being of β, so that α would be necessary per se and per aliud, but this is impossible; or
(ab.) being β necessary in itself, α may be necessary in itself and not through β, but if their relation were as such, there would be no way to justify their concomitance and their bond, and this goes against the premises of the analysis. Hence, the second main option (b.) is to be accepted. α is not necessary in itself, so that α is possible in itself and necessary through the intervention of β. However, this second conclusion is also false according to the premises. After thorough examination of their relationship, since the being of α entails the being of β, we reach another set of alternative options:
either (ba.) the same ontological condition of α also applies to β; or
(bb.) α and β are not in the same ontological condition.
The latter conclusion cannot be accepted, for the premise entails that the two necessary beings are coaequalia, so that we should suppose that both comply with the same conditions; this would mean that both of them are possible per se and necessary through the intervention of the other. However, this would mean the alternative options are:
(baa.) either the possible being of β causes the necessary being of α, and then the necessary being of α causes the necessary being of β, but if it were true that the possible being of β causes the necessity of α, the being of the latter would not entail any necessity and the relation would be between a cause (β) and a caused (α); and
(bab.) or the necessary being of β causes the necessary being of α, and at the very same time the necessary being of α causes the necessary being of β. However, this latter option implies that the necessity of being of each presupposes itself, for it is both the cause of the necessary being and essentially subsequent to the same necessity of being.
Therefore, it is impossible that two necessary beings which are homologous in their being exist simultaneously.
Second argument Footnote 17 (hypothesis of the two relatives) – the argument is based on the hypothesis of two necessary beings of which neither is the cause of the necessity of the other but their simultaneous presence causes the necessity of both their beings. It is possible, according to Avicenna, to make two different claims concerning such beings. Firstly (a.) if we consider the existence and essence of α and β on the basis of their relationship, we must choose between two options:
(aa.) the existence and essence of α and β consist of and derive from their mutual relation, but in this way they are in themselves possible beings, for their necessity depends on the other; and
(ab.) the existence and essence of α and β do not consist in their mutual relation, but if this were true, they would not be relatives if not accidentally, for their relationship would be extrinsic and added to their essence. Hence, once excluded these first options, we should admit that (b.) the relative beings under examination have the same essence. Accordingly, the relation between their essences has to be reduced to one of these eventualities, i.e.:
either (ba.) the essential definition of α causes the essential definition of β through the same essence of α, but in this case it is the being of α that causes that of β, and so the first is a cause and the latter is caused; or
(bb.) there is no causal bond between the essential notion of α and β, and they integrally consist of their relationship. This would mean, however, that their essence derives from an extrinsic cause, their bond is accidental, and their being is caused.
So it is impossible that two relative necessary beings exist.
Third argument Footnote 18 (uniqueness of the definition of necessary being) – if we suppose the existence of manifold necessary beings, they will be:
(a.) in a situation where each is identical to the others in terms of their intellectus essentiae, and so the essential notion of α would be identical to that of β; or
(b.) each of them is different from the others in terms of their essential notion.
But if we admit the first alternative (a.), supposing an identity of their essential notions, the being of α could not be differentiated from the being of the other through its essence, and its differentiation (and identification) would be given by an accident or unessential attribute that are inherent to its essence. But this would mean the supposed accident would be inherent to the essence of α:
(aa.) through the quiddity of α, which is identical to every other supposed necessary being, and so there would be no principle of identification; or
(ab.) through an external cause, but this would mean its individuation is derived from a cause and the necessity of both would derive from their mutual presence, and this scenario was already refuted.
Hence, it is impossible (a.) that their essential notion is identical. But at the same time, it is unacceptable that (b.) their intellectus essentiae is different, because in that case we face the alternative between:
(ba.) the eventuality that the essence of each one coincides with the necessity of its being, but if this were true, there would be no principle of differentiation between them; and
(bb.) the eventuality that their essence is different from their necessity, but in this case the essential definition of each one would be accidental, for their necessity of being does not derive from their essence, and they would be necessary beings before the supposed acquisition of necessity, and since there would be no possibility of an essential differentiation between them, we are again confronted with the first hypothesis (a.).
Fourth argument Footnote 19 (division of the necessary being according to the species and the individual) – the initial hypothesis is that the definition of the necessity of being can be considered as a species characteristic of many individuals. This division should happen in one of the following two ways:
(a.) through specific difference (as the genus is subdivided into species), but the species can account for the actual being and not for the definition, and for this reason it is impossible that such a difference could provide the necessary being to a thing; or
(b.) through accidents (as the species is subdivided into individuals), but as the individuals agree on their essence and are differentiated by accidents, this possibility leads back to the first alternative (a.). Hence the necessity of being cannot be constituted as a generic or specific being.
Fifth argument Footnote 20 (necessity of being as a proper characteristic of the necessary being predicable of a multiplicity of beings) – Following the assumption that the necessity of being is a proper and predicable characteristic of the necessary being, the same characteristic has to be a necessary being, and we are confronted with an alternative regarding the possession of this characteristic. If the belonging of characteristic y is necessary to the necessary being α to which it belongs, the necessity of α is true only if it possesses y, so α is possible and only y is necessary. If, however, the belonging of the characteristic y is only possible for the necessary being α, we face a contradiction, since the main characterization of the possible being is the possibility of being or not being, so the contradiction resides in the same possibility of being/not being as a characteristic of the necessary being, which is impossible. The necessity of being cannot therefore be a characteristic belonging to something. This is also true if we take the characteristic in a general sense, since:
(a.) if we consider that the characteristic y is a necessary being and a particular being, y is that particular being; and
(b.) if we admit that y is a necessary being but not a particular being, that being is particularized either (ba.) thanks to its essence, but in this way it is a particular being, or (bb.) thanks to an external cause, and y would actually be a possible being.
Having proved the unrelated uniqueness and unity of the necessary being, Avicenna can finally state that the necessary being is one in name and number, completely different from the possible being, which is always characterized by the possibility of being or not being until the necessary being causes it.Footnote 21
THE ALTERATION STRATEGY IN GUNDISSALINUS’ HERMENEUTICS OF METAPHYSICS I, 6–7
The De processione mundi is perhaps Gundissalinus’ most mature philosophical work. In this text, the author examines creation from a cosmological and ontological point of view,Footnote 22 describing on the one hand the different stages by means of which the world was created and on the other the ontological composition of the created being following the Gabirolian doctrine of universal hylomorphism.
The first pages of the treatise are dedicated to resolving a series of epistemological issues, directly concerning the very possibility of a theoretical investigation of the first causes. Once the conditions of possibility are clarified, Gundissalinus starts his analysis by descending from the first Cause, God, i.e. the necessary being who causes all the subsequent beings, and whose possibility of existence is separate from the possibility of non-existence. It is in this passage that Gundissalinus presents the long quotation from Metaphysica I, 6–7 that we have just examined. This quotation plays a fundamental role in the continuation of his analysis, since the ontological difference between necessary and possible being is the theoretical basis on which Gundissalinus’ universal hylomorphism is founded.
Thus, the focus shifts from the cosmological description to the analysis of creatural being, and the different logical stages that lead from the first creation of matter and form to the first substance, and then to the particularization of that substance into the first, eternal caused beings: the celestial intelligences, the planets and the four elements. These first caused beings – first from an eminently ontological point of view, as from a logical perspective their ontological composition involves more stagesFootnote 23 – serve as secondary cause in the institution of the creation. After having described the procession of the beings from God to the secondary cause, Gundissalinus ends his treatise with a very brief exposition of some arithmological pertinences.Footnote 24
In this philosophical scheme, the doctrine of the necessary and possible being serves a double aim. On the one hand, through Avicennian theory Gundissalinus states the main ontological difference between God and creation and, at the same time, the proofs there presented are judged by Gundissalinus as fully sufficient for the treatment of the existence and characterization of God. On the other hand, the theory of the necessary and possible being plays a fundamental role in the evaluation of the creatural being: Gundissalinus connects this doctrine to the Gabirolian universal hylomorphism, interpreting the causal transition from the possible being to the necessary being per aliud as the actualization of a potential state, characteristic of both matter and form, through their union. In this way, Avicennian ontology is merged into the universal hylomorphism, forming a new theory in which the possible being is that characteristic of both matter and form before their union,Footnote 25 the necessary being per aliud is the being of the hylomorphic compound after the union of matter and form and the necessary being per se is the being of God only, first eminent cause of everything.Footnote 26
It is clear that, from a certain point of view, Gundissalinus is taking Avicenna's theory to its extreme consequences here, directly linking the necessity/possibility of being to the potency/act dialectic, and then to hylomorphic dynamics. However, it is clear that it would have been difficult for Avicenna to plainly admit such an ontological perspective. This is why Gundissalinus often needs to modify some passages of his sources, in order to offer a stronger consistency to his treatise.
Not every passage where we see an alteration has to be considered as a voluntary change introduced by Gundissalinus, and not every alteration voluntarily introduced by the author responds to a theoretical aim. Gundissalinus seems to be keen to improve the elegance of the Latin rendering of the Arabic text he translated, introducing into his quotations some textual changes which, from a theoretical point of view, are quite irrelevant, at least in the present study. We will therefore focus only on the changes that are (a.) deliberately introduced by Gundissalinus (b.) with a theoretical aim.
Gundissalinus presents several quotations in his philosophical works, since the doctrinal adoption he accomplishes of many concepts from his source usually corresponds to a direct quotation of the source itself. It is hence possible to define some criteria for the peculiar use Gundissalinus made of his sources, highlighting the main aspects that characterize his modus operandi.
A first criterion (1.) concerns the length of these quotations. In his work, Gundissalinus inserts (1.1) a high number of very short quotes, often no longer than a simple phrase, covering at best a period. These brief excerpts could be interconnected to other quotations of the same length, and they are always placed within a very fluent textual ensemble, in which – and this is another typical temper of Gundissalinus – an explicit mention of the original author never appears. Due to this conciseness, the most prominent effect of this procedure is that, by changing the semantic context of a quote, Gundissalinus often modifies its original sense as well.Footnote 27 However, we cannot say, at least for now, whether or not Gundissalinus was fully aware of the extent of these modifications.
A second class of quotations (1.2) includes excerpts whose text is, approximately, longer than a period but within the limits of a paragraph.Footnote 28 In this kind of quotations, textual alterations are not sporadic, and seem to be subordinated to a global philosophical rereading of the source used by Gundissalinus, who seems to modify the quoted texts, through a semantic bending, in order to adapt them to his speculative perspective.
The last kind of quotation according to this first criterion (1.3) comprises excerpts much longer than a paragraph, texts that are structured in different conceptual kernels and that may arrive to include whole chapters from the employed sources. Among the more eloquent examples of this approach, one may include the Summa Avicennae de convenientia et differentia subiectorum, in the final part of the De divisione philosophiae,Footnote 29 the presentation of the Avicennian doctrine of necessary and possible being in the De processione mundi and, more peculiarly, the whole structure of the De scientiis.Footnote 30 The length of these passages often offers a series of textual modifications and meaningful alterations, introduced in different ways by Gundissalinus.
The second aspect we have to point out in order to establish valuable criteria to analyse Gundissalinus’ “modus referendi” is (2.) the philosophical range of each quotation. As a matter of fact, as the number of direct quotes in Gundissalinian texts is so high, it is possible to distinguish between (2.1) a class of passages that quote a textual passage with an important doctrinal kernel, or a quotation that performs a distinctive role in the treatise's theoretical structure;Footnote 31 (2.2) a group of passages, often very short, that have minor importance on the doctrinal level both in the original and in the derived text;Footnote 32 and (2.3) those textual passages that are of little relevance in the source text but that play a primary role in Gundissalinus’ treatises.Footnote 33 The last group is especially important, as it reveals the focus of Gundissalinus’ critical perspective.
The last criterion we have to define is (3.) the proper alteration of the text, a phenomenon already highlighted by Hasse.Footnote 34 Even in this case, it is possible to single out three main categories of textual modifications, namely:
(3.1) – introduction of new textual material in the quoted excerpt
(3.2) – elimination of a part of the original text in the passages presented by Gundissalinus
(3.3) – proper textual alteration, divided into a) lexical and morpho-syntactical alterations, and b) changes to the structural order.
All three kinds of alteration are found in the presentation of the doctrine of necessary and possible being Gundissalinus provides in his De processione mundi by means of the long quotation from Metaphysica I, 6–7. Regarding the first two criteria we have just presented, it clearly appears that this excerpt is one of the longest quotations (1.3), including two chapters of the first book of Avicenna's work. Moreover, the doctrine of necessary and possible being is of primary importance for both Avicenna and Gundissalinus (2.1). It performs a fundamental task in the economy of Gundissalinus’ cosmological treatise, since it proves the existence of God and introduces the ontological difference between first cause and the series of caused beings, the theoretical substrate for the whole exposition of the De processione mundi.Footnote 35
The two versions of the text are very similar to one another, but on closer analysis a large number of discrepancies appears. We will focus here only on the most significant instances of Gundissalinus’ alteration strategy.
A first aspect clearly revealed by the textual confrontation is Gundissalinus’ modification of the order of the arguments. Gundissalinus modifies the structure of the exposition on at least two occasions: the De processione mundi presents us with a textual articulation that starts with the analysis of the possible being and then moves on to examine the necessary being, whereas in the Liber de philosophia prima we find a reverse order.Footnote 36 Likewise, the negative attributes of the necessary being have a different position in the text in the first Avicenna translation, that is to say as premise to the treatment of the necessary and possible being, whereas in the De processione mundi this passage figures as a conclusion and as a summary of the previous demonstrations.Footnote 37
In addition to these changes in the order of the exposition, the textual comparison also reveals a high number of textual alterations characterized by the insertion, elimination or the plain modification of passages or words. Often these changes seem to fit the aim of yielding more textual intelligibility, both in exposition and lexicon. Such a need for clarity originates from the significant textual complexity of the original Arabic text and leads Gundissalinus to seek terms that could be clearer and often more precise, and to break the long sentences of the first, literal translation from the Arabic, in order to give more perspicuity to the philosophical reasoning. This is the case with many word modifications aimed at reinforcing the sense of Avicenna's line of reasoning, such as, for example, the change of the connective vel with aut in order to reinforce the strict necessity of the alternative presented, or the choice of the periphrasis “they equally accompany themselves” (aequaliter se concomitantur) instead of “they are equal in the association of being” (coaequalia in concomitantia essendi) in the Metaphysica.Footnote 38 Another typical modification of this sort, with some doctrinal implications, is the tendency to substitute Metaphysica's “necesse esse” with the attributive “necessarium esse”,Footnote 39 which seems to respond to Gundissalinus’ focus on the ontological patterns of Avicennian theory.
Nevertheless, this clarification does not always obtain the expected results: Gundissalinus’ choice to change “and it is also impossible that the other one be similar or not similar” (et impossibile est etiam quin illud aliud vel sit similiter vel non sit similiter Footnote 40) to “and so it is impossible that the other one be so or not be so” (et tunc impossibile est, quin illud aliud aut sit sic aut non sit sic Footnote 41) is far from satisfactory, since, in addition to be cacophonous (at least in the Latin text), it is less clear than the former expression, even though its insertion springs from the probable aim to strengthen Avicenna's argument, by replacing the “similiter” – which implies a likeness between the terms – with “sic”, which conveys a stronger identity for the two beings supposed in this comparison.
These kinds of alterations may also involve a straightforward doctrinal modification. An example of this procedure is provided by the following passage taken from the discussion of the two possible beings, where we read:
The discrepancy here is not only lexical but also doctrinal: in the original version, the “intention of not being” (intentio non essendi) derives from that very cause which is the privation of the cause of the intention of being, following a logic-ontological perspective typical of Avicennian modus speculandi. In the Gundissalinian version, however, the definition of not-being derives from a cause that deprives the being of the thing, a supposed negative cause that subsists as a being that deprives other things of their being. It is clear, then, that the two perspectives are different and the textual change is subordinated to a different understanding of the argument.
In other cases, Gundissalinus modifies the text by depriving it of some elements and/or re-elaborating the whole passage in a synthesis of Avicenna's reasoning. An example of this may be seen in the presentation of the second of Avicenna's arguments,Footnote 42 where Gundissalinus provides a synthesis of the premise of Avicenna's demonstration, presenting the results of his reasoning and displaying a clear syntactical restyling.
As we have said, the alterations Gundissalinus introduces are often aimed at a doctrinal sharpening of Avicenna's perspective. Two passages are in my opinion paradigmatic in this regard. The first is related to the first proof of the uniqueness of the necessary being. This argument is based on the impossibility of the essential definition of the necessary being to be manifold.
The first part of this demonstration refers to the sub-hypothesis according to which two necessary beings are supposedly identical in their essential definition. In this case, even though they share the same essential definition, all the same they are separate from one another, so that a differentiating factor must be supposed, such as an accident or an un-essential attribute. At this point, the argument bifurcates into a further alternative concerning the relationship of these supposed accidents with the necessary being. And it is here that Gundissalinus felicitously modifies the text.
Hence, the first possibility of the alternative is linked, both for Avicenna and Gundissalinus, to the eventuality that the diversity between the supposed two beings inheres in their essence in virtue of their very quiddity, that is, thanks to their essence: however, this would mean that what should be the principle of their differentiation is identical for both the supposed necessary beings, and that is contradictory.
With regard to the second possibility of this alternative, we find an interesting discrepancy between the two versions of the text:
This second alternative is not only different but presumptively opposite in the two versions, as we read “by an extrinsic cause and not by its own quiddity” (ex causis extrinsecis non ex ipsa sua quidditate) in the Metaphysica's text, and “by an intrinsic cause not a cause assigned to the essence” (ex causis intrinsecis non causis essentiae addictis) in the De processione mundi.Footnote 43
In the Liber de philosophia prima, the hypothesis of an external pertinence is directly linked to the previous hypothesis regarding the possibility of an internal pertinence by the quiddity of the necessary being. If we admit an external cause that grants its necessity to that being, said necessity would derive from something other than the supposed being itself, meaning it could not be said to be a necesse esse per se.
Rather, a differentiating factor is supposed in the De processione mundi, which is developed by a cause inherent to the essence of that necessary being, thanks to which the two beings should differ from one another. This would mean, however, that the necessity of these two beings derives from this same internal cause, so that, regarding the definition of their essence, they are possible beings; this latter scenario is inconsistent, however, as they would be simultaneously possible and necessary beings. Thus, the outcome of the two arguments seems to be the same, even if their course is different.
As a matter of fact, the consequence of the argument presented by the Metaphysica is that the pertinence of an external cause is at variance with the premise of the reasoning itself: the two beings were supposed to be necessary per se but they are necessary per aliud. Gundissalinus’ result is minutely different: here, too, the pertinence of a cause involves the two supposed beings being necessary per aliud, because they are caused. Indeed, everything that is caused is a possible being whose possibility has been granted by its cause: through this distinction of the possibility of being from that of not-being, as we have seen, the possible being becomes a necessary being but not per se: it becomes a necesse esse per aliud. In contrast with the Liber de philosophia prima, here the cause is presented as internal, and so we might suppose that for Gundissalinus there is a contradiction between the inner necessity of the two beings and the fact that their necessity is caused: they would be simultaneously necessary per se and per aliud, which is not possible. The two versions of this argument therefore lead to the same conclusion: the impossibility of two necessary beings.
The last paradigmatic case of Gundissalinus’ alteration strategy I would like to underline here belongs to the fourth Avicenna's argument and is directly linked to the definition of the necessity of existence:
The question here concerns the possibility of dividing the necessity of being as a genus in its species, and thus supporting a kind of multiplicity. Nevertheless, species are not part of the definition of their genus, since throughout their existence, species only actualize the genus, i.e. they give it its being in act. This principle can be grasped using the example of the relationship between the species of rational beings and its animal genus: in the definition of animal, rationality is not included, since not all animals are rational; nevertheless, however, rational animals – like the other animal species – through their existence actualize the genus’ existence. For this reason, if there were specifications of the necessity of being, they would not give their genus the necessity of being but its subsistence in act, and that is not possible. In fact, the same definition of necessary being implies, as we have seen, the impossibility of not being as its proper definition, and a lack of difference between essence and existence.
We can detect in this passage the careful approach Gundissalinus adopted in his work. If, stylistically, the text presented in De processione mundi is clearer, how Gundissalinus modifies the Avicennian terms intentio and certitudo with the simpler and univocal “essence” (essentia) is very significant: these changes are probably due to the ontological stance the author wanted to emphasize in his treatise, linking the whole demonstration to the other parts of his text. At the same time, it is very representative of his modus operandi that Gundissalinus modifies the definition of the necessity of being from the “impossibility of not being” (impossibilitas non essendi) to the more ontological “unceasing activity of being” (incessabilitas essendi), as well as the subsequent comparison to the animalitas,Footnote 44 which is different from the necessity of being in the Metaphysics since the animalitas is “an intention beyond the necessity of being that accompanies it and that adds to it” (intentio praeter necessitatem essendi, et est esse comitans illam et superveniens illi), while in the De processione the same difference is expressed by the being of the animalitas, that “is subsequent to it or that adds to it” (est consequens illam vel superveniens illi).
How can we explain these discrepancies? And what motivates Gundissalinus to conclude this passage by modifying the Avicenna's “whence the acquisition of the necessity by the necessary being is the acquisition of the certainty of its necessity” (unde acquisitio necessitatis necessitati essendi est acquisitio condicionis de certitudine suae necessitatis) into “therefore the differences are not useful to the necessity of being except only for what is between its essence and its form” (ergo differentiae non prosunt necessitati essendi nisi ad id tantum, quod est inter essentiam suae formae)? As far as the first question is concerned, it seems plausible that Gundissalinus was driven by the intention to clarify the passage, as “being subsequent” (consequens) implies a strict logical consequence that “being companion” (comitans) does not support, bringing the text closer to the specific sense of Avicenna's line of reasoning.
Regarding the second question, the Metaphysics states the indifference of essence and existence for the necessary being, which entails the impossibility of the acquisition of its actual being, since it is implied in its very definition. In the De processione mundi, the text is subtly different: it declares that the differences can be useful (i.e. they can have an active consequence, prosunt) only in relationFootnote 45 to the essence of the necessary being's form, but in this way those differences would cause the existence of the same essence whose existence is part of its definition: a possibility already previously excluded in the argument. Also here, the textual alteration entails the insertion of new doctrinal elements – in this case, the form of the necessary being and the direct mention of its essence – that are absent from the Metaphysics translation: insertions and modifications that reveal Gundissalinus’ understanding and hermeneutics of the text.
CONCLUSIONS
A thorough examination of these passages allows us to draw some conclusive remarks. Firstly, we have to consider whether the doctrinal shifts these changes involve are intentional or not. It seems to me possible to claim that Gundissalinus is not only aware of the theoretical implications of his textual alterations, but that these doctrinal modifications are actively sought in order to provide his cosmological treatise with a coherent philosophical outlook. And for this same reason it is possible to speak of Gundissalinus using a deliberate alteration strategy.
This first remark is directly linked to the question regarding the translation from the Arabic original. It seems implausible to me that the quotation we find in the De processione mundi could be derived from a second version of the Latin Metaphysica, and this for two reasons. Firstly, because there is no real coherence between the minor lessons of the Metaphysica's stemma codicum and those of De processione, and thus, from a strictly philological point of view, there is no grounds for such a supposition. Secondly, if it could be true that the changes introduced by Gundissalinus are aimed at a clearer understanding of the text, at the same time we have seen how the purpose of these alterations also envisages their insertion into the doctrinal system presented in the De processione mundi. These remarks, however, do not deny that Gundissalinus may have elaborated these quoted passages translating directly from the original Arabic text: on the contrary, since many modified terms show a stronger proximity to the Arabic text, it seems plausible to claim that Gundissalinus undertook a new reading of the Avicennian text in Arabic, using that text for the drafting of the De processione mundi, checking it and choosing the best terms for his purposes. Nevertheless, this would imply that the De processione mundi was written in a late period of Gundissalinus’ life, once he could have acquired a good knowledge of Arabic, leaving aside the necessity of a collaborator for direct access to the text.
A last remark concerns the philosophical importance of this alteration strategy: its analysis shows how Gundissalinus does not read his sources in a sterile way. On the contrary, the “modus referendi” and the alterations Gundissalinus made in his De processione mundi are parts of a textual and doctrinal balance between different sources, which are interpreted, re-elaborated and blended together by Gundissalinus in order to build his own personal philosophical analysis of the cosmogenesis and the ontological composition of creatures.
APPENDIX
Textual comparison between De processione mundi, pp. 5, 19–17, 1 and Metaphysica I, 6–7
