Options
Article

Imagination and Its Object: Recovering Suhrawardī's Suspended Images

Author
  • Reza Hadisi orcid logo (University of Toronto)

Abstract

In this paper, I offer a modern adaptation of a theory of imagination that we can find in the writings of Shihāb al-Dīn Suhrawardī (1154–1191) and his followers in the Ishrāqī tradition (‘illuminationism’). I turn to this philosophical tradition because it offers a rare historical example of a type of theory that treats imagination as a primary source of knowledge. That is, on the Ishrāqī account, even an ideal human knower must know certain things via imagination. What I offer is a ‘modern adaptation’ of the theory in that I try to extract the theory of imagination from the cosmological and theological framework of the Ishrāqī thought. The resulting theory is still Ishrāqī in that it relies on a unique form of Platonism about knowledge that, I suggest, we can find in Suhrawardī's writings. Finally, I discuss how the reconstructed Ishrāqī conception of imaginative knowledge is especially appealing as a theory of moral knowledge.

Keywords: imagination, Platonism, Ishraqism, theory of forms, moral knowledge

How to Cite:

Hadisi, R., (2026) “Imagination and Its Object: Recovering Suhrawardī's Suspended Images”, Philosophers' Imprint 26: 22, 1–21. doi: https://doi.org/10.3998/phimp.4235

Downloads:
Download XML
Download PDF
View PDF

710 Views

99 Downloads

Published on
2026-07-01

Peer Reviewed

License

Creative Commons Attribution-NonCommercial-NoDerivs 4.0

1. The outline

It is often assumed that, unlike reason or sensibility, imagination is a secondary source of knowledge, in that its role in learning is, in principle, dispensable. For instance, although we often learn the truths of geometry or arithmetic through the inexact yet efficient use of imagination, ideally we would answer a mathematical question through purely intellectual reflection. Similarly, although ideally I would know whether my son is happy at school by perceiving him during the day, I can know this kind of truth about the actual state of the world through the less reliable and less “vivid,” yet more readily available, mental simulation of how he likes to play with other children. Seemingly then, to the extent that we do rely on imagination, that reliance may reflect a limitation on our part or a necessary concession to human imperfection.1

To uncover a radically different account of the epistemic role of imagination, I turn to the medieval Persian philosopher Suhrawardī (d. 1191), the central figure behind the Ishrāqī philosophical tradition. He offers a rare historical example of a theory that does not treat imagination’s epistemic role as merely a surrogate for reason or perception.2 For Suhrawardī, imagination is a primary source of knowledge in that even an ideal human knower must rely on imagination to learn certain kinds of facts.

This account focuses on ideal human epistemic agents in a specific sense: we consider an ideal agent who must acquire knowledge through inquiry. This qualification is particularly significant in the historical context of the Arabic tradition, where philosophers theorized about ideal knowers in two markedly different ways. First, there is extensive discussion of ideal human learners: those who begin with observation, engage with the world, and strive for complete knowledge in an ideal manner. Second, there are also many attempts to characterize knowledge abstracted from the conditions of learning. The latter is often occasioned by questions about the nature of perfected persons in the afterlife, and so, quite literally, it concerns an ideal epistemic agent at the end of inquiry. Here, we are concerned with the former: theories that examine the ideal epistemic agent as an ideal learner. Thus, according to the standard account I’ve just contrasted with Suhrawardī’s, an ideal inquirer would rely solely on reason and perception, and our reliance on imagination reflects a limitation on our part.

To approach Suhrawardī’s account, I draw on a venerable tradition in epistemology that traces back to Plato.3 According to a prominent reading, the Platonist method of ideal theorizing in epistemology begins by distinguishing kinds of objects. Only then, and in light of this distinction, can we theorize about primary sources of knowledge for learning and inquiry.4

The most familiar shape of this Platonic method can be viewed as a way of understanding the distinction between purely intellectual and perceptual knowledge. In this model, we first identify two types of objects: Platonic forms and appearances. Roughly, appearances are sensible particulars that are spatiotemporally located and stand in causal and part-whole relations to other spatiotemporally located objects. They are subject to generation and corruption. They exemplify different types of properties to an approximation (e.g., no object of appearance is “purely” or “precisely” circular). Platonic forms, on the other hand, are non-spatiotemporal objects that are stable and permanent. They contain their qualities “purely” and “precisely.” Etc.

In this Platonist model, in light of that ontology, we can distinguish between two types of faculties for knowledge: we know appearances via perception and Platonic forms via intellect.5 Importantly, this distinction is made at a high level of abstraction about psychological capacities. For instance, to attribute a “faculty” of perception to an agent is to say that she has a set of capacities in virtue of which appearances are possible objects of her knowledge. This capacity may be realized by a single actual psychological function or by a complex set of them. However, the general point remains: in the Platonic tradition, we identify two types of cognition as primary, in virtue of acknowledging two radically distinct types of objects of cognition. This is a metaphysics-first way of doing epistemology.

I will approach Suhrawardī against this broadly Platonist background.6 Most of the medieval philosophers to whom Suhrawardī responded postulated that there are two primary types of knowledge: intellectual and sensible. As early as al-Kindī (d. ca. 873), this epistemological distinction was introduced based on a metaphysical difference between different objects of knowledge. For him, whereas the object of intellectual knowledge is eternal and unchanging, the object of sensibility is “in flux” (al-Kı̄ndı̄ 2012, 285–86). Later, al-Fārābī (d. ca. 950) offered a similar analysis of intellection and sensation based on the immutability and mutability of their respective objects of knowledge (al-Fārābı̄ 1961, 42–43). Most influentially, Avicenna (d. 1037) upheld this bifurcated epistemology on metaphysical grounds by arguing that the object of intellectual knowledge is necessarily immaterial, while the object of sensation is material (Avicenna 1027/1981, chap. 6).7

By contrast, Suhrawardī argues that in addition to intellectual and sensible knowledge, we must acknowledge the reality of another primary kind of knowledge, namely, imaginative knowledge. We must do so because we must acknowledge at least three fundamentally different kinds of objects of knowledge: Platonic forms, sensible particulars, and what he calls “suspended images” [muthul muʿallaqa]. Whereas we cognize the forms via intellect and the sensible particulars via perception, we cognize the suspended images via imagination. Thus, imagination is a primary source of knowledge in that it is the capacity to cognize a unique aspect of reality, one that is populated by suspended images. Hence, even an ideal human knower can know some things about the world only via imagination.

Suhrawardī’s reasons for postulating this further category of entities as proper objects of imaginative knowledge are multifaceted. In recent decades, many scholars have explored Suhrawardī’s reasons within the framework of his cosmology, theory of the afterlife, and philosophical theology.8 However, as a perhaps unintended consequence of their historical sensitivity and depth, these accounts sometimes leave readers with the impression that unless we adopt the entire cosmological framework of Ishrāqism, we cannot accept or even understand the underlying epistemological theory. By contrast, and in a spirit of experimentation, I will try to extract and reconstruct Suhrawardī’s epistemology of imagination from much of the cosmological commitments of Ishrāqism. The resulting theory remains Ishrāqī in that it identifies an epistemological strand in Suhrawardī’s thought and relies on some of his central metaphysical doctrines. Nevertheless, it is a ‘modern adaptation’ of the theory in that, as much as possible, I try to disentangle the view from the more demanding parts of the Ishrāqī ontology.

2. Imagination’s Own Object

Suhrawardī’s initial postulation of the “suspended images” as a type of object can be traced back to an obscure debate about the state of the soul in the afterlife.9 However, the theory of imaginative knowledge that Suhrawardī eventually developed was general and went beyond questions of eschatology. He argued that dreams, ordinary acts of imagination, mirror perception, and eschatological events all involve encounters with the same unique kind of object. As Lit (2017) has pointed out, while these were his initial examples, Suhrawardī held that the suspended images could appear anywhere in the world. We can encounter them in inanimate objects like “mountains, seas, lands,” in “amazing sounds,” in interpersonal interactions, and even in “written sentences” (Suhrawardı̄ 1186/1999b, sec. 256). That is, insofar as we can imagine things in any context, what we cognize is a suspended image. Notably, the examples he provides illustrate episodes of imagining that are occasioned by perceptual experience (e.g., I see color pigments on the canvas and imagine an elephant; I read a sentence and imagine Shīrāz; I hear a sound and imagine the sea near home). As we will see, Suhrawardī maintains that all acts of imagining share a similar structure: namely, we employ a sensible perception, whether through external or internal senses, as a vehicle to reach the proper objects of imagination.

I will thus focus on Suhrawardī’s overarching and independently motivated theory of imagination, which extends well beyond eschatological questions. Given the Platonic model that I outlined above, we should begin by investigating the nature of the proper objects of imagination.

2.1. Suspended images as objects of knowledge

According to the Ishrāqī account, just as we speak of the “world” or “realm” of Platonic forms, we can also speak of the world of suspended images [ʿālam al-muthul al-muʿallaqa] or the world of images [al-ʿālam al-mithālī]. In a yet-to-be-determined sense, suspended images are a set of mind-independent entities that populate an aspect of the real world. Suhrawardī refers to them using the language of a “fourth world”:

[There are] four worlds: [1] the worlds of the dominating lights [Platonic forms], [2] of the man­aging lights [the souls], [3] of the barriers [particular material bodies], and [4] of the dark and illumined suspended images (Suhrawardı̄ 1186/1999b, sec. 247)

Elsewhere, Suhrawardī underscores his realism about these objects by describing them as possessing a life of their own, populated with “cities,” “people,” and “objects”:

If you hear among the statements of the ancients [about] the existence of a magnitudinous world [ʿālam miqdārī], different from intellect and soul, containing cities impossible to count, with among them what the Lawgiver has called Jābulqā and Jābulsā, then do not be too quick in dismissing it. (Suhrawardī, al-Mashāriʿ, “faṣl fī l-ajsām wa-l-muḥaddad wa-l-makān” (Suhrawardı̄ 1952/2009a); translated in Lit (2017, 59))10

The underlying proposal is to consider the objects of imagination as entities with mind-independent existence and features we can discover. We could begin approaching this idea by reflecting on certain intuitions about artistic appreciation and creation. Consider someone who reads a novel but fails to form the correct picture of the characters, the plot, or the scenes. This reader makes a cognitive error, not because they fail to grasp a sensible object, but because they misapprehend the novel’s fictional objects. In Suhrawardī’s terms, the reader fails to navigate the world of suspended images correctly. Likewise, consider a novelist who drafts a scene and realizes it fails to capture what she was aiming for. She might say, “This isn’t it, this isn’t what I was looking for!” In Ishrāqī language, the author is struggling to locate the appropriate suspended images. Or, in Suhrawardī’s more vivid phrasing, she is trying to find the right “city”, “street”, or “person” among the suspended images. In short, imagining, even in the creative mode of fantasy-making, is characterized as a cognitive activity: it has a direction of fit from the mind to the world of suspended images.

Suhrawardī’s strong realism about the ‘realm’ of suspended images, and his corresponding view of imagination as a faculty aimed at discovering an independent reality, were well received in the Ishrāqī tradition. For instance, his most influential early commentator, al-Shahrazūrī (d. ca. 1288), expands upon the above passage and supplements it with vivid descriptions of the independent reality of the world of imagination:

If you hear of the discussion of the ancients that there is a magnitudinous world in existence, which is not the sensory world nor the intelligible. Its cities are countless without end. Among them are Jābulqā and Jābulsā, which are two of the cities of the world of image, both having a thousand gates and uncountable creatures are in it, who are unaware that God created Adam and his offspring. [If you hear all of that,] Believe in it and let your faith in it not be hard on you, as the prophets and the divine sages recognize this world. For the wayfarers there is, among the manifestation of wonders and cases where the natural order is interrupted, what they desire and aim for. Those who excel in sorcery and soothsaying see and bring out wonders from it. The world of image is parallel to the evermoving world of sense. As the elements are constantly receptive of what is appropriate from the spheres and stars, so the spheres and the stars of the world of image are [also] constantly in motion. The elemental apparitions and its composites are receptive of the influence of the imaginable spherical motions and the illuminations of the intelligible worlds. From that occur endlessly different kinds of forms. (Shahrazūrı̄ 2001, 554; trans. in Lit 2017, 93).

Suhrawardī’s realism entails that through acts of imagination, we can aim to discover a suspended image that exists independently of any particular episode of imagining. Suspended images are, in other words, self-subsistent [qāʾim bi-nafs] (Suhrawardı̄ 1186/1999b, sec. 225), a qualification he also applies to Platonic Forms (Suhrawardı̄ 1186/1999b, sec. 94; Kaukua 2022, 160–61).

Here, we should pause to consider a radical implication of the view. According to Suhrawardī, each episode of imagining has truth as its standard of correctness in the sense that its success consists in the accurate cognition of a suspended image. But we may worry that this claim is too strong. For instance, I can imagine an elephant driving a car while knowing that it is false. By default, this act of imagination does not appear to involve any error, since, seemingly, truth is not a default standard of correctness for imagination (even if we can occasionally impose a voluntary constraint on imagination and attempt to imagine whether something is the case in the world (Kind 2016a)).

I will respond by drawing an analogy to Platonic forms. Consider how we may aim to cognize a Platonic form (or, more generally, consider how we may aim to cognize an idea or a concept). There are two ways that such cognitive acts could fail. First, there can be an application error: we may be correctly apprehending a Platonic form by having a valid conception of the type of structure it imposes on reality, but misapply it to the wrong particular object. Second, there may be misapprehension: we may err with respect to the form itself by mischaracterizing the kind of structure it imposes on reality.11 Platonic forms considered in themselves do not err because they do not judge. But we do err in our judgments about either their nature or objects that fall under them. Accordingly, for anyone who is a realist about Platonic Forms (or likewise, about concepts, abstract objects, and so on), every attempt to cognize a Platonic Form, and every attempt to cognize with a Platonic Form, has truth (or veridicality) as its standard of correctness.

Although Suhrawardī does not elaborate on this point with respect to suspended images, we can extrapolate from the above considerations concerning Platonic forms and speculate how he would think of veridicality conditions for suspended images. Imagine an elephant driving down the street. The object of your imagination, i.e., the suspended image, cannot err because it does not pass judgment about anything. However, you may err either in apprehending the character of what you imagine or in using it to make further inferences. In our example, you might misapprehend the image by taking what you imagine to be a raptor driving down the street. Alternatively, you might make an error in application, for instance, by inferring from the image of an elephant driving down the street that, in that imagined scenario, there must be a storm in Toronto.

To be clear, the point is not that we always err when we imagine raptor-elephants. Rather, in imagining an object under an aspect (e.g., imagining an elephant as driving, imagining a house as flying, etc.), we aim to represent the object and its environment as resembling reality in other respects. In Suhrawardī’s terms, when we imagine an elephant driving down the street, we aim to cognize a suspended image in which other features resemble reality, but elephants can drive cars. We could, of course, aim to cognize a different suspended image, one in which elephants drive and are raptors. But to say that every imaginative act has truth as a standard of correctness is to say that, in imagining, we aim to cognize one or another suspended image, and we may miss the target.12

In short, the Ishrāqī realism about objects of imagination commits us to the possibility of error in every act of imagining. When I aim to imagine an object x as A, I aim to cognize a suspended image whose content is x-as-A other things being equal. But that suspended image exists independently of me. Thus, I may err in apprehending the internal structure of x imagined as A, or I may err in apprehending how it relates to other possible images or facts.

2.2. Suspended images as images

In what sense are the suspended images distinct from Platonic Forms and sensible objects? In addressing this question, we confront my methodological constraint head-on. For Suhrawardī, suspended images constitute a unique and fundamental ontological type (as we have seen, he maintains that they comprise one of the four “worlds” or “realms” of being). Thus, we must determine how much of his metaphysical theory is necessary to make sense of his account of imagination. To that end, I will offer a minimalist reading of Suhrawardī’s ontology. That is, what I directly attribute to Suhrawardī will be explicitly asserted by him, though I will also offer reconstructive proposals, clearly flagged as such. Hence, with due warning, I will set aside many of Suhrawardī’s other commitments related to his broader cosmology and metaphysics that do not directly bear on the theory of imagination.

Earlier, when we examined Suhrawardī’s account of the four distinct worlds or realms of being, I set aside the fact that these metaphysical distinctions are articulated through his characteristically Ishrāqī terminology of “lights.” Suhrawardī describes Platonic Forms as “luminous,” sensible objects as “dark” and “barriers to the light,” and suspended images as “partially luminous and partially dark” (Suhrawardı̄ 1186/1999b, sec. 246). We now need to unpack these claims by clarifying the terminology of “light.”13

For Suhrawardī, ‘light’ [nūr] is the most fundamental ontological type in that different kinds of entities are understood as different kinds or degrees of light. Thus, to say that suspended images are a type of mind-independent real thing is to say that they are a mode of light. According to most commentators, this is a metaphorical use of “light” in that we are not dealing with the physical light but a metaphysical one.14 What, then, is light as a fundamental metaphysical category?

In his magnum opus, the Philosophy of Illumination [Ḥikmat al-Ishrāq], Suhrawardī maintains that ‘light’ cannot be defined because there is no concept “more evident” by which it could be elucidated (Suhrawardı̄ 1186/1999b, sec. 107).15 However, he offers his readers a constitutive ‘rule’ [ḍābiṭ] that provides a helpful formulation of what he means by light:

Light-Rule: “Light is that which is evident in its own reality and by essence makes another evident” (Suhrawardı̄ 1186/1999b, sec. 117).

The Light-Rule articulates the basic principle of ontology by providing a criterion through which we can recognize the most fundamental distinctions between different types of entities. On my minimalist reading of Suhrawardī, the Light-Rule identifies two epistemic functions that serve to distinguish metaphysically between kinds of objects. I reformulate the Light-Rule by splitting off these two functions16:

Reflexive epistemic function: an entity x is reflexively illuminating just in case one can know the nature of x by knowing x itself.

Non-reflexive epistemic function: an entity x is non-reflexively illuminating just in case one can know the nature of some other entities by knowing x.

For our purposes here, I want to underline how, for Suhrawardī, the mode of being of an object is determined via its possible epistemic roles, that is, modes of being are functions of modes of making things knowable or “manifesting.” As a result, when Suhrawardī distinguishes between the modes of being of Platonic forms, sensible particulars, and suspended images, his distinction must be understood as a distinction between different ways of making things knowable or manifesting. Construing luminosity, and thus being, in terms of manifesting [ẓuhūr] results in what Sinai (2015) aptly describes as “hardwiring an epistemological dimension into the very foundation of […] ontology” [283]. On the Ishrāqī account, to say that there is a mind-independent object is to say that there is something that makes itself or other things, in principle, knowable. Suhrawardī insists that this is an “in principle” epistemic relation by noting that the luminous objects “would not cease to be light” and would not cease making things knowable “were there no people and nothing at all possessed of senses” (Suhrawardı̄ 1186/1999b, sec. 117). Thus, for him, it is a mind-independent metaphysical fact whether an object makes its own nature or the nature of something else knowable. As Kaukua (2022) observes, here, manifesting “is both an epistemological and a metaphysical concept” [120].17

Now, with my minimalist reading, I promised to make certain aspects of Suhrawardī’s epistemology accessible to those unwilling to adopt his full metaphysical framework. Does the identification of being and manifesting already place that goal out of reach? Of course, satisfying every reader is unlikely, but I believe the core idea behind Suhrawardī’s account is not overly extravagant. It is reasonable to think that one thing metaphysically grounds another insofar as the former explains the latter (e.g., Fine 2012; Schaffer 2009). For instance, the cause of an event grounds the event by causally explaining its existence and nature. In other words, the epistemic relation of being-the-explanation-for-another captures at least one dimension of metaphysical grounding: the nature of the grounded object is, in principle, knowable through its ground. We can interpret Suhrawardī as identifying this epistemic aspect of grounding as wholly constitutive of it. On this view, to assert the mind-independent existence of something is to assert that it has a mind-independent ground, that is, something that manifests it or renders it knowable.18

We can now see how Suhrawardī characterizes Platonic Forms, sensible objects, and suspended images as modalities of light, that is, as modalities of manifesting or making knowable. For Suhrawardī, “the Forms of Plato are luminous [nūriyya]” (Suhrawardı̄ 1186/1999b, sec. 246) in that they can satisfy two epistemic functions: (1) to know a form is to know its essence (i.e., forms make their “own reality evident”), and (2) a form enables knowledge of what falls under it (the form “makes another evident”). Let me briefly unpack this.19

To be a Platonic form A is to be something in virtue of which two kinds of things are made knowable. First, it is to be something in virtue of which a particular object becomes knowable by manifesting A-structures. For instance, to say that there is a horse in this barn is to say certain horse-structures are, in principle, knowable in this object. Further, these structures that make this object knowable as a horse exist independently of their particular instantiation in the barn. In contemporary terms, one might say that Suhrawardī is committed to the independent reality of natural kinds and holds that they are more fundamental than material objects, insofar as the latter are, in principle, knowable by virtue of the former.20 This is what I have called the non-reflexive function of forms, which aims to capture what Suhrawardī called a form’s capacity to “make other things manifest.”21 Second, to be a form A is to be something in virtue of which the A-structure as such is made knowable (e.g., the horse form exists because the horse-structure as such is made knowable by it). This is what I have called the reflexive function of forms, which roughly22 captures what he calls Platonic forms’ capacity to “make their own reality manifest.”

Table 1.

Suhrawardī’s categories of being in relation to the dual functions of light.

Reflexive Non-reflexive
Pure light (Platonic forms)* That which makes possible knowledge of its own nature That which makes possible knowledge of the nature of other objects
Dark and Light (Suspended Images) That which makes possible knowledge of its own nature only partially That which makes possible knowledge of the nature of other objects only partially
Dark barriers (Sensible Particulars) That which does not make possible knowledge of its own nature That which does not make possible knowledge of the nature of other objects

* God and separate souls also belong to this category. However, in this paper, I focus on the Platonic Forms, as they are more directly relevant to understanding the nature of suspended images.

In sharp contrast to Platonic Forms, Suhrawardī classifies sensible particular objects as “barriers” (barāzikh) that are “dark” (ẓulmāniyya) in themselves, in that they cannot satisfy either of the two epistemic functions. On this account, to know a sensible particular object, one needs to know it “in light of” something else. For example, I can know this horse (as existing, or as existing in this-or-that way) only through or “in light of” something like the universal form of horse. Likewise, a sensible particular, at least on its own, does not make knowledge of other things possible. This material object as-a-horse might partly explain the nature of that material object as-a-foal, but that explanation depends on the Platonic forms of horse and foal, which make these material things knowable as this-or-that kind of entity standing in this-or-that kind of relation. In short, material objects qua material are “dark” in that they stand at the terminus of explanatory chains.

Accordingly, for Suhrawardī, particular sensible objects exist only insofar as they are informed by structures that render them knowable, that is, only insofar as they are “illuminated.” For him, an ordinary sensible object (“dusky substance”) taken on its own (that is, abstracted from the structures that explain its nature) is merely an intellectual postulate (Suhrawardı̄ 1186/1999b, sec. 111). But this should not be mistaken for nihilism about sensible particulars. Rather, for him, a sensible object exists insofar as and because there is something about it that is, at least in principle, knowable. That which makes the object knowable as having this or that kind of property is what confers upon it its mode of being. In this way, even “dark” sensible objects participate in being, albeit derivatively, by virtue of their relation to that which renders them intelligible.

“These suspended images,” Suhrawardī writes, “are not the Forms of Plato [muthul aflāṭūn], for the Forms of Plato are luminous [nūriyya],” whereas the suspended images are both “dark and illumined [ẓulmāniyya wa mustanīra]” (Suhrawardı̄ 1186/1999b, sec. 246). Unfortunately, Suhrawardī does not elaborate on this intermediate ontological status in much detail. The task of unpacking the claim is thus left to sympathetic readers. I propose that we adopt Suhrawardī’s Light-Rule, as outlined above, and apply it as shown in Table 1.

Of course, everything hinges on how we understand the idea of partial satisfaction of the reflexive and non-reflexive functions. To explain this, we need to see in what sense the suspended images are suspended.

2.3. Suspended images as suspended

According to Suhrawardī, whereas Platonic forms are “fixed [thābita]” (Suhrawardı̄ 1186/1999b, sec. 152) “[t]he suspended images may be renewed and destroyed like the images in mirrors and the imaginative faculty [takhayyulāt]” (Suhrawardı̄ 1186/1999b, sec. 247). The reference to mirrors, as we will see, represents a general idea about the nature of the content of suspended images. To unpack this, I start by looking at section 225 of the HI, which first introduces the notion of suspended images:

The truth is that the forms in mirrors and the imaginative forms are not imprinted. Instead, they are suspended fortresses, which have no place of inherence at all. Though they may have a place of manifestation, they are not in them. Thus, the mirror is the place of manifestation for forms in a mirror, though they are suspended; they are not in a place nor are they in a place of inherence. [Likewise] the imaginative faculty is the place of manifestation of the forms of the imagination, and they are suspended. [Suhrawardı̄ (1186/1999b), sec. 225; emphasis added, and translation modified]23

The background to this passage is Suhrawardī’s discussion of mirror-vision in section 104 of the HI. To summarize that earlier section: Suppose you successfully see yourself in a mirror. Where is the image that successfully represents you? An intuitive answer is that the veridical image is on the mirror. But Suhrawardī rejects this view by emphasizing how radically different the image on the mirror is from the object it reflects: for instance, the image does not smell like the object, it is two-dimensional, it is inverted, etc.24 Thus, we are left with two options: either mirror images represent their objects in fundamentally distorted ways, or we must rethink the route via which a mirror represents its object.

Suhrawardī takes the second option.25 On his account, when the image in the mirror appears to depict a person accurately, it is not the mirror’s surface that achieves this, but something else, something capable of presenting that person veridically. In other words, what enables someone to see themselves in the mirror is something richer and more substantial than the image on the mirror’s surface. That other thing, as we will see, is a suspended image. Likewise, he holds that when one sees a cat and then imagines it moving away from the laptop by generating a mental image, that mental image is not the proper object of imagination. It is, at best, a vague or nonvivid sketch, insufficient to account for the richness of what imagination can know. For Suhrawardī, the mental image merely points to something else. The same holds even in cases of pure fantasy, when one is not attempting to imagine anything corresponding to external reality. For instance, when someone imagines the fictional cities of Jābulqā and Jābulsā, the mental imagery produced is not, on his view, the proper object of imagination. What, then, is?

Let me begin with a visualization of the mirror example. Sam and Sara are standing in the same room, both looking into a mirror, but each sees a different image (Figure 1).

Figure 1.
Figure 1. Mirror vision. Original composition using public domain vector elements from Openclipart.org

As we saw, Suhrawardī underlines that the forms of the images each person sees in the mirror differ radically from the actual sensible forms of the objects in the room. The mirror flips spatial orientations and presents a two-dimensional image lacking smell, heat, and many other sensory qualities. Suhrawardī’s point is that when Sam and Sara imagine successfully, they treat the two-dimensional sensible form on the mirror’s surface as a site for manifesting a richer, three-dimensional world of interrelated images with smell, taste, and texture. The image on the mirror’s surface functions as a gateway for cognizing those interrelated images. As a result, although Sam and Sara see different sensible gateways, they can enter one and the same world of imagination. These images, so to speak, hover above the sensible objects, because we access them by manipulating the sensible forms before us. In other words, the perception of a sensible form in the mirror offers a slice of a complex imaginal world,26 one we navigate by enriching that sensible form with additional content and dimensions.

While the mirror example is Suhrawardī’s own, the experience of engaging with art may be more intuitive. For instance, when I look at a movie screen, I immediately perceive a sensible object with an arrangement of light and colors (Figure 2).

Figure 2.
Figure 2. Still from Qeysar (dir. Massoud Kimiayi, 1969). Public domain. Retrieved from Wikimedia Commons (accessed March 27, 2026).

On this account, the immediate sensible forms that are perceived function as gateways into the imagined world of the movie. Or, to put it in Suhrawardī’s terms, the complex world of the suspended image makes itself knowable (manifests itself) through the sensible object on the screen. The scene in Figure 2 prompts my imaginative grasp of a suspended image in which I might step into the room, draw closer to the figures, rest a hand on one man’s shoulder, feel the tension in their posture, and so on. Note that the world I enter is not the actual filming location, either, but the world of the image of this-or-that thing as related to other images. In the filming location, camera crew are present; in the imagined world, none are. A person on set can perceive only from a single vantage point, whereas the imagined world can be navigated from multiple, even incompatible, perspectives and by multiple subjects. For instance, I can imagine observing the scene from no particular point of view, or from several at once (Aronowitz and Helton 2024). In this sense, insofar as I succeed in imagining, what I cognize is neither on the television screen nor at the filming site. In Suhrawardī’s language, the object of imagination is not in these places, but is manifested through them. It is a world of suspended images, overflowing with content that we must imaginatively navigate, while the sensible content on the screen remains relatively shallow. For Suhrawardī, the imagined world possesses a richness that hangs over and above the sensible image. The latter, by contrast, suffers from an “essential poverty” in relation to the imaginal world it makes accessible (C.f., Sartre 1940/2004, 125).

As section 225 of the HI indicates (see above), Suhrawardī suggests that the relationship between suspended images and sensible forms in mirror perception generalizes to all uses of imagination. Now, it is important to see how this is meant to apply when I engage the faculty of imagination to contemplate a wholly fictional scene. Here again, I must produce a sensible representation (or better, some mental imagery) within the bodily organ of imagination. However, notably, for him, this sensible content is not itself the proper object of imagination, but functions as a “place of manifestation” for it. That is, even in the case of purely fictional imagination, what we imagine is not “in the head.” Rather, mental imagery is like an internal canvas composed of pigments of color, sound, and other sensory elements, which serves as a gateway for cognizing a more complex set of mind-independent interrelated images suspended above it. On Suhrawardī’s account, once the suspended image is accessed, it can be cognized directly, or “by presence.”27 Thus, the place of manifestation is not a representation of the suspended image; it is simply the condition that enables direct access to it.

The suspended image that is cognized in this way contains structures and patterns that articulate how the sensible content could be if we were to manipulate it, enrich it, or subtract and add further dimensions. This unified set of images is thus “suspended” over the particular sensible content in the sense that it determines part of its range of imaginative possibilities. Whether generated internally or prompted externally, the sensible image that initiates an imaginative episode is merely a slice of the broader world of suspended images, one that specifies how this content could appear from multiple perspectives and across various possible transformations.

So far, I have tried to elucidate the positive sense in which suspended images are “suspended.” But, as noted earlier, there is another contrastive sense that requires unpacking. Suspended images, unlike Platonic Forms, are not fixed [thābita]. How are we to understand this distinction?

Let me begin by stating the conclusion, and then I will present an argument for it. For Suhrawardī, the Platonic Forms are fixed in that they determine immutable and eternal aspects of reality.28 In other words, the Platonic Forms enable knowledge of the world's necessary and universal structures. By contrast, suspended images make knowable those structures that are not eternally fixed. These are features that, while still formal (that is, not reducible to merely sensible facts about actuality) are neither universal nor necessary. To use a contemporary metaphor, to know the Platonic Forms is to grasp how the world is carved at its universal and necessary joints; to know the suspended images is to grasp how the world is carved at its particular and transient joints. In short, suspended images enable knowledge of structures that are real and informative, but contingent and transient.

Here is my argument for these claims. As noted above, Suhrawardī maintains that suspended images can make things knowable only by modulating a sensible image of what is actual, that is, what is sensibly perceived either through external perception or through internally constructed sensible forms in the faculty of imagination. This, in turn, reveals something about the content of what suspended images make knowable: namely, the possibilities that this-or-that sensible form could take. Like many others in the tradition, Suhrawardī holds that sensible objects are in flux and are necessarily and essentially subject to change (Suhrawardı̄ 1396, bk. III.6.9.193). Hence, suspended images make knowable possible structures of objects that are not part of the unchanging fabric of the universe.

By contrast, it is commonly assumed that Platonic Forms make knowable universal and necessary structures and patterns of the world in such a way that they make no reference to changing, transient, sensible objects or events.29 For example, the Platonic Form of a house, insofar as it discloses a certain kind of structure in the world, does not refer to contingent facts about any particular neighborhood, culture, material, or group of people. This is because, arguably, if the Platonic Forms were constituted by contingent and changing entities, they would not be carving the world at its universal and necessary joints. Put differently, the Platonic Forms are essentially immutable in that they make no reference to entities subject to generation or corruption.

Suspended images, by contrast, are not fixed because the content of a suspended image makes knowable possibilities that hover over transient and ever-changing sensible objects in this world. For example, on the Ishrāqī account, when I imagine a house, I must engage with a complex network of images that correspond to manipulations of an actual sensible form (whether that form is present to me via mental imagery, on a canvas, or in my perceptual field). But since that sensible form is itself subject to change, the suspended image associated with it is also subject to change. These structures come and go because the things they structure come and go.

Thus, as I read Suhrawardī, in identifying the suspended images, we are attempting to isolate a unique category of forms. They qualify as forms in that they are not reducible to the sensible forms over which they range (e.g., the world of images hovering over a movie screen is not reducible to the sensible form on the television screen). But unlike the Platonic Forms, these forms are contingent and non-universal, because they make knowable structures and patterns that are constitutively about particular relations among sensible objects. In that sense, they are not fixed or eternal, since the objects that constitute them are themselves neither fixed nor eternal. As we saw in al-Shahrazūrī’s commentary, this temporality, reflecting the flux of the sensible world, is a defining feature of Illuminationist realism about suspended images: “The world of image is parallel to the evermoving world of sense” (Shahrazūrı̄ 2001, 554).

However, this might raise a concern that Suhrawardī’s realism about suspended images, and his view of their ontological status as explanatory grounds of sensible objects, is under threat. Is Shahrazūrī suggesting that suspended images come into being at the same time as the corresponding acts of imagination? If so, can we still speak of discovery? And if suspended images are occasioned by sensible objects, in what sense do they then explain those sensible objects?30

As we saw, Suhrawardī distinguishes between images “suspended over” external sensible objects and those suspended over internal ones. First, suspended images attached to external perception (e.g., possible images of my cat) exist independently of whether anyone is currently imagining them. Even if no one is actively imagining my cat, there are facts of the matter about how he could be imagined, e.g., as turning around, growing wings, and so on. So long as my cat exists, these images remain suspended over him and are, in principle, discoverable.

Second, suspended images attached to internal mental images (e.g., those associated with my imaginary friend) are more closely tied to acts of imagination and memory, since the object over which the suspended images hover is a mental object. Yet even here, Suhrawardī does not require that the mental image be actively imagined at every moment to grant the mind-independent existence of the suspended image. After all, mental images may persist in short or long-term memory. As long as there is a mental image, whether stored or occurrent, there are mind-independent suspended images that determine its contingent structures. Thus, while this class of suspended images would determine mind-independent structures of mental entities, they themselves are not mental constructs.

Now, Suhrawardī does hold that suspended images are co-temporal with the objects over which they are suspended; not with the occurrent perception of those objects or with occurrent mental imagery. That is, suspended images come into being with the corresponding object they determine, whether that object exists externally in the sensible world or internally as a stored mental image. As long as the object exists, the suspended image exists.

However, this co-temporality should not be mistaken for the claim that sensible objects ground suspended images. As we saw, on Suhrawardī’s account, the grounding relation is reversed. Although suspended images come into being simultaneously with the sensible objects they hover over, they are nonetheless the grounding entities, for they make the contingent and non-eternal modal features of those objects knowable. Their co-temporality is thus not that of two equal correlates, but of explainer and explanandum. Of course, suspended images are not eternal grounds like the Platonic Forms. They exist only as long as the sensible objects they illuminate exist. Yet during that span, they serve as the condition for the intelligibility of those objects under certain aspects.

Finally, we can see why the ability to cognize suspended images is indispensable for an ideal learner. What we have said about their content shows that the relationship between Platonic Forms and suspended images cannot be one of simple deduction, since what is made knowable through suspended images is not already “contained” in the Platonic Form. For example, when I look at my cat and imagine him eating my dinner, I access a set of suspended images that specify how this cat, whom I am perceiving, could be. The particular sensible cat is, in other words, reflected and then augmented in the suspended image. But the Platonic Form of “cat” contains no reference to this or that individual cat. The suspended image thus reveals something unique about the possibilities hovering over this particular cat. In this sense, suspended images constitute a genuinely distinct type of light, that is, a genuinely distinct mode of making things knowable.

To be clear, the suggestion that suspended images pick out the contingent and non-universal structures or forms of the universe is a reconstructive proposal, not something Suhrawardī explicitly asserts. It is an attempt to draw out implications from what he does say about suspended images, even though he never formulates this conclusion himself. Suhrawardī does explicitly describe suspended images as both “dark” and “luminous,” and as transient entities that come into and go out of existence. Al-Shahrazūrī formulates this idea in terms of a “parallel” to the “ever-moving world of sense.” In sum, my proposal aims to make sense of Suhrawardī's description of suspended images as both 'dark' and 'luminous' and as transient, in the following way:

  1. The content of suspended images can be characterized as the set of possible manipulations and enrichments of actual sensible forms. Insofar as sensible forms partially constitute suspended images, the latter are partially “dark”, i.e., their constituent parts are knowable only in virtue of something else. However, because a suspended image reflects an informationally richer field of possibilities for a given sensible form, it also renders knowable how that form could be, beyond what the sensible form itself presents. In this respect, suspended images are also “light”; they make the contingent possibilities of a given sensible form (whether externally or internally given) knowable.

  2. Consequently, suspended images are transient objects, since their constitutive parts (the sensible contents over which they are suspended) are themselves transient. Their “light,” that is, their capacity to make things knowable, is not eternal or unchanging in the way that Platonic Forms are.

2.4. Realism about suspended images

Suhrawardī’s own arguments for realism about suspended images take two forms. The first set of arguments may not persuade an uninitiated reader. He suggests that mystical experience can provide direct phenomenological evidence for the existence of such objects. For instance, he writes: “I myself have had trustworthy experiences indicating that there are four worlds,” including “the dark and illumined suspended images” (Suhrawardı̄ 1186/1999b, sec. 246).

The basic idea behind this kind of appeal to one’s epistemic position is not as outlandish as it may initially seem. Just as one may require intellectual training to recognize Platonic Forms as objects of intellection, one may likewise need to cultivate the faculty of imagination for recognizing its proper objects. Nevertheless, I set this phenomenological argument aside. Given what we have said so far, it is relatively straightforward to reconstruct another line of reasoning on Suhrawardī’s behalf, namely, an argument from indispensability, which may prove more promising.

Arguments from indispensability are familiar in contemporary defenses of Platonism or realism about concepts: unless we posit Platonic Forms, we cannot explain certain features of the world and our knowledge of them. We’ve already seen a similar move in Suhrawardī’s defense of suspended images. As we saw in section 104 of the HI, he argues that sensible forms and perception are too thin to account for the depth of what is made knowable to us. He illustrates this with mirror perception, where the sensible content is cognitively linked to a richer set of images that render its possibilities intelligible. That knowledge, he contends, cannot come from sensible content alone; nor, as I argued, from Platonic Forms, which account only for its universal and eternal structures. Suspended images, by contrast, contain structures that govern the contingent possibilities of particular sensible objects. Suhrawardī thus posits a distinct kind of entity, one that makes knowable the modulations of sensible particulars. He then generalizes: all acts of imagining share this structure, modulating over sensible (internal or external) content to disclose an interrelated world of images.

In this section, I aim to supplement Suhrawardī’s original argument by offering a line of reasoning inspired by one of his other writings. I will argue, on his behalf, that without positing suspended images, we cannot make sense of certain aspects of practical reality and our knowledge of it. This is a supplementary argument, as it rests on assumptions about moral cognition that are not explicitly Suhrawardī’s. Nevertheless, the proposal seeks to illuminate an enigmatic passage at the beginning of one of his philosophical allegories, The Red Intellect (ʿAql-i Surkh). In a sense, this is a case of applying the theory: I aim to show how Suhrawardī’s account of suspended images could prove fruitful for moral epistemology.

The Red Intellect opens with the story of a falcon who escapes its cage but remains fettered, still seeking full liberation. This kind of bird symbolism was widespread in Persian philosophical and poetic literature. As Pourjavady explains, in this genre the caged bird symbolizes “the soul, an exile in a foreign world, stranded far from its homeland, [and] yearns for its original home or ‘place of origin’ (mabdaʾ) or ‘place of return’ (maʿād)” (Pourjavady 2009, 293).31 In this context, the falcon encounters an “elder”:

[…] In that wilderness, while still shackled, I saw a person approaching. I moved forward and greeted him. He responded most kindly. […] When I looked at him, I noticed that his beard and complexion were red, so I thought he was young. I said, “O young man, where are you coming from?”

He replied, “O child, you are addressing me incorrectly! I am the firstborn child of creation, and you call me young?”

I asked, “Why has your beard not turned white then?”

He said, “My beard is white, and I am a luminous elder. But the one who captured you in his snare, put you in these shackles, and placed these guardians to watch over you; he cast me into a dark pit long ago. You see my color as red because of that. Otherwise, I am white and radiant. Any luminous whiteness, when mixed with blackness, appears red, like the twilight at the beginning of dusk or the break of dawn. It is [originally] white, for it belongs to the sun’s light, but since one side of it is toward the white light, while the other side is toward the darkness, it appears red.” [Suhrawardı̄ (1999a), 21–22 (4); trans. modified)

Although the falcon has escaped the hunters, he remains fettered and still seeks guidance.32 After this dramatic opening, the elder begins to teach the falcon the path [ṭarīq] of self-liberation, describing how he must fly over seven mountains, each representing a stage of Sufi training.33 Clearly then, the elder symbolizes the wise guide on the Sufi path.34 But note that Suhrawardī portrays his wisdom as “red,” as something poised between day and night, fire and smoke. Why, then, does Suhrawardī not depict the guide as pure light, with features that are simply luminous?

As indicated above, Suhrawardī identifies the Platonic Forms as pure light and particular sensible objects as darkness. We translated the “luminosity” of the Forms in terms of their capacity to make universal and necessary structures knowable. In this sense, the Platonic Forms are luminous because they allow us to carve the universe at its “eternal” joints. Sensible material objects, by contrast, are in themselves dark because, on their own, they do not afford such knowledge.

I speculate that Suhrawardī aligns, in part, with Aristotle when he characterizes the wise guide’s wisdom in this world as not purely luminous. Practical wisdom is not simply knowledge of universal and eternal laws of good action. It also requires knowing what to do in particular situations, amid contingent and ever-changing circumstances, interests, instruments, obstacles, agents, and relations [Aristotle (1985), 1094b–1095a10; 1141b]. But this does not mean that practical wisdom consists of wholly particularized or inexplicable choices. In other words, practical wisdom can neither guide us in this world by just relying on the universal ideas of friendship as such, justice as such, etc., nor can it guide us by just relying on a mere description of this friendship, this unjust situation, etc. Practical wisdom requires going beyond knowing what is in front of one’s nose, that is, beyond the sensible particulars as they happen to be. It involves knowing what could be in front of one’s nose: how particular persons and situations could or should change.

We thus arrive at a picture of practical wisdom that steers between uninformative particularity and implausible generality. On the one hand, Suhrawardī affirms the possibility and necessity of knowing universals. In the practical domain, one must grasp the universal forms of friendship, justice, courage, and so on. These forms may impose side-constraints on the essential and eternal moral features of the world. On the other hand, he suggests that such knowledge suffices only for an agent who does not engage the ever-changing, contingent world of particulars. As we saw, the wise teacher insists he is originally “white and luminous.” But having “fallen,” having descended into the world of sensible particulars, his wisdom has turned “red”: “Every white thing that is connected to light appears red when admixed with black” (Suhrawardı̄ 1999a, 21–22 (4)). As I read the allegory, Suhrawardī is noting that practical wisdom in the world requires further knowledge: that of suspended images, which help determine what one ought to do in transient, non-universal situations, within the constraints set by the universals. The fallen ideal agent must know not only the universal form of justice or friendship, but also the contingent and transient forms of this friendship with that person, that injustice in twenty-first-century America, or respect for those people in light of their particularities. In short, the fallen agent must learn to imagine well.

To put it another way, this supplemented argument from practical considerations suggests that, given the background condition of dealing with transient particulars (which, for Suhrawardī, may be seen as the non-ideal condition of the “fall”) there are formal features of the world that even an ideal agent can access only through imagination. These features allow the agent to carve “practical reality” at its contingent and shifting joints, within the bounds imposed by its universal and eternal structures. On this view, moral imagination does not exhaust the task of moral learning (since intellectual reflection is also required), but it is an indispensable part of it. Whereas intellectual reflection affords us only purely formal structures, imagination delivers structures that, though general at one level, essentially include particulars as constitutive elements. Without imagination, even an ideal agent cannot access this crucial class of moral facts.

3. Conclusion

This has been, by design, an experimental reading of Suhrawardī. I have deliberately set aside many complexities of his theory, not only because they are notoriously difficult to interpret, but also because they were hard to reconcile with my aims. Instead, I adopted an incremental method. Rather than beginning with a full overview of his ontology and the various conceptions of “light” it contains, I introduced only those aspects that became relevant as the argument progressed. I chose this approach because of my interest in Suhrawardī’s practical philosophy. On his account, practical wisdom requires an irreducibly imaginative form of knowledge. Downstream, this means that excellence in practical agency results from immersive imaginative exercises in ascetic and meditative practices, through which the agent habituates ways of engaging with the world that are saturated with imaginative elements.35 But my focus here has been on the upstream story: I wanted to understand how much of Suhrawardī’s radically alternative philosophical commitments are needed to make sense of the claim that practical wisdom requires imaginative knowledge in an irreducible way.

First, I began by suggesting that Suhrawardī’s conception of imaginative knowledge calls for a metaphysics‑first approach to epistemology. Yet as I developed his distinctive metaphysics of light, it became clear that, within his framework, metaphysical types are classified according to their possible epistemic functions. So although I initially adopted a “metaphysics‑first” approach, I came to see that Suhrawardī’s metaphysical and epistemological reflections are deeply intertwined.

Second, Suhrawardī’s project speaks most directly to those sympathetic to Platonism about knowledge or something close to it. I say “something close” because one might find his theory promising so long as one accepts realism about ideas, concepts, or similar postulates, understood as structures that explain the world in their own ways. For such a reader, Suhrawardī introduces a new category of explainers: the suspended images.

Third, I explained his distinction between Platonic Forms and suspended images through a minimalist reading of his ontology of “lights.” Notwithstanding the constraints of theorizing through the history of philosophy, I hope this reconstructive engagement helps sketch at least the outlines of a theory of imagination, one that seeks to identify its proper objects as distinct from those of perception, conceptual thought, and mind‑dependent mental images.

Acknowledgements

Many thanks to the anonymous referees at Philosophers’ Imprint for their thoughtful and constructive feedback; I learned a great deal from their comments. I am also grateful to the audiences and organizers at the Workshop on Analytic Islamic Philosophy at the University of Missouri–St. Louis; the Center for Middle Eastern Studies at the University of Arizona; the virtual Third Annual Conference for Imagination Domination; the Department of Philosophy at Carleton University; the Medieval Islamic and Jewish Philosophy Conference at the University of Illinois at Chicago; the Department of Philosophy at Concordia University (Montreal); the Collaborative Program in Ancient and Medieval Philosophy at the University of Toronto; and the Department of Philosophy at the University of Toronto. For comments and conversations about the paper, I would like to thank Sara Aronowitz, M. Fariduddin Attar, Antony Aumann, Billy Dunaway, Aidan Gray, Amy Kind, Peter Langland-Hassan, Jon McGinnis, and Mohammed Rustom. A summer research award from the Theology, Science, Knowledge Project of the Templeton Foundation supported my work on an earlier version of this paper.

Notes

  1. Arguably, most recent accounts of the epistemic function of imagination focus on the uses of imagination as a surrogate for perception or reason (Yablo 1993; Chalmers 2002; Goldman 2006; Williamson 2016; Currie and Ravenscroft 2002; Balcerak Jackson 2016; Kind 2016b; Kind 2016a; Kind 2020; Aronowitz and Lombrozo 2020; Munro 2021). They thus give us an account of imagination as a secondary source of knowledge. However, these accounts are often compatible with the view that I will develop in this paper, since their advocates do not hold that imagination can be a source of knowledge only if it is used as a surrogate for reason or perception. ⮭
  2. Of course, Suhrawardī is not the only philosopher, even in the history of Arabic philosophy, to treat imagination as a primary source of knowledge. However, the exact way in which he develops the view is unique, and as I hope to show, promising. Elsewhere, I have argued that al-Ghazālī (Hadisi 2022; especially in al-Ghazālı̄ 1111/1998) and Ibn ’Arabī (Hadisi 2021) also treat imagination as a primary source of knowledge, though in quite different ways. ⮭
  3. Suhrawardī’s affinity with Plato, at least as understood by his medieval contemporaries, runs deep. As Corbin reports, Suhrawardī’s disciples were labeled “friends of Plato” [aṣḥāb Aflaṭūn] (1969, 21). For a detailed account of his debt to Plato, see Kaukua (2022, chap. 7). Cf. Benevich (2019). ⮭
  4. For a detailed account of this methodology in Plato, see Moss (2021, chap. 2 ‘Plato's Object-Based Epistemology’). ⮭
  5. On Moss’s reading (2021), according to Plato, we have two different kinds of successful cognitive relation to forms and appearances: we can know the forms, while we can have true beliefs about appearances. If so, we should rephrase and say: Platonists theorize about different kinds of cognition in light of their metaphysics. ⮭
  6. To be clear, I think that in reading Suhrawardī and his contemporaries, it is helpful to begin with a metaphysics-first approach, as it is often their explicitly endorsed method. That is, to determine whether they regarded imagination as a primary source of knowledge, we must ask whether they recognized any fundamental kind of object as proper to it. However, as I will show in section 2.2, Suhrawardī’s distinctive account of basic metaphysical types as types of manifesting ultimately blurs the line between “metaphysical” and “epistemic” types. Arguably, this is a feature that makes Suhrawardī’s version of Platonism distinctive. I am grateful to an anonymous referee for bringing this point to my attention. ⮭
  7. For Avicenna, the intellect may know the immaterial aspects of material objects, too (e.g., see [Avicenna (1027/2005), bks. 1; ch. 1]. Moreover, a classification based on properties such as immutability may, extensionally, coincide with one based on immateriality. ⮭
  8. For example, see Aminrazavi (1996); Corbin (1969); Gutas (2003); Lit (2017); Marcotte (2011); Nasr (1964); Sinai (2015); (Walbridge 1992; 2000; 2001); Ziai (1990); for a notable outlier, see, Yazdi (1992). ⮭
  9. In the al-Talwīḥāt, Suhrawardī explicitly defends the Avicennian view of the afterlife by relying on his theory of imagination (Suhrawardı̄ 1945/2009b, 89–90; Lit 2017, 44–47). Due to space limitations, I have omitted a section on the historical background of Suhrawardī’s theory of imagination. For detailed analysis, see Sinai (2015) and Lit (2017). ⮭
  10. Contrast with Corbin’s highly non-literal translation of the passage (1977/1989, 118). ⮭
  11. For Suhrawardī’s discussion of errors in conceptualizing about forms, see al-Muqāwamāt (Suhrawardı̄ 1396, sec. 1.2.2). For discussion, see Kaukua (2022, chaps. 1 and 9). ⮭
  12. Note that even when we err in imagining, we may still be apprehending a possible object of imagination, just not the one we were aiming to cognize. Of course, because of error, we become prone to inferential and application errors. The status of impossible objects of imagination (e.g., imagining a square with circular sides) is different, however. Presumably, an impossible suspended image is a non-existent object. Suhrawardī holds that non-existent objects possess only mental existence (Suhrawardı̄ 1945/2009b, 176.13–23). On his account, then, when we aim to imagine an impossible suspended image, what we entertain is merely a combination of mental images; one that does not correspond to a mind-independent suspended image. ⮭
  13. A complete account of the “light” terminology is a complete account of Ishrāqī metaphysics. For the reconstructive task ahead, I will discuss the ‘light’ metaphysics by only focusing on parts that I find relevant to understanding the theory of imagination. ⮭
  14. For dissent, see Zhang (2023). ⮭
  15. This mode of beginning a metaphysical treatise is familiar within the tradition. For instance, Avicenna opens his own metaphysical writings by observing that concepts such as “the existent,” “the thing,” and “the necessary” require no further definition, since there are no “better-known” notions by which to clarify them (Avicenna 1027/2005, 22). ⮭
  16. As I have formulated the knowability clauses in both rules, what is at stake is the knowledge of an object (e.g., a sensible particular, a Platonic Form, or a suspended image), rather than necessarily the knowledge of a proposition. It may well be the case that knowing an object entails knowing propositions about that object. For reasons of space, I will set these questions aside. Still, it is worth noting that aspects of Suhrawardī’s view bear directly on them; for instance, his account of “knowledge-by-presence” [al‑ʿilm al‑ḥuḍūrī] appears to suggest the primacy of objectual knowledge over propositional. For detailed analysis, see Yazdi (1992) and (Kaukua 2013; 2015, 125–42). ⮭
  17. See footnote 6. ⮭
  18. Again, this is meant to be a minimalist reconstruction of the Ishrāqī concept of light. For a recent discussion of the other aspects of the metaphysics of ‘light’, see Kaukua (2022, chaps. 5 and 6). ⮭
  19. At least on one prominent reading (Kaukua 2022, chaps. 5–7), for Suhrawardī, the essences of Platonic Forms taken in isolation (“internally”) are simple and identical (i.e., luminosity), but when taken in relation to other things (“externally”), they exhibit distinct essences. These external essences are not strictly speaking real, though Suhrawardī acknowledges their utility in scientific and everyday discourse (Kaukua 2022, chap. 9). For present purposes, I simplify this account by treating the Forms as having distinct essences simpliciter. This, I hope, preserves their epistemic function while avoiding the need to reconstruct Suhrawardī’s full ontological apparatus. ⮭
  20. Of course, anyone uneasy with such forms of realism about natural kinds would be uneasy with Suhrawardī’s view. But I take it that this falls within the range of metaphysical assumptions that many are willing to make. ⮭
  21. Roughly speaking, this corresponds to what Kaukua (2022, chap. 5) calls a form’s “external” function of illumination. I think it is uncontroversial that Suhrawardī’s Platonic forms have this external or non-reflexive function. For my reconstructive purposes, I am staying silent about Suhrawardī’s account of the metaphysics of how they come to have this function. ⮭
  22. See footnote 19. ⮭
  23. Here, Suhrawardī is using maḥall in a technical sense. I prefer Sinai’s translation as “locus” or “place of inherence” (Sinai (2015, 290)). ⮭
  24. For the relationship between Suhrawardī’s argument and Avicenna, see Sinai (2015, 287–90). ⮭
  25. In section 104, Suhrawardī leaves the issue open-ended, posing a rhetorical question about how mirror images and imaginative representations can correctly represent their objects. He notes that the solution to this puzzle “will come later” (Suhrawardı̄ 1186/1999b, sec. 104). He fulfills this promise in section 225, cited above, where he explains that suspended images account for the phenomena of mirror vision and imagination. ⮭
  26. I use “imaginal,” rather than “imaginative,” in the technical sense common in the secondary literature. The neologism highlights the distinction between suspended images and mental image. As we will see, for Suhrawardī the mental image functions like the mirror’s surface: a mind‑dependent site where an aspect of the mind‑independent suspended images is manifested. ⮭
  27. It is beyond the scope of this paper to discuss Suhrawardī’s theory of “knowledge by presence.” Put roughly, the theory holds that genuine knowledge consists in a non-representational, purely relational connection between the knower and the known. For a detailed analysis, see Eichner (2011). ⮭
  28. I am setting aside a complication. For Suhrawardī, everything is ultimately grounded in eternal realities, or in an eternal “light” (Benevich 2019; Kaukua 2024). This raises a difficulty: how can one make sense of a distinction between eternal and non-eternal structures within an ontological framework that is, at its core, eternalist? Still, as we’ve seen, Suhrawardī does invoke such a distinction when contrasting Platonic Forms with suspended images, namely, between things that come into and pass out of existence, and things that are eternally fixed (Suhrawardı̄ 1186/1999b, 149–50). To be sure, fully situating the theory of suspended images within his broader metaphysics would require a more detailed account of how he justifies non-eternal entities. But since the distinction is philosophically intelligible apart from his system, and given the aims of this paper, I am content to grant that Suhrawardī can, in some way, accommodate the difference between fixed and transient modes of being within his eternalist framework. ⮭
  29. Suhrawardī distinguishes between universals “in the sense used in logic” and universals in the metaphysical sense operative in Illuminationist ontology (Suhrawardı̄ 1186/1999b, sec. 169). The former are conceptual constructs that yield nominal definitions and exist only in the mind. There, Suhrawardī explicitly denies that Platonic Forms are identical with universals in this logical sense. However, he affirms that universals (understood metaphysically) are identical with Platonic Forms. In this sense, they are mind-independent structures that make the universal features of the world intelligible at its eternal and necessary joints. Thanks to M. Fariduddin Attar for pressing me on this issue. ⮭
  30. Thanks to anonymous referees for raising these worries. ⮭
  31. The bird symbolism reappears in several philosophically significant allegories. These include Risālat al-ṭayr (“The Epistle of the Bird”), controversially attributed to Avicenna (d. 1037); Risālat al-ṭuyūr (“The Epistle of the Birds”) by Aḥmad Ghazālī (d. 1126); and perhaps most famously, Manṭiq al-ṭayr (“The Speech of the Birds”) by Farīd al-Dīn ʿAṭṭār (d. ca. 1225). An earlier version of a similar narrative appears in the Syriac Hymn of the Pearl (Poirier 2021). For a helpful overview of this tradition, see the opening pages of Saberi-Najafi (2022). ⮭
  32. This is a recurring theme in the genre. The underlying idea is that recognizing one’s moral predicament in this world is one thing; overcoming it is another. ⮭
  33. The mystical symbolism of the number seven is a common theme in Sufism. Many Sufi manuals speak of “seven stations” of mystical growth. For instance, in his influential encyclopedia of early Sufism, Kitāb al-Luma [The Book of Light], Abū Naṣr al-Sarrāj (d. 988) names seven stages of Sufi growth (al-Sarrāj 1914, 42). ⮭
  34. In the Sufi tradition, the path to “liberation” requires a guide or teacher, who may be either a living person or a figure encountered solely in imagination. Many of the greatest Sufis (sometimes referred to as the Uwaysīs) claim to have been instructed by imaginal figures, most commonly the figure of Khiḍir (Corbin 1969, 32–33). ⮭
  35. For an overview, see Aminrazavi (1996). ⮭

References

al-Fārābı̄ (1961). Fuṣūl al-Madanı̄: Aphorisms of the Statesman. Translated by D. M. Dunlop. Cambridge University Press.

al-Ghazālı̄ (1111/1998). The Niche of Lights. Translated by David Buchman. Brigham Young University Press.

al-Kı̄ndı̄ (2012). On the quantity of Aristotle’s books. In The Philosophical Works of Al-Kindı̄, edited by Peter Adamson and Peter E. Pormann (279–296). Oxford University Press.

al-Sarrāj, Abū Naṣr (1914). Kitāb al-Lumaʿ fı̄ l-taṣawwuf [The Book of Flashes of Light in Sufism]. Edited by R. A. Nicholson. Leiden: Gibb Memorial Trust.

Aminrazavi, Mehdi (1996). Suhrawardi and the School of Illumination. Curzon Sufi Series. Curzon Press.

Aristotle (1985). Nicomachean Ethics. Translated by T. Irwin. Indianapolis: Hackett.

Aronowitz, Sara, and Grace Helton (2024). Subjectivity in film: Mine, yours, and no one’s. Ergo: An Open Access Journal of Philosophy, 11(2). doi:  http://doi.org/10.3998/ergo.5707.

Aronowitz, Sara, and Tania Lombrozo (2020). Learning through simulation. Philosophers’ Imprint, 20(1), 1–18. https://hdl.handle.net/2027/spo.3521354.0020.001.

Avicenna (1027/1981). Avicenna’s Psychology. Translated by Fazlur Rahman. Hyperion Press.

Avicenna (1027/2005). The Metaphysics of The Healing. Translated by Michael E. Marmura. Brigham Young University Press.

Balcerak Jackson, Magdalena (2016). On the epistemic value of imagining, supposing, and conceiving. In Knowledge Through Imagination, edited by Amy Kind and Peter Kung (41–60). Oxford University Press.

Benevich, Fedor (2019). A rebellion against Avicenna? Suhrawardı̄ and Abū l-Barakāt on 'Platonic forms' and 'Lords of species'. Ishrāq: Islamic Philosophy Yearbook, 9, 23–53.

Chalmers, David (2002). Does conceivability entail possibility?. In Conceivability and Possibility, edited by Tamar Szabó Gendler and John Hawthorne (145–200). Oxford University Press.

Corbin, Henry (1969). Creative Imagination in the Sufism of Ibn ’Arabi. Translated by Ralph Manheim. Bollingen Series 91. Princeton University Press.

Corbin, Henry (1977/1989). Spiritual Body and Celestial Earth: From Mazdean Iran to Shi’ite Iran. Translated by Nancy Pearson. Bollingen series XCI:2. Princeton University Press.

Currie, Gregory, and Ian Ravenscroft (2002). Recreative Minds: Imagination in Philosophy and Psychology. Oxford University Press.

Eichner, Heidrun (2011). 'Knowledge by presence', apperception and the mind-body relationship: Fakhr al-Dı̄n al-Rāzı̄ and al-Suhrawardı̄ as representatives and precursors of a thirteenth-century discussion. In In the Age of Averroes: Arabic Philosophy in the Sixth/Twelfth Century, edited by Peter Adamson (117–140). Warburg Institute.

Fine, Kit (2012). Guide to ground. In Metaphysical Grounding: Understanding the Structure of Reality, edited by Fabrice Correia and Benjamin Schnieder (37–80). Cambridge University Press.

Goldman, Alvin (2006). Imagination and simulation in audience responses to fiction. In The Architecture of the Imagination: New Essays on Pretence, Possibility, and Fiction, edited by Shaun Nichols (41–56). Oxford University Press.

Gutas, Dimitri (2003). Suhrawardı̄ and Greek philosophy. Arabic Sciences and Philosophy, 13(2), 303–309. doi:  http://doi.org/10.1017/S0957423903003138.

Hadisi, Reza (2021). Creative imagining as practical knowing: An Akbariyya account. Res Philosophica, 98(2), 181–204. doi:  http://doi.org/10.11612/resphil.2036.

Hadisi, Reza (2022). Ghazālı̄’s transformative answer to skepticism. Theoria, 88(1), 109–142. doi:  http://doi.org/10.1111/theo.12331.

Kaukua, Jari (2013). Suhrawardı̄’s knowledge as presence in context. Studia Orientalia, 114, 309–324.

Kaukua, Jari (2015). Self-Awareness in Islamic Philosophy: Avicenna and Beyond. Cambridge University Press.

Kaukua, Jari (2022). Suhrawardi’s Illuminationism: A Philosophical Study. Vol. 118. Islamic philosophy, theology and science. Texts and studies. Brill.

Kaukua, Jari (2024). Future contingency and God’s knowledge of particulars in Avicenna. British Journal for the History of Philosophy, 32(4), 745–765. doi:  http://doi.org/10.1080/09608788.2022.2088469.

Kind, Amy (2016a). Imagining under constraints. In Knowledge Through Imagination, edited by Amy Kind and Peter Kung (145–159). Oxford University Press.

Kind, Amy (2016b). Introduction: Exploring imagination. In The Routledge Handbook of Philosophy of Imagination, edited by Amy Kind (1–11). Routledge.

Kind, Amy (2020). What imagination teaches. In Becoming Someone New: Essays on Transformative Experience, Choice, and Change, edited by Enoch Lambert and John Schwenkler (133–146). Oxford University Press.

Lit, L. W. C. van (2017). The World of Image in Islamic philosophy: Ibn Sı̄nā, Suhrawardı̄, Shahrazūrı̄ and Beyond. Edinburgh Studies in Islamic Apocalypticism and Eschatology. Edinburgh University Press.

Marcotte, Roxanne D. (2011). Suhrawardı̄’s realm of the imaginal. In Ishrāq: Islamic Philosophy Yearbook, edited by Y. Eshots (68–79). Vostochnaya Literatura.

Moss, Jessica (2021). Plato’s Epistemology: Being and Seeming. Oxford University Press.

Munro, Daniel (2021). Imagining the actual. Philosophers’ Imprint, 21(17), 1–21. https://hdl.handle.net/2027/spo.3521354.0021.017.

Nasr, Seyyed Hossein (1964). Three Muslim Sages: Avicenna–Suhrawardı̄–Ibn ’Arabı̄. Harvard Studies in World Religions. Harvard University Press.

Poirier, Paul-Hubert (2021). L’Hymne de la Perle des Actes de Thomas: Introduction, Texte, Traduction, Commentaire. Turnhout: Brepols.

Pourjavady, Nasrollah (2009). Genres of religious literature. In General Introduction to Persian Literature, edited by J. T. P. Bruijn (270–311). I.B. Tauris.

Saberi-Najafi, Navid (2022). The Sı̄murgh in Chach: Bird symbolism in the Rawḍat al-farı̄qayn. British Journal of Middle Eastern Studies, 49(4), 580–594. doi:  http://doi.org/10.1080/13530194.2020.1833707.

Sartre, Jean-Paul (1940/2004). The Imaginary: A Phenomenological Psychology of the Imagination. Translated by Jonathan Webber. Routledge.

Schaffer, Jonathan (2009). On what grounds what. In Metametaphysics: New Essays on the Foundations of Ontology, edited by David Chalmers, David Manley, and Ryan Wasserman (347–383). Oxford University Press.

Shahrazūrı̄, Shams al-Dı̄n (2001). Sharḥ Ḥikmat Al-Ishrāq. Edited by Hossein Ziai. Institute for Humanities and Cultural Studies.

Sinai, Nicolai (2015). Al-Suhrawardı̄ on mirror vision and suspended images. Arabic Sciences and Philosophy, 25(2), 279–297. doi:  http://doi.org/10.1017/S0957423915000053.

Suhrawardı̄ (1396). Al-Muqāwamāt. In Al-Ḥikmat Al-Ishrāqiyah, edited by Mohammad Maleki (19–273). Qom: Adyān wa Madhāib University Press. 1396 HSh (Iranian calendar).

Suhrawardı̄ (1999a). The Philosophical Allegories and Mystical Treatises. Translated by Wheeler M. Thackston Jr.. Bibliotheca Iranica: Intellectual Traditions Series 2. Mazda Publishers.

Suhrawardı̄ (1186/1999b). The Philosophy of Illumination. Translated by John Walbridge and Hossein Ziai. Islamic Translation Series. Brigham Young University Press.

Suhrawardı̄ (1952/2009a). Al-Mashāriʿ. In Majmūʿa-yi Muṣannafāt-i Shaykh-i Ishrāq, edited by Henry Corbin and Seyyed Hossein Nasr. Institut d’Études et de Recherches Culturelles.

Suhrawardı̄ (1945/2009b). Al-Talwı̄ḥāt. In Majmūʿa-yi Muṣannafāt-i Shaykh-i Ishrāq, edited by Henry Corbin and Seyyed Hossein Nasr. Institut d’Études et de Recherches Culturelles.

Walbridge, John (1992). The Science of Mystic Lights: Quṭb al-Dı̄n Shı̄rāzı̄ and the Illuminationist Tradition in Islamic Philosophy. Harvard Middle Eastern Monographs 26. Harvard University Press.

Walbridge, John (2000). The Leaven of the Ancients: Suhrawardı̄ and the Heritage of the Greeks. SUNY Series in Islam. State University of New York Press.

Walbridge, John (2001). The Wisdom of the Mystic East: Suhrawardı̄ and Platonic Orientalism. SUNY Series in Islam. State University of New York Press.

Williamson, Timothy (2016). Knowing by imagining. In Knowledge Through Imagination, edited by Amy Kind and Peter Kung (113–123). Oxford University Press.

Yablo, Stephen (1993). Is conceivability a guide to possibility?. Philosophy and Phenomenological Research, 53(1), 1–42. doi:  http://doi.org/10.2307/2108052.

Yazdi, Mehdi Ha’iri (1992). The Principles of Epistemology in Islamic philosophy: Knowledge by Presence. SUNY Series in Islam. State University of New York Press.

Zhang, Tianyi (2023). A Philosophical Enquiry into the Nature of Suhrawardı̄’s Illuminationism: Light in the Cave. Vol. 17. Investigating Medieval Philosophy. Brill.

Ziai, Hossein (1990). Knowledge and illumination: A study of Suhrawardı̄’s Ḥikmat al-Ishrāq. Vol. 97. Brown Judaic Studies. Scholars Press.