Visual metaphors for knowledge are pervasive in the early Buddhist texts. Although such usage is not a cultural universal, the same can also be seen in modern English. The expression “I see,” for example, can convey a form of understanding that need not involve something perceived through the eyes, such as when saying: “I see what you mean” or “I see your point.” In fact, seeing is a primary source of information about the world that provides comparably more data than the other physical senses. This is due to a superior ability of vision to select something specific from its perceptual field, compared to the perceptual fields accessed by the other senses, and also due to the capability to sense stimuli at a greater distance than the other senses. Hence, to relate understanding to seeing, rather than to the other senses, is in a way natural. Commenting on the Buddhist employment of visual metaphors, McMahan (2002, p. 65) observes:

Of the numerous ocular metaphors found in Buddhist literature, perhaps the most pervasive, and that which grounds some of the others, is the correlation of awakening, or the most profound kind of knowing, with vision. From the earliest stages of Buddhism, seeing has connoted the direct cognition of truth. Even verbal truths, when understood fully and directly, are “seen” … The frequent pairing of knowing and seeing indicates intimate, direct, and full knowledge.

At the same time, however, explicit references to the intentional employment of visualization are rare in the early Buddhist discourses. According to McMahan (1998, p. 253), although “Buddhist vocabulary was rife with visual metaphor, vision in a literal sense and visual imagery were not emphasized as a way of communicating the teachings.” Nevertheless, as pointed out by Gethin (2006, p. 99):

We should perhaps also consider that one reason for the relative lack of formal instruction in “visualization” may also be that the visual aspect of certain passages — the fact that they were meant to be imagined and brought to life — may have been largely taken for granted. If we conceive of these texts as being composed orally, being transmitted orally, then the visual dimension may have come alive more or less spontaneously to those reciting and listening to the texts. An oral culture may well nurture a more active visual imagination than a culture transmitted via TV, cinema, and computer screens.

In addition to the unique nature of an oral culture, the basic difference between reading and hearing a text would also be of relevance. When reading, the eyes are already to some extent occupied. When hearing, this is not the case to the same degree, more easily allowing for the possibility of looking at something that is present or else closing the eyes and engaging in visual imagination in line with what is heard. On this understanding, the oral delivery of the early Buddhist discourses could easily have acquired a visual dimension in one way or another. In fact, historical–critical textual study enables identifying evidence that points to the impact of what is visual on the orally transmitted texts. Two examples may suffice to make the main point:

One example is the notion that the Buddha’s body was endowed with thirty-two marks, whose description appears to be the result of a cross-fertilization between ancient Indian art and textual accounts (Anālayo 2017a). In other words, a textual depiction, in itself believed to originate from someone actually having seen the Buddha, would naturally have stimulated visual imagination. The process of concretization initiated by such imagination appears in turn to have impacted ancient Indian art. Seeing an image of the Buddha in art would then have guided visual imagination, which in turn appears to have influenced textual descriptions.

Another example is the well-known hagiographical narrative of the Buddha escaping from his palace at night in order to go forth, after having for the first time in his life encountered someone sick, old, or dead during pleasure outings. These descriptions do not tally particularly well with relevant indications found in the early discourses (Anālayo 2017b, p. 10), making it quite possible that pictorial descriptions of his insight into the human predicament have led to the formulation of these textual accounts (Weller 1928, p. 169).

Fire Imagery 

The early discourses abound in similes (Rhys Davids 1907, 1908). These operate in a range of ways, often taking up something that relates directly to the lived experience of the ancient Indian audience, and such cases can involve a visual dimension. This is not to take the position that such instances are already actual forms of visualization, but rather to point to aspects of orality that come embedded with a visual component. In fact, at times such illustrations are explicitly linked to something that is present or happening, as if to invite taking in the visual image as a converging and reference point for the teaching that is being delivered.

Several early discourses reflect a fascination with fire imagery of various types (Anālayo 2022a, p. 59). A particularly drastic example is the report of the Buddha seeing what appears to have been a wildfire and then inviting his disciples to look at it. Based on the impact of what must have been a stark image of conflagration, a teaching on the dire repercussions of ethical misconduct unfolds. The teaching begins with the Buddha asking the monks present what they would consider preferable, to hug a beautiful woman or to embrace that raging fire? The monks of course prefer the first alternative. In reply, they are told that it would actually be preferable for them to embrace the fire, rather than doing something liable to lead them to commit misconduct and for this reason then end up in hell in the next life. The discourse continues with further inquiries of this type, several of which employ fire imagery. The concluding section offers the following report:

And when this exposition was spoken, sixty monks vomited blood; sixty monks gave up the training and returned to the lower life, [thinking]: “Blessed One, it is hard to do, it is really hard to do”; and the minds of sixty monks were liberated from the influxes through non-clinging.

(AN 7.68: imasmiñ ca pana veyyākaraṇasmiṃ bhaññamāne saṭṭhimattānaṃ bhikkhūnaṃ uṇhaṃ lohitaṃ mukhato uggañchi, saṭṭhimattā bhikkhū sikkhaṃ paccakkhāya hīnāyāvattiṃsu: dukkaraṃ bhagavā, sudukkaraṃ bhagavā ti; saṭṭhimattānaṃ bhikkhūnaṃ anupādāya āsavehi cittāni vimucciṃsū ti).

When this teaching was spoken, sixty monks were liberated from the fetters with the eradication of the influxes, and sixty monks gave up the precepts and returned to the household life.

(MĀ 5: 說此法時, 六十比丘漏盡結解, 六十比丘捨戒還家).

At that time, when this teaching was spoken, the minds of sixty monks were liberated through the eradication of the influxes, and sixty monks left the monastic robes and became white-clothed [laity].

(EĀ 33.10: 爾時, 說此法時, 六十比丘漏盡意解, 六十比丘還捨法服而作白衣).

Leaving aside the reference to vomiting blood, found only in the Pāli version, the parallels agree that the stern instruction delivered in the face of a wildfire had a rather strong impact on the audience: motivating either a return to lay life or the reaching of full awakening. As each case involves the same number of monks, the report balances these two outcomes. The depiction of such outcomes throws into relief the power of combining the witnessing of a conflagration with an imagining of having to endure the repercussions of immorality, which in the ancient Indian imagination closely relates to the raging fires of hell. Although the texts do not report the Buddha explicitly inviting his monks to carry out a visualization, the role of the visual dimension in the delivery of this teaching seems to be clearly implicit.

Another also rather stark image involves the depiction of the arising of multiple suns (Anālayo 2019, p. 120). The main point here is a teaching on impermanence, based on describing how at some time in the future additional suns will arise (up to seven altogether) eventually leading to an incineration of even the largest of mountains. The image of additional suns expresses in a visual manner an increase of solar heat that has to some extent a counterpart in the contemporary scientific prediction that the sun will eventually incinerate the Earth (Schröder & Smith 2008). Without intending to pretend that these two descriptions must be referring to the same phenomenon, it does seem fair to propose that they share some commonalities. This much granted, the motif of seven suns arising one after the other would indeed offer a useful approach for conveying an increase in solar heat radiation in a way that can easily be visualized. Such a visual image can safely be expected to have had quite an impact on its ancient Indian audience.

Besides serving in these various ways as an illustration in the course of giving an oral teaching, the vision of fire can also function as a meditation object in its own right in the form of a “totality” (Pāli kasiṇa, Sanskrit kṛtsna, Chinese 遍處定, Tibetan zad par). Here, the chief purpose is to develop concentration (Anālayo 2022c). According to later exegesis, the meditative procedure for cultivating the totality of fire requires looking at an actual fire until one is able to see the same image with closed eyes, which can then become the object for a deepening of concentration (Vism 171).

Visual Stimulation

Fire is of course not the only visible theme that can be employed when giving an oral teaching. An alternative approach can be seen in the following introductory narrative in a Pāli discourse and its two Chinese parallels:

The Blessed One saw a great log of wood being carried along by the stream of the river Ganges. Having seen it, he addressed the monastics …

(SN 35.200: addasā kho bhagavā mahantaṃ dārukkhandhaṃ gaṅgāya nadiyā sotena vuyhamānaṃ. disvāna bhikkhū āmantesi).

At that time, the Blessed One was observing the river Ganges. He saw that in the midst of the river Ganges there was a great tree floating downstream. He said to that monastic …

(SĀ 1174: 爾時, 世尊觀察恒水. 見恒水中有一大樹, 隨流而下. 語彼比丘; the text is based on adopting a variant that adds 恒 to the first instance of 水, in keeping with the formulation 恒水 found right afterwards).

At that time, the Blessed One saw that in the midst of the water of the river there was a great log of wood that was being floated [downstream] by the water. He sat down on a seat by the side of the river under a tree. At that time, the Blessed One said to the monastics …

(EĀ 43.3: 爾時, 世尊見江水中, 有大材木為水所漂, 即坐水側一樹下坐. 爾時, 世尊告諸比丘).

The parallels continue with the Buddha employing this readily available image to deliver a description of overcoming obstacles to progress to Nirvana, comparable to the log of wood floating down the river unimpeded until it reaches the ocean. The vivid illustration provided in this way was apparently so inspiring that a cowherd overhearing this instruction wanted to go forth on the spot (Anālayo 2022b). Although the parallels do not provide explicit instructions on visualization, it seems fair to assume that those receiving this teaching would have kept in mind the image of the log of wood carried along by the river and perhaps even imagined how this log gradually moves closer to the ocean.

The few selected instances surveyed thus far show the Buddha ready to take up some visual occurrence from the nearby environment to deliver an instruction, which in view of the reported inspiring effects would indeed seem to be a skillful way of teaching in an oral setting. At times, the discourses show a disciple eliciting such a mode of teaching. An example is the report in the introductory narration of a discourse that the Buddha had gone with a group of monastics to a mountain peak. Apparently looking down from a precipice, one of the monastics is on record for posing the following question to the Buddha:

Now, venerable sir, is there another precipice that is steeper and more frightening than this precipice?

(SN 56.42: atthi nu kho, bhante, imamhā papātā añño papāto mahantataro ca bhayānakataro cā ti?).

Blessed One, this is a very steep precipice. Yet, is there another very steep precipice, a steep precipice wherein one can become very frightened?

(SĀ 421: 世尊, 此極深嶮. 然復有一極深嶮, 極嶮於此甚可怖畏者不).

The Buddha of course obliges by expounding the frightful precipice of the failure to understand the four noble truths and thereby remain subject to the dukkha/duḥkha of the cycle of rebirths (saṃsāra). The reported query is clearly meant to inspire the Buddha to use the visible impact of the precipice as an illustration of a teaching. Once again, with the Buddha and his audience apparently being at the edge of this precipice, the visual impact of looking down from this height must have been an integral part of the teaching delivered.

Another instance takes as its occasion the arrival of a group of young men by chariot, well dressed and adorned. According to a range of textual sources, on seeing them arrive from afar, the Buddha told his monastics to look at them, as for those who had not personally beheld the Celestials of the Thirty-three this was an opportunity to see what they look like. The idea behind this instruction would be that some monastics were endowed with the meditative prowess to visit that heaven, but others were unable to do so. Hence, this was a unique opportunity for members of the latter group to acquire a visual impression of what these celestials look like.

Several parallels extant in Chinese indicate that the visual appearance of the young men was not different from the Celestials of the Thirty-three (DĀ 2: 與此無異; T 5: 如此 … 無有異也; T 6: 如此無異; T 7: 等無異也). The same holds for the three main Asian editions of the Pāli version (Be, Ce, and Se of DN 16: tāvatiṃsasadisan); the relevant part in a Sanskrit manuscript parallel has unfortunately not been sufficiently preserved (Waldschmidt 1951, p. 180). Without intending to deny that this may just be meant as a remark with no ulterior purposes, in view of the preceding examples it could also be of further significance. It could be related to a form of mindfulness practice described in the early discourses, namely recollection (Anālayo 2022d).

The early discourses regularly refer to six forms of such recollection, which can take as their respective objects the Buddha, his teaching, the community of realized disciples, morality, generosity, and celestial beings. The last involves calling to mind different celestials and then reflecting that the qualities that lead to being reborn in such a heaven are likewise found in oneself. Parallel versions of such instructions agree in mentioning the Celestials of the Thirty-three as the second member in their listings (AN 6.10: santi devā cātum(m)ahārājikā, santi devā tāvatiṃsā; SĀ 931: 有四大天王, 三十三天; SĀ2 156: 所謂四天王, 三十三).

The reports of the recommendation by the Buddha to look at the beautifully dressed young men arriving in their chariots in order to acquire a visual impression of the Celestials of the Thirty-three come without any explicit indication that this was for purposes of recollection. Nevertheless, according to the Pāli commentary, one of the reasons for this instruction to acquire a visual idea of the glory of the Celestials of the Thirty-three was to rouse energy in monastics who are weak in this respect (Sv II 545). In view of this commentarial suggestion, it is perhaps not too far-fetched to allow for the possibility that this encouragement had a purpose similar in kind to recollection of celestials.

Such a relation holds quite definitely in later texts for modalities of the first of the six recollections, which takes the Buddha as its object (Harrison 1978, 1992). A central purpose of such recollection is to serve as a means of arousing faith and inspiration (Rotman 2009). An instruction in this practice, found in a discourse in the Ekottarika-āgama, speaks of contemplating the Buddha’s form without taking one’s eyes off (EĀ 3.1: 觀如來形, 未曾離目), which then leads over to recollecting his qualities. This is a later text, compared to the bulk of the early discourses, seemingly intending some form of a Buddha image to look at. However, what this text makes explicit may to some extent already have been implicit in earlier forms of recollecting the Buddha. Such practices could in principle have included a visual dimension in the form of some mental image, to be aroused in reliance on textual descriptions of the Buddha’s thirty-two marks. This could then result in what Kinnard (2004, p. 128) has aptly called “iconographic thought” about the Buddha.

Visualizing Impermanence

Another relevant passage concerns the tale of the former king Mahāsudassana/Mahāsudarśana, a previous life of the Buddha reported in a range of sources (Waldschmidt 1948, p. 205). A summary of the Pāli version provided by Collins (1998, p. 477) is perhaps the best way to introduce the main trope:

Ānanda says that the unimpressive town of Kusinārā is not a fitting place for the Buddha to die. The Buddha counters by telling at great length the story of Kusinārā in its glory days. The extraordinary number of scenes of wealth and enjoyment repeated throughout the story — there are any number of lists of 84,000 things (palaces, elephants, wives, cows), all of them repeated more than once as the story unfolds — seem clearly meant to prepare for the Buddha’s closing statement: “Ānanda, see how all these conditioned things are past, ceased, changed (into something else).”

The tale of Mahāsudassana/Mahāsudarśana is situated within the sequence of events leading up to the Buddha’s passing away. In the course of a groundbreaking study of the significance of this tale, Gethin (2006, p. 88) notes that “the story of Mahāsudassana’s life and (especially in the Pāli version) death forms a literary counterpoint to the story of the Buddha’s death.” In other words, the narrative is a teaching on impermanence related to the Buddha’s own impending passing away. However, it achieves this purpose in a rather specific way, which involves not only references to various aspects of meditation but also appears to be intended for visual enacting. Based on a survey of instances of such meditative dimensions, Gethin (p. 93) concludes that this discourse presents “a very straightforward and clear narrative of the Buddhist path.”

Of particular relevance to the visual dimension of the discourse are the rather repetitive descriptions of various features of the royal capital being made up of different precious materials. In relation to a similarly repetitive description in a Mahāyāna text, the Larger Sukhāvatīvyūha, Harrison (2003, p. 121) reasons: “What, then, if we read it differently, not as describing a world, but as constituting it, that is, as [a] prescription?” The same approach would facilitate reading such aspects of the tale of Mahāsudassana/Mahāsudarśana. Harrison (p. 122) proposes the following mode of reading:

Seen in this way, the passage passes from being static to being kinetic, since now we are ourselves creating and manipulating the images, setting them in motion. This gives us a new way of reading the text, as a template for visualization, the sheer detail of which now makes sense. What we are left with on the printed pages resembles the wiring diagram for a television set, of interest only to electricians, baffling and tediously complex to anyone else. But when we “do” the text rather than read it, when we perform its operations ourselves, it suddenly becomes a little more interesting … we get quite different results if we read … texts like it, not as descriptions of something already existing, but as blueprints for something which is to be constructed in the mind.

From a historical perspective, as noted by Gethin (2006, p. 94), “there can, I think, be little doubt that the descriptions of Mahāsudassana’s city are earlier and that they provide something of a template for the descriptions of comparable cities in the Mahāvastu and Divyāvadāna, and of Amitābha’s Pure Land.” In other words, the text under discussion appears to have set a precedent for such descriptions, or more precisely such visualization scripts. When viewed in this way, it becomes clear, as noted by Shaw (2021, p. 176), that the tale of Mahāsudassana/Mahāsudarśana “anticipates some of the great traditions of Buddhist meditative practice throughout central Asia, Tibet, China, and Japan.”

Visualizing Death

Another passage relevant for appreciating dimensions of early Buddhist orality that are related to visual images and visualization occurs in the Satipaṭṭhāna-sutta and its Chinese parallels. Among the modes of practice emerging from the passages surveyed thus far, the present instance stands out for being explicitly introduced as a mindfulness practice. In fact, it is part of the factor of right mindfulness in the noble eightfold path, which calls for a cultivation of the four establishments of mindfulness (satipaṭṭhāna, smṛtyupasthāna, 念處, dran pa nye bar gzhag pa). The relevant exercise is the last in a survey of different body contemplations. It quite vividly describes stages of decay a human body would go through if left out in the open. The meditator is then instructed to bring to mind the image of such decomposition and employ it to drive home the fact that one’s own body is similarly subject to falling apart.

As though one were to see a corpse thrown away in a charnel ground … one compares this same body with it: ‘This body is also of the same nature, it will be like that, it is not exempt from that fate.’

(MN 10: seyyathā pi passeyya sarīraṃ sivathikāya chaḍḍitaṃ … so imam eva kāyaṃ upasaṃharati: ayam pi kho kāyo evaṃdhammo evaṃbhāvī etaṃ anatīto ti).

One contemplates another’s corpse … having seen it, one compares oneself to it: ‘This body of mine now is also like this, it is of the same nature, and in the end it cannot escape [this fate].’

(MĀ 98: 觀彼死屍 … 見已自比: 今我此身亦復如是, 俱有此法, 終不得離).

One contemplates a corpse … one contemplates that one’s own body is not different from that: ‘My body will not escape from this calamity.’

(EĀ 12.1: 觀死屍 … 自觀身與彼無異: 吾身不免此患).

Commenting on the Pāli version of these instructions, Yamabe (1999, p. 7) notes that it involves “vivid visual images, and one may call this practice ‘visualization.’” However, “a concrete method of visualizing these images is not provided. It is not clear whether the practitioner needs to see a corpse literally or [if] simply imagining various stages of decomposition in meditation is sufficient.”

In line with instructions on actually looking at a corpse given in later exegesis (e.g. Vism 179), the Chinese versions give the impression that looking at an actual corpse is intended. This finds support in a set of verses in the Theragāthā, which describes a monk cultivating insight when seeing an actual corpse in decay lying in a burial ground (Th 393‒398). Another set of verses in the same collection reports how a different monk, after having encountered a similar sight, left the burial ground to sit and meditate, with the result of reaching awakening (Th 315‒319).

From a practical viewpoint, however, it would seem less probable to assume that each time a meditator wishes to do this type of mindfulness practice, a corpse has first of all to be found. Instead, as explicitly indicated in later exegesis, the idea would rather be that, whenever a corpse becomes available, one employs it to get a clear visual image. Later, one then simply brings to mind the visual memory of what one had earlier seen. The proposed interpretation does not appear to be only relevant for later exegesis but already for the early discourse. This can be seen from a passage describing the need to make an effort at protecting the sign (nimitta), in the sense of the visual image, that has arisen from seeing a corpse in various stages of decay (AN 4.14 and its parallels Allon 2001, p. 128, SĀ 879, and EĀ2 11; for a survey of different listings of such visual images of a corpse to be employed for meditative purposes see also Lamotte 1970, p. 1311). The description clearly reflects the fact that one needs to make an effort in order to ensure that the visual image taken from an actual corpse is still available later for meditational recall. This in turn implies that such a form of practice was already known at the time this description came into being.

The same can also be seen from a discourse in the Madhyama-āgama and its Sanskrit fragment parallel. The two versions agree in describing that a practitioner arrives at a place to view corpses in various stages of decay. The actual undertaking of meditative contemplation happens subsequently, after the practitioner has returned to the monastic dwelling, washed the feet, and sat down on a mat or seat (Matsuda 2021, p. 70: śayanāsanam āgamya pādau prakṣālya ca smṛtaḥ, mañcake sanniṣīded vā bṛsyāṃ vā pīṭhake’tha vā and MĀ 139: 還歸至本處, 澡洗於手足, 敷床正基坐). This confirms the impression that the exegetical tradition only makes explicit what would already have been implicit in the description of the cemetery contemplations in the Satipaṭṭhāna-sutta and its Chinese parallels, in that a visual sign taken from having seen an actual corpse will then be recalled when sitting in meditation.

Notably, the formulation employed in the relevant instructions in the Satipaṭṭhāna-sutta differs from its Chinese Āgama parallels insofar as it employs the expression “as though one were to see” (seyyathā pi passeyya). This may well be intended in the sense that one recalls the visual image of a corpse seen earlier. Alternatively, however, it could also be read to invite an act of meditative imagination. From a practical perspective, to leave open such an alternative seems a sensible option, in that a meditator may start this mindfulness practice by simply imagining a corpse. At least until it becomes possible to see an actual corpse, such visual imagination would serve as a meaningful substitute and can reasonably be expected to fulfil the purposes of the practice at least to some extent. In fact, Kloppenborg and Poelmeyer (1987, p. 83) comment that in order to “facilitate the emergence of the images, and to strengthen the experience of ‘reality’ of their vision, the meditators could visit cemeteries.” On such reasoning, the actual vision of a corpse would not be indispensable, although of course being desirable to strengthen the practice. Be that as it may, the practice of this exercise from the first establishment of mindfulness, found similarly in the parallel versions, does appear to amount to a form of visualization.

Other practices found under the same heading of contemplation of the body involve vivid imagery to illustrate the purpose of the respective meditation. In the case of contemplation of the body as made up of different elements, for example, the example of a butcher cutting up a cow illustrates the gist of such mindfulness meditation (Anālayo 2011, p. 84). For such cutting up to be employed for the purposes of illustration, it can safely be expected to have been a common sight in ancient India. Given that much, it seems fair to assume that someone receiving such instructions may bring to mind the memory of having seen a butcher cutting up a cow. Doing so would certainly enhance the power of the simile. From this viewpoint, then, it is perhaps less surprising to find a description of this type of meditation in a later text, the so-called Yogalehrbuch, encouraging the actual visualization of a butcher (Schlingloff 1964, p. 165). As noted by Yamabe (2021, p. 13), “a vision of a butcher appears to the meditator, slaughters a cow, and divides its body in front of him. Then the meditator imagines his own body slaughtered and divided in a similar fashion.” Such dividing up of one’s own body, mirroring the butcher’s division of a cow, in order to gain insight into its composite nature, would offer an approach, albeit a drastic one, for actualizing the purpose of this mindfulness exercise in the Satipaṭṭhāna-sutta and its parallels.

A similar tendency of active visualization being applied to one’s own body can be seen in relation to the main practice under discussion here, the contemplation of a corpse. According to Green (2021, p. 27), in some exegetical texts such practice still requires looking at an actual corpse or an image of it: “The meditator first contemplates it, grasps its image (nimitta), and then imagines his or her own body in a similar condition.” In later texts, however, the same basic practice becomes “a purely imaginative exercise, without the need for an external corpse of any kind.” On adopting this approach, “a meditator simply imagines his own body transforming into a skeleton and then focuses his mind upon it.”

Returning to the early discourses, it is worthy of note that in this type of text it is only in relation to the contemplation of a corpse that an act of visualization becomes explicit. Since the mode of practice that emerges in this way involves the intentional and conscious generation of a mental image, rather than unexpected visions during meditation that have not been called up intentionally, it would qualify as a form of “visualization.” Moreover, according to Cousins (2003, 4) the “cemetery meditation on the stages of decomposition of a corpse is not recorded as a Jain practice and may well have been typically or even uniquely Buddhist at this time.” In this way, an apparently typical Buddhist employment of mindfulness takes the form of a visualization. The present exploration thereby shows once again the broad scope and remarkable range of meditative activities associated with mindfulness in early Buddhist thought (Anālayo 2020).