The City Parable

A Comparison of EĀ 39.4, AN 7.67, and MĀ 3

Introduction

I’m not aware of a detailed study of this particular set of early texts in English. I began editing my translation of MĀ 3 some time ago, which led to a deeper look at its parallels. That deeper look then led to the decision to translate EĀ 39.4, given how significant it is to understanding the origins of MĀ 3 and AN 7.67.

AN 7.67 is an oddity as a member of AN’s Book of Sevens. It contains a list of seven metaphors that makes its placement in the Book of Sevens ostensibly appropriate, but it also adds a second list of four items. Effectively, it has eleven items, which would make it fit better in the Book of Elevens. The existence of EĀ 39.4 and MĀ 3 places AN 7.67 in a larger context that suggests a developmental history for this sūtra. EĀ 39.4 is probably the closest of the three versions to the original. MĀ 3 represents the endpoint of its development into a middle-length sūtra. And AN 7.67 represents an intermediate stage of development that retained the older outline but expanded the list of metaphors.

In this study, I want to focus on three aspects of these three versions: 1) The list of defenses possessed by a frontier city, 2) the list of skillful qualities possessed by a noble disciple, and 2) the overall outline of the sūtra. These discussions will not delve into detailed analysis of the wordings found in these sūtras. Readers are encouraged to consult the translations of these texts, which are linked above.

1. The Frontier City’s Defenses

The table belows sets the three versions of the city defenses side by side. We can see that the version in EĀ 39.4 is substantial different than the other two, while the lists in the AN and MĀ versions are almost identical.

The items that are directly parallel with each other are in bold, and the items that are similar in meaning are in italics. For example, I consider the sturdy gates and the high walls in the two versions to be very similar since sturdy gates are of little use without walls around the city. All three versions have an astute leader, but they each call him a different kind of leader. Still, the leader is described as astute and talented and keeps the population of the city safe.

We can see here that EĀ 39.4 had items that were placed in both lists in the expanded version, showing that the four item list was not entirely fabricated. Rather, a couple items that could be considered types of food or consumables were placed in a separate list and then a couple more added to it to make four. The reason it needed to be four items was because it was designed to correspond to the four meditations (P. jhāna, S. dhyāna).

Moving those two items from the original list to a new list meant also that two more items were needed to keep a list of seven city defenses. The garrison and patrol paths were inserted to fill the gaps created by the move.

2. The Noble Disciple’s Skillful Qualities

When we turn to the skillful qualities that are metaphorically paired with these city defenses, we see a greater difference between EĀ 39.4 and the other two sūtras. But the same basic outline of a monk’s religious practice from start to finish can be discerned in both.

Again, there are a couple items in the two versions that are direct parallels, but the correspondence is much more vague overall compared to the city defenses. The strongest parallels are those involving learning the Dharma and the four meditations, which are called progressive (or higher) mental states in both EĀ 39.4 and MĀ 3. This is treated an alternative epithet for the four meditations in AN 7.67.

In the case of learning the Dharma, the third and fourth qualities in EĀ 39.4 have been combined into one in MĀ 3 and AN 7.67. The last item of the seven qualities in both versions represent wisdom, but they describe it in different ways: For EĀ 39.4, it is the ability to discern the aggregates, senses, and elements as well as dependent origination. For AN 7.67 and MĀ 3, it represents the discernment of impermanence and gaining the knowledge required to end suffering.

If it’s true that MĀ 3 and AN 7.67 are versions that have altered a list of seven qualities to create a list for each of the four meditations, then we might suspect that the four meditations originally were the sixth quality before wisdom. In that scenario, mindfulness would have been added to fill the blank created by moving them to a separate list.

Was the rest of the list revised as well to make it somewhat different than the one we find in EĀ 39.4? Possibly. But it’s also possible the two versions had already differentiated by the time the sūtra was enlarged. The lack of corresponding descriptions points to these lists being different from the start, unlike the descriptions of the city defenses. They would appear to be two different outlines of the gradual path, one beginning with becoming faithful as a layperson, and the other beginning with practicing the Vinaya as a renunciate. It seems likely one of the two versions was changed to fit a sectarian mold.

3. The Overall Outline of the Sūtra

The outline of these sutras is quite simple, yet it’s still worth noting an important difference and incorporating it into our overall conclusions about the historical development of this text.

The unique feature of MĀ 3 compared to the other two parallels lies in its overall structure rather than its content. It adds a section that presents the disciple’s skillful qualities and easy attainment of the progressive mental states separate from the metaphors of defense. This seems innocent enough, but the only purpose it serves is to lengthen the sutra overall and provide a repetition of these items. This adds more emphasis to the noble disciple’s abilities and provides more practice at memorizing them.

When texts in the Madhyama Āgama have parallels in AN and SN, they often show this type of expansion. Sometimes more related information is collected together. Sometimes additional repetitions are added. These changes may not have been motivated by a desire to add the text to a collection of middle-length sūtras but instead had the effect of making them candidates for such inclusion. Whatever reason it may have been, we can see that this sūtra reached its largest size in the hands of Sarvāstivādins.

This feature of MĀ 3 is more interesting when we look at AN 7.67. That version of the City Parable agrees with MĀ 3’s additions to the lists of defenses and qualities of a noble disciple but not in adding this extra repetition of the skillful qualities. It may be that AN 7.67 looked more like EĀ 39.4 before adopting the larger lists, or that the expansion of MĀ 3 happened after their adoption.

Conclusion

This is a good example of the development that’s evident in Buddhist texts when multiple recensions are available, whether they are EBTs or later works. These developments usually took the form of adding related content, increasing specificity, and adding structures designed to aid memorization. They also sometimes changed key doctrinal points and passages to fit later teachings. The latter cases are often most challenging since they beg the question as to which is earlier and which is later.

These comparisons also highlight the fact that comparing Theravāda and Sarvāstivāda texts is not always illuminating when it comes to discerning their historical development. Both corpuses represent late, well-developed canons that appear to have copied each other to one degree or another. It’s when we compare them to other sources like the Ekottarika Āgama that this becomes apparent.

Author: Charles Patton

Last Revised: 25 April 2025