Dependent Origination. Causes or conditions, sufficient or necessary?


Hi discoursers, (especially @Erik_ODonnell, @Martin, @Brahmali, @Javier)

Picking up on a thread in this topic (which is now closed) where this talk was brought up (part of this course).

I took a break from here, so I apologize, my reply is a bit late. (Only by about 8 months or so!) But I wanted to clarify two things:

  • on using “cause” in context of Dependent Origination, which I don’t think is always incorrect, but can imply more than what the Buddha was literally saying,
  • and on the kind of conditionality in DO, which I think is just “dependency on”, or, to put it in modern terms, just necessary conditions.

Underneath lies a more fundamental point about the language of the suttas being much easier than the kind of language we often see when Dependent Origination is discussed (including, unfortunately, in this very post).

I’d like to clarify that it was me, not Ajahn Brahmali, who is responsible for these slides. So all shortcomings are mine.

Yes, you have a point. But if you listen to the talk, I think you’ll hear me saying that people use “cause” in different ways. And I think “modern accounts of causality” is not what people use in daily speech, as you kind of attest when you say “only in the last decade became mainstream”.

I belief the Buddha taught in simple, everyday language. Such language I find in the Pali, and I feel incredibly fortunate to be able to read it directly. When I read on Dependent Origination, I feel the Buddha tried his very best to make it as theoretically simple and clear as possible. But then I read English explanations or even just translations of DO… and I often find the exact opposite! I find a lot of logic which, first of all, is what Dependent Origination is not about, and secondly, confuses it for people who I belief would be perfectly capable of understanding it if it was put in simper language. This “frustrates” me, and led me to present the things you replied to.

I mean, just look at all the different views out there on DO! That’s not just because of a lack of insight in people, it’s also a tendency to proliferate beyond what the the suttas are actually saying. So in the talk I attempted to get back to the basic ideas I find in the texts, and to not go beyond that. We can always elaborate from there, but at least it’ll be clear what comes from the Buddha and what comes from us.

So we are coming from different places here. You come from modern logic, I from what I think is more true to the original message of the Buddha. That’s doesn’t mean I think you are wrong, it’s more about what I feel is pragmatic for the majority of people and what is true to the Pali.

In that case, with birth and death, I agree it can be useful to use ‘cause’. But is it useful for every link? Do feelings cause craving? I agree with @Martin who said it’s better to say ignorance causes craving. And what about consciousness causes namarupa, and namarupa causes consciousness? There it gets even more troublesome for me. These things don’t really cause one another, in my understanding and daily use of the word. If they would, it sounds to me like the cycle is endless. These things come together, dependent on one another, but don’t cause one another.

The Buddha does use words that could potentially be translated as ‘cause’ (like hetu). But not in the core teachings on DO and definitely not in paticca/paccaya. Moreover, hetu can perhaps also be translated as ‘reason’.

But as you say, it is also a personal thing.

Thanks for listening. :slight_smile:

Given all the different views that exist, it may be quite rare to find two people agreeing on as much of Dependent Origination as Ajahn Brahmali and me. :smiley: Yet we did not agree on the sufficient conditions issue. Ajahn thinks there is sufficiency and necessity in all the links, whereas I think it’s simply just necessity. I agree there is sufficiency in some links, but not in all, and moreover, I don’t see this expressed in the texts. (Again, I try to get back to the most basic idea of the texts.)

First of all, necessity is what dependency is all about. If you depend on something, you need it. It doesn’t imply any sufficiency. I think the conditionality behind DO is as simple as that (on the textual level, anyway–reality is always more complicated). Paccaya means ‘dependent on’. So when the Buddha says “There will be B, only if there is A. B is dependent on A” (like in SN12.10), I understand the two sentences to mean the exact same thing, namely necessity. It’s just expressed in different ways.

Ajahn Brahmali disagrees and thinks there is also sufficiency. He said in the thread I linked to above: “‘When there is ignorance, there will be willed actions’ is a statement of sufficient conditionality.” Which I agree is true. But it’s true in English. I don’t think it is in Pali. This comes down to the difference between ‘if’ and ‘only if’ (or ‘when’ and ‘only when’) that exists only in English, not in Pali. (For the Pali nerds, see Wijesekera on Locative Absolute, which he states can “signify a condition that exists or should exist”, i.e. is necessary.) “There will be B if there is A” means something fundamentally different from “There will be B only if there is A”. The former is indeed a sufficient condition, but the latter is a necessary condition. I read the Pali as the latter. For example, I read a link as “there will be craving only if there is feeling” (i.e. you need feelings for craving). And then this link makes sense. But the translation “there will be craving if there is feeling” to me doesn’t make sense, because it implies you’ll always have craving, whether you are enlightened or not.

For Ajahn this is resolved by assuming no single link in DO applies to the arahant. Let’s for a moment assume that to be true. Then I still don’t agree, because even if you are not enlightened, you can feel, yet not crave. (An extreme example is in jhanas.) So even then feelings are not sufficient for craving.

But I think the assumption is incorrect. It implies ignorance lies behind each link of the origination sequence, so that in effect they say things like “if there is feeling and ignorance, there will be craving”. But I think each link is valid in and of itself, without having to imply ignorance.

To me this just makes sense, that every link should be a complete truth in its own right. You can compare it to the cessation sequence. Take any link there, it doesn’t matter whether you start from ignorance or not, they work by themselves (e.g. if craving ceases, fuel (upadana) will cease). And if we drop the concept of sufficiency, all the arising links work in isolation too (e.g. there will be fuel only if there is craving). The cessation and origination sequences are just two different ways of looking at the same concept of necessity.

This can be useful also from a pragmatic point of view. Take the mutual link between namarupa and consciousness. That link works, and is useful, even if you disregard ignorance. You can’t have consciousness without namarupa, whether there is ignorance or not. This is analogous to the link between contact and feeling. You can’t feel without contact, whether there is ignorance or not. Or birth and death. Birth (i.e. being alive) leads to death even for an arahant; death doesn’t suddenly disappear when there is no ignorance. I think these are powerful ideas that we lose if we assume no single link of DO applies to the arahant.

That ignorance doesn’t need to be implied in every link is also illustrated by suttas wherein the Buddha uses the “only if …” phrases, especially those where he starts his analysis at suffering, only coming to ignorance at the very end. Take SN12.10, where before his enlightenment he asked: When is there suffering? Well, only if there is birth. You need birth for there to be suffering. How can there be birth? Only if there is continued existence (bhava)… and so on … And how can there be willed actions? Only if there is ignorance. And only then, at the end, he had it all figured out, only then he thought of ignorance. He didn’t start at ignorance and then wondered what follows from it (which would imply he could see ignorance before he could see birth and death, which is of course not the case).

This idea is also supported by suttas that start the sequence at the feeling-craving junction, not mentioning ignorance (eg SN12.57). Of course ignorance is implied here too, in a sense. But the shorter sequence should also be sound by itself, I feel, and it just isn’t if we imply sufficiency in all links. Again, because it would say you’ll crave if you feel anything, even if you’re enlightened. I think this shorter sequence is actually very important to consider, because it is so close to the four noble truths, where craving is also the starting point. In other words, I think you can capture the essence of what DO is about without always needing ignorance. That’s what the four noble truths do, and that’s what suttas such as these do. There’s quite a variety of suttas that omit ignorance in the sequence, which I think quite makes the point.

Take SN12.67, which mentions the mutual dependence of namarupa and consciousness. If the dependence between these two were sufficient, it would be a never-ending cycle, where one leads to the other, leads to the other, leads to the other… Sure, one could assume ignorance needs to be there. But in this sutta also, ignorance is actually not in the sequence, which simply starts at this mutual dependency (after working back from suffering). The simile of the bundles of reeds also illustrates the fundamental idea is necessity, not sufficiency. The two bundles need one another to stay upright; if you put one bundle upright it is not sufficient for another to be also. The other doesn’t automatically stand up by itself.

The oft-mentioned four nutriments (e.g. SN12.11) are another clear image. Nutriments (say food for the body) are necessities, not a sufficiency.

Let me approach it from yet a different angle. Suppose you had ignorance your whole life but get enlightened tomorrow, so you won’t get reborn. That means somewhere in the chain between ignorance and birth sufficiency didn’t apply–and not just tomorrow when you get enlightened, but even today when you aren’t. After all, the ignorance that exists now, before your enlightenment, won’t lead to birth, so it isn’t sufficient for birth. This illustrates not all links can be sufficient conditions, because not all ignorance leads inevitably to birth.

So I think the links are not about sufficiency, and instead just say, in effect, that you need A for there to be B. That’s the necessary conditionality, that’s the dependency. There is nothing much more to it than that, on the linguistic and logical level, anyway. It’s simple, and easy to understand, just one single concept. Why infer more if we don’t have to?

That brings me back to my view that the Buddha tried to make things a theoretically simple and obvious as possible. When he presents the four statements on dependency (imasmiṁ sati idaṁ hoti, etc e.g. SN12.37), I read them as clarifications of a single underlying principle (dependency/necessity), whereas some others read in them various elaborations. They disagree on what these elaborations are, and how they work out exactly (which illustrates they may not actually be there). There are distinctions made such as: present moment vs future, causality vs dependency, and, in this case, sufficiency vs necessity. What I see happening here, is reading too much of modern logic back into the suttas, or else relying too much on translations. The only essential distinction I see in the four statements is origination vs cessation. But that is agreed upon by all, because it’s obvious.

If you have a certain idea of the Pali (or English), it can take quite some reevaluating and rereading to get a feel for how another idea may work. So I hope readers will consider the idea of necessity, and that translators will consider “only when/if” instead of “when”.

In short, I think the simplest possible interpretation is to be preferred, which is that the DO links are all just dependencies (i.e. necessary conditions). Sure, some links are also sufficient conditions, as is clear from their nature (i.e. birth is sufficient for death) though not for all (i.e. feelings for craving). But I don’t see this explicitly in the Pali. And if it’s not obviously there in the Pali, perhaps it’s best left out of the basic explanations, especially if it confuses things, which in my experience it does.

So if I’m right, and the Buddha didn’t make this distinction between sufficiency and necessity, then should we? I personally don’t think it’s all that helpful. But if we do talk about it, it should not be with reference to the Pali.

Anyway, I’m starting to repeat myself, and perhaps I’m just exacerbating the problem I’m trying to solve… :smiley: So if this is all more confusing than helpful, I’m sorry. But it’s exactly this kind of discussion I think people in the Buddha’s time didn’t have. Either because they didn’t have, or otherwise didn’t care about, the kind of logical constructs I’m arguing against.

Please let me know if these ideas resonate with you, or if you think I’m off track and why. I may one day turn the slides of the talk into a more worked-out writing, less rambling than this. Let me know if that interests anybody, and I’ll consider it, after the vassa, and probably after at least another 8 months! :wink:

(Sorry, I won’t be here to reply until November, after the vassa. Please somebody defend my case on my behalf while I’m meditating. :smiley: )


Hi, Bhante! :slight_smile:

I hope you experience my comments as constructive and not as some sort of scathing critique or anything :slight_smile: Dependence, condition, cause; I am happy with either and I am certainly not any authority to decide which is best.

I still want to argue for ‘cause’, because I think DO is an interesting and edifying topic, and maybe I will further my own understanding by doing so.

My feeling is that “modern accounts of causality” are much closer to daily speech meanings, because the modern accounts of causality is essentially a formalization of our basic causal intuitions.

I think this touches on views on how the human mind works. I would argue that causality is like the fish’s water, it’s so fundamental to our thinking that we don’t see its presence.

Like, we know that getting a diagnosis from the doctor isn’t what makes us sick, we know that the rooster crowing doesn’t make the sun rise. It’s inconceivable that what happens now should affect the past, yet we know that what happens now will affect the future.

When the Buddha says “It’s because of ignorance people get reborn” we know that ignorance is the cause of rebirth, stopping ignorance stops rebirth. When this is, that is; when that is not, this is not. Causality is just baked into our cognition and language IMO.

In English, there are tons of constructions to deal with causality: “It’s because of…”, “It was X that made it happen”, “It was X that did it”, “X is responsible for Y”, “It was X’s fault!”, “This wouldn’t have happened if not for X”, etc.

It’s totally possible that I am wrong though, and I hereby formally claim the right to be wrong! :grinning_face_with_smiling_eyes:

I think there’s an interesting difference between the “logic” of logic, and the “logic” of causality that’s worth considering.

A thing that has been pointed out by causality scholars is that formal logic is not really an appropriate language for causality. This is because formal logic cannot model the asymmetric relationships implied by causality.

To give an example, I’ll use ⇒ for a logical implication and → for a causal relationship.

A ⇒ B has the symmetry that ¬B ⇒ ¬A.

If we read DO in this way, i.e. as ignorance ⇒ .... ⇒ suffering, we may get into trouble, e.g., because the absence of any of the conditions downstream of ignorance implies that ignorance is not there. Thus, someone who is not craving is not ignorant. This kind of forces us down the road of momentariness/mind moments.

The whole point of A → B is to model the asymmetric relationships that are part and parcel of our daily lives.

A → B is asymmetric in the sense that it tells us that intervening on B will have no effect on A, but intervening on A will have an effect on B.

Flipping the light switch turns on the light, but smashing the light bulb has no effect on the position of the switch. We take this way of thinking so for granted it can be hard to see it’s there. Moreover, formal logic cannot model this asymmetry / directionality of cause to effect.

IMO, reading DO in a causal way solves a lot of the problems you detail regarding sufficiency and necessity.

From a causality standpoint, only an intervention on ignorance (removing it) produces the desired effect (no more rebirth). Jhana can be seen as a temporary intervention on craving. Rebirth in a high heaven realm can be seen as a temporary intervention on suffering.

Just to reiterate, DO as a causal sequence puts asymmetry at the forefront: different kinds of rebirth do not change one’s ignorance, but changing one’s ignorance does change one’s rebirth.

I hope you have an excellent vassa, bhante :pray:


Craving could only come from feeling, that way we could call it as feeling born craving or just craving

Saying feeling causes craving doesn’t work because arahant still have feeling yet still have craving

Furthermore I still believe that this dependent origination still apply to arahant because of its universality

Now birth causes death is still correct but saying death is only caused by Birth is no problem

I believe that partial dependency is what buddha is talking about because full dependency means you must cease feeling in order to cease craving

In case of birth and death relationship, it’s a full dependency you must cease birth to cease death

So among 12 links of dependent origination some are full dependency like birth-death while others are partial dependency like feeling - craving


@Erik_ODonnell: I hope you experience my comments as constructive and not as some sort of scathing critique or anything :slight_smile: Dependence, condition, cause; I am happy with either and I am certainly not any authority to decide which is best.

Definitely constructive. No need to be so defensive! There is no authority when it comes to translation, you know. :smiley: (Otherwise I’d apply for the job!) Neither is there “the best” translation. The best in what sense? :slight_smile: For me that would be, as I said, what’s “pragmatic for the majority of people and what is true to the Pali”, but that’s impossible to quantify, and those two factors don’t necessarily go together. So thank you for your input.

I think I got a bit lost, though. I don’t know whether we are discussing the same issue. I also don’t follow after you introduce the arrows. So let’s see if we can make our difference a bit more concrete, at least for me.

Perhaps we should distinguish between the noun ‘cause’ and the verb ‘cause’, which are used somewhat differently, I feel. My dictionary says for the verb, among other uses: “to actively produce as a result”. And it’s in this sense that I prefer not to talk about ‘cause’. For example, in this sense feelings don’t cause craving. They don’t actively produce it. Same with many other links.

That’s just one way to use the word ‘cause’, of course. That’s why I said in the OP: “‘cause’ in context of Dependent Origination, […] I don’t think is always incorrect, but can imply more than what the Buddha was literally saying”. We can see more in the word than just a condition. I do, anyway, and I think most people will. Do you see what I mean?

I was just scanning through Ven. Payutto’s Buddhadhamma that was posted here a bit ago, and by pure accident found this. Although judged by my very brief scan I don’t agree with most of what he writes on DO, especially the single-life take he has on it, he seemed to agree in some way on the issue of ‘cause’. Perhaps he can pinpoint it better than me:

Although the twelve factors are said to be interdependent and act as conditions for one another, this is not the same as saying they are ‘causes’ for one another. As a comparison, there are more conditions other than the seed itself that permit a plant to grow: soil, water, fertilizer, weather, and temperature all play a part. And these interrelated conditions do not need to follow a set temporal sequence. Similarly, a floor acts as a condition for the stability or positioning of a table. p157

Although Payutto does later use ‘cause’ and ‘causality’ himself (or rather, his translators), it is not with reference to any specific links. And perhaps that’s the crux of our disagreement. “Cause” works fine for me with most links, but not all of them. As pointed out above, and also by others in the original thread, if we say “feelings cause craving” we can miss the point. Or does namarupa cause the six senses? To me that doesn’t really make sense with how I feel about the word cause. The two just go together, one dependent on the other, like Payutto’s floor doesn’t cause the table. I brought this up in the OP also, but with other links. If you could reply to that, maybe it’ll help us find exactly where we disagree.

@Erik_ODonnell: I will further my own understanding by doing so.

Me too.

Talking about “understanding”, let me clarify for the sake of others (I should have done this earlier): We’re only discussing on a textual and theoretical level. You’d be perfectly able to understand DO in reality, yet not know what the Pali means, or how to put it into English the “best” way. Two arahants may well disagree on how to translate it, or on what kind of necessary/sufficient conditions there are, if any! Language is not the same as reality. I want to put this out here because I don’t want some people to get scared by all the theory and linguistics. The actual insights are not of such a nature at all.

Still, to find suitable translations that speak to people, I feel is a good thing to do. So thanks for helping me with this. I see some of the points you made before, and will take them into account in the future. I perhaps should not have made a strong separation of ‘cause’ VERSUS ‘dependent’, and nuance it a bit more. (But then again, nuance is boring! haha) Sorry if I caused unnecessary confusion.

1 Like

IMO, if each link is conceived of as a Process (think .*.exe :slightly_smiling_face:) , rather than as a factor/ cause, the inherent tendency to attribute ‘thingness’ is much reduced. It is then easy to conceptualize Ignorance as a process going on over many lifetimes, with no discernable beginning - yet being sustained by further underlying processes such as the hindrances, wrong attention, bad friends, etc. It is also easy to see how certain processes such as feeling must not only run, but also reach a certain level of maturity before the next level process of craving can get started. If either of contact.exe and feeling.exe fail to initialize, craving.exe cannot complete bootup… :rofl:

(Just my idiosyncratic way of thinking… if it helps, good - if not, never mind!)


I actually wrote various long-ish replies but the more I think about it, for me it boils down to “what do you mean ‘condition’ or ‘depend on’?”

The only way I can make sense of those words is if they refer to causality. In the DO context, they surely don’t mean ‘condition’ as in the ‘terms of and conditions’ of a contract, or ‘depend on’ like ‘I depend on my friends for emotional support’?

I guess I just find it hard to understand what these words mean in the context of DO if it’s not causality :man_shrugging:

1 Like

Paticca Samuppada isn’t simple arising. I think conditionality is correct, not mere direct causality. And the mode of conditionality should be understood as in Patthāna.