The Intellectual Traditions of the East For Beginners, Part 6 | Nāgārjuna and Śūnyatā

NOTE: This is Part 6 of a series of articles on Eastern Philosophy. Read Part 1, Part 2, Part 3, Part 4, and Part 5. The individual articles in this series are written so that they are self contained but occasionally may refer to older articles for some details.

Having seen the history of the development of Buddhism and understanding the difference between a substance metaphysics and a Buddhist anti-substance metaphysics through dependent origination, in this article, we shall begin our study of the works of a very important Buddhist philosopher Nāgārjuna, who developed a more comprehensive critique of substance metaphysics. While his own school of Mahāyāna Buddhism enjoyed success in India, after the decline of Buddhism in India, its full flourishing happened in Tibet where Nāgārjuna is revered as “The Second Buddha”. Nāgārjuna is dated to the first three centuries CE and he was a brāhmaṇa, probably from the modern day region of Andhra Pradesh. Some of his famous works (or famous works that are attributed to him) include Mūlamadhyamakakārikā (Verses on the Middle Way), Yuktisastika (Sixty verses of reasoning), Śūnyatāsaptati (Seventy verses on emptiness), Vidalyasūtra (Devastating discourse), and Vigrahavyāvartanī (Reply to Objections). A large body of works is attributed to him and later hagiographical accounts assign to him an extraordinarily long lifespan to account for such a prolific output but modern scholarship attributes multiple authorship to that collection over a longer period of time. In this work, we shall primarily focus on some chapters from his most important work - Mūlamadhyamakakārikā while exploring some other works.

Nāgārjuna belonged to the Mahāyāna school of Buddhism, which if you remember from the previous article, has its sūtras in Sanskrit that are of mysterious origin. There were broadly three schools in Mahāyāna Buddhism and Nāgārjuna was the founder of the Madhyamaka school that was both chronologically and positionally the middle one. His works serve as the foundational text for the Madhyamaka school. The other two schools are the earlier Abhidharma school (from which Nāgārjuna broke to found the Madhyamaka school), and the later Yogācāra school. As usual, the foundational text Mūlamadhyamakakārikā (abbreviated as MMK in the succeeding sections) is very cryptic and in a sūtra format (although Buddhists use the word sūtra exclusively to refer to Buddha’s own sayings) and necessitates later commentaries to be understood by us. The chapters of our focus from Mūlamadhyamakakārikā in this article have lots of commentaries and interpretations - some of the most important among them are by Āryadeva (~2nd century CE), Buddhapālita (~3rd century CE), Bhāvaviveka (~5th century CE), Candrakīrti (~6th century), Avalokitavrata (~8th century CE), Tsöndrü (12th century - Tibet), Tsongkhapa (~14th-15th century - Tibet). In this article, I shall focus primarily on those aspects that all or most commentators agree upon along with a brief explanation of those aspects of disagreement, if required. The most influential among the commentaries is the one by Candrakīrti.

Nāgārjuna’s Śūnyatā - Empty of Svabhāva

The philosophical position of Nāgārjuna is summed up by a single Sanskrit word śūnyatā that means emptiness. This leads to the popular notion he is a nihilist who believes that nothing exists but this is a misconception. Nāgārjuna does not say that all things are absolutely empty - he says that all things are “empty OF SOMETHING”. What is that SOMETHING? It is the Sanskrit word svabhāva. He says that everything is empty of svabhāva. This Sanskrit word can’t be translated exactly into English and hence deserves an elucidation. A difficulty is the fact that the concept of svabhāva does not have any straightforward equivalent among the concepts discussed in Western philosophy. This is not to say that it is a fundamentally alien concept to the West, but merely that the Indian term svabhāva combines a number of notions which are regarded as distinct in the Western philosophical tradition. Thus, I will never translate this word and use it in its Sanskrit original. Literally at an etymological level, the word splits up into two parts - sva (self/own) + bhāva (existence). It has been translated into English as “essence”, “inherent/independent existence”, “own being”, “intrinsic nature” or “substance” depending on the context, because it indeed actually can mean all of the above notions. The underlying idea is that something has existence (bhāva) all on its own - independent of anything else.

One of the peculiar aspects of Nāgārjuna’s way of doing philosophy is that he does not argue for any position or viewpoint (dṛṣṭi) of his own (at least explicitly) as to why he thinks that all things are empty of svabhāva. He simply uses the various arguments used by his various opponents (who think that things do have svabhāva) and then finds loophole(s) in each of their own arguments and thus making the opponent’s arguments fall apart on his own account due to its own internal contradictions. This method is called reductio ad absurdum (also: method of contradiction) in Western philosophy and logic - if you remember in high school, this is how the square root of 2 is proved irrational - by finding internal contradictions among the implications that follow assuming that it is rational. This style of argument is called prasaṅga in Sanskrit. But there is a difference in the issue of interpretation. One set of commentators (Buddhapālita, Candrakīrti, etc) read Nāgārjuna as exclusively propounding reductio arguments that refute his opponents’ claim based on his own terms, while other set of commentators (Bhāvaviveka, Avalokitavrata) argue that in this process of doing so, Nāgārjuna, in the process of refuting various opponents on their own terms, ends up implicitly advancing independent arguments of his own, to support his thesis that all things are devoid of any svabhāva.

Coming back, when it is said that all things lack svabhāva - it means that no thing exists (has bhāva) on its own, independently of “anything else” - nothing can exist as a sole thing in an otherwise empty cosmos and still be what it is. What exactly are those “anything else”?

1. Parts: First, if something has parts, then it has no svabhāva (independent existence). Because, it cannot exist without its parts. A chariot has no independent existence because it depends on its component parts (wheels, axles, seat, etc.).
2. Causal: Even if something has no parts, but if it is causally produced, then it depends on its cause for its existence. A tree depends on the seed (that it previously was) for existence. Hence, it has no independent existence.
3. Conceptual: Could a thing that is indivisible and uncaused, still fail to possess svabhāva? Yes, if it depends on something that is conceptualized in our minds. If we cannot conceive of that thing’s existence as it is, without any of our conceptualizing minds present in this cosmos, then it still can fail to possess svabhāva even if it has no parts and is self caused. For example, the number 9 is an idea in my mind - there can exist 9 apples or 9 people or 9 mountains but the number 9 itself is an idea that depends on my mind for existence. So, it has no svabhāva.

Thus, if someone claims that something has svabhāva, all that Nāgārjuna simply does is to show that it is either breakable into parts, or is causally produced, or is conceptualized in the mind. He does not provide an independent explicit argument to show that nothing has any svabhāva. He simply turns the other people’s arguments that things have svabhāva upside down by simply attacking them with either of the above three analyses. The opponent that he is focused on the most is his predecessor Abhidharma school.

The Abhidharma school that preceded Nāgārjuna, accepted the first of these three criteria - i.e. No composite thing has independent existence as its existence depends on the existence of its parts. If we continue breaking down everything to its constituent parts, we must arrive at the maximally simple objects which have no parts. In Buddhism, such objects are called dharmas (note that dharma already means lots of things in Hinduism but in Buddhist usage, it specifically means an indivisible object). The scientific discovery of atoms and later of the subatomic particles, are a part of the quest to find these ultimate dharmas of reality. The ancients obviously did not know modern particle physics and believed in four or five element components for the material world (air, fire, water, earth, ether) and hence talked about atoms of air, atoms of fire etc for constituents of material objects. The various sub-schools of Abhidharma Buddhism contain many lists of such dharmas. Buddhists use the term dharma to refer to any such simple atom that cannot be broken down into its constituent parts. Until this point, Nāgārjuna goes with the previous Abhidharma school which also agrees that any composite object made up of many parts does not have any independent existence and breaks everything down to their constituent dharmas. Now, having done so, do these dharmas have independent existence, just because they have no parts? The Abhidharma school does agree so but it is here when Nāgārjuna goes further. All that he needs to do is to prove that these dharmas have no independent existence in the sense of the second or third criteria to prove that nothing has svabhāva. Nāgārjuna gives many arguments to prove that these dharmas have no independent existence. The Abhidharmas broke down composite things into their constituent parts to disprove their independent existence. But a dharma by definition has no parts. How will Nāgārjuna break a dharma down further? He will break it apart into its properties and see.

Decomposition of Dharmas - Property Tropes

Consider a dharma which is a substance that is indivisible into parts - like an electron that is indivisible at least according to the current scientific knowledge. (Of course, Nāgārjuna and his commentators did not know about electrons or photons, so he uses fire atoms and water atoms - the prevalent atomic theories of their day.) An individual electron that has a number of properties (its charge is so and so, mass is so and so, spin is so and so, etc.). Now Nāgārjuna asks this crucial question - what is the relation between an individual electron and all its properties that define it?

Question A: Is an individual electron something distinct from the combination of all the properties that are observed from it? (i.e. Is an individual electron something more fundamental upon which all its properties are attached to? Or is an electron just a bunch of all its observable properties?)

Answer A (if NO): An individual electron is simply a bunch of all its properties - Is an individual electron distinct from the combination of all the properties that are perceived from it by our minds? - If we assume NO, then we see that an individual electron is just a collection of a bunch of all of its properties. That is, all I see is charge -e, mass so and so, and spin so and so - there is no such thing as an individual electron - it is just a name for a particular coincidence/co occurrence/coperception of all of these individual properties of mass, charge and spin. If it is so, then an electron decomposes into a bunch of its properties in the same way that a chariot loses its independent existence when analyzed into parts. Just as the aggregate of a particular collection of four wheels, seat, etc, designates a conventional notion of a chariot, a particular aggregate of a bunch of properties (charge = something, spin = something, mass = something) designates a conventional notion of an electron. So, what we assumed as indivisible (an electron) is now divided and hence an electron is not a dharma - each of the electron’s observed individual properties now become the real dharmas. So, in this picture, there are no individual electrons - what we call conventionally as an electron is just a bunch of its various observed properties - the technical jargon name for an individual manifestation or occurrence of a particular instance of a property (because it belongs to a particular substance) is called a trope.

A note on tropes: The fundamental idea here is that the redness of an apple is not regarded as one thing inhering in different red objects. The redness of the apple, that of a tomato, and that of a postbox are rather regarded as three distinct property-particulars which are sufficiently similar to be classified under the common name “red.” These three distinct property-particulars (redness of an apple, redness of a tomato and the redness of a postbox) are what are called tropes. So, in this conception and worldview, there are no substances having properties. There is no such thing as a tomato. What we conventionally name as a tomato is simply a combination of its tropes - the redness of the tomato, the roundness of the tomato, the juiciness of the tomato and so on. Also, there is no such thing as redness in this worldview. Redness is merely the conventional name that we give to the aggregate of the redness of a tomato, redness of an apple, redness of a post box etc. So, in this trope worldview, all that exists are tropes which are particular properties of particular things. A thing is a conventional name for the collection of all of its tropes. A property is the conventional name for the collection of all of its manifestations in all the individual objects of the world. A “tomato” is basically an aggregate of its redness, roundness, juiciness, and so on. A “red” is basically an aggregate of redness of tomatoes, redness of apples, redness of post boxes etc.

So, in this worldview, each electron is a bunch of its tropes - its charge trope (which says that its charge is so and so), its spin trope (which says that its spin is so and so), and its rest mass trope (which says that its rest mass is so and so). Like we call a collection of a particular set of wheels, axles, seats and so on as a chariot, we call a collection of a particular set of of charge trope (charge = so and so), mass trope (mass = so and so), spin trope (spin = so and so) that are observed by us, as an electron. But how do we tell apart a charge trope of one electron from the charge trope of another electron when all electrons have the same sets of charge, rest mass and spin. How is a charge trope having charge value equal to -e of one electron told apart from another? One can propose that individual electrons also have particular properties that differentiate one electron from another - like their location in space-time can enable us to tell apart charge trope of one electron from another. But wait - being at a particular location (in space) and being at a particular moment (in time) are also properties of an individual electron - so they also become tropes (let’s call them - place trope and time trope). Place tropes and time tropes are examples of what are called as individuated tropes - tropes corresponding to properties that are unique for a given particle. But properties like charge, mass, spin etc that a particle shares with another particle are examples of what are called as universal tropes. So, we say that a charge trope of one electron (say electron 1) is distinguished from the charge trope of another electron (say electron 2) because the charge trope for electron 1 is identifiable distinctly because it is co-present with the place trope of a particular location and the time trope of a particular moment (say X1, T1), while the charge trope for electron 2 is identifiable because it is co-present with the place trope of another location and the time trope of another moment (say X2, T2).

So, universal tropes corresponding to different individual particles are identifiable because they occur merged with the individuated tropes of their corresponding particles. This procedure allows us to individually identify the different universal tropes of different particles (by putting them together with different particular tropes) but then these universal tropes no longer have an independent existence. A charge trope of a given electron 1 (say Q1) depends for its existence, its co-existence with a space trope at the location X1 and a time trope at the moment T1. If none of the space and time tropes existed, then the very concept of charge trope of a PARTICULAR electron like Q1 - cannot make sense because without being co-existent with a space trope at the location X1 and a time trope at the moment T1, we can no longer call the charge trope Q1 as Q1 - the 1 in Q1 won’t make sense if Q1 is not co-existent with the place and time tropes of X1 and T1. Without the existence of a space trope at X1 and the time trope at T1, there cannot be a charge trope for electron 1 because it is the space and time tropes themselves that enable us to talk about charge tropes of various distinct electrons like 1 and 2.

So, without the existence of individuated tropes, the identity of universal tropes of various particles, like Q1, cease to exist. So, the independent existence of these tropes are violated and hence they possess no svabhāva. On the other hand, if we abandon all together, the notion of particular tropes and keep only universal tropes, then we are forced to conclude that a particular combination of the universal tropes constitutes a particular particle (say that a particular collection of charge, spin and mass tropes is designated as constituting a particular electron 1 by convention). Just like a particular collection of wheels and axles and so on constitutes a particular chariot. But this would mean that these tropes lose their independent existence because without the existence of mass tropes and spin tropes, the charge tropes alone cease to exist as we made the identity of each individual electron dependent on the particular combination of its universal tropes with which it occurs - and hence what makes a charge trope Q1 of electron 1, the charge trope of electron 1 (and not electron 2) is not something intrinsic to it but dependent on its combined co-existence with something else. So, these universal tropes of individual particles lose their independent lonely existence - their fundamental identity (whether that of particle 1 or 2) comes about only with reference to other tropes and not by anything that is from itself. So, if everything is a bunch of various universal tropes of various particles, then these universal tropes lack independent existence or svabhāva. Because their identity (whether they are of electron 1 or electron 2) depends on their co-existence in combination with other universal tropes or with other particular tropes. One cannot conceive of just charge of electron 1 without thinking of also the mass of electron 1 or the location of electron 1.

Decomposition of Dharmas - Bare Particulars

So, upon assuming an answer of yes to question A leads us to the fact that the most fundamental components of all that exist - which are tropes, lack independent existence or svabhāva. How about we assume the other alternative.

Question A: Is an individual electron something distinct from the combination of all the properties that are observed from it? (i.e. Is an individual electron something more fundamental upon which all its properties are attached to? Or is an electron just a bunch of all its observable properties?)

Answer A (if YES): An individual electron is more than a bunch of its properties. Is an individual electron distinct from the combination of all the properties that define it? - If YES, then we have all the characteristic properties of electrons on the one hand and then something else on the other hand - an individual electron is not simply just a bunch of its various properties (whether universal or particular) but has something individually unique. This extra element of individuality in philosophical parlance is called the bare particular - a kind of a pillow on which various pins (properties) are attached to. It is that thing (distinct from any property) that gives to each particular electron its individuality. What problem would come with assuming such a picture? The problem seems to be this. Let us assume that there was indeed a bare particular left over once we abstracted the property of charge, mass and spin from an electron. Assume furthermore that this particular would have its existence and its bare-particular-ness independently, intrinsically and essentially. But then, how can we know that it is the bare-particular of an electron and not that of another particle like a photon? You might say - well, it is the bare particular of an electron because its charge is -e but hey wait - we constructed a bare particular of an electron by separating it from its properties and abstracting its properties away - charge, mass and spin. In this case charge / mass / spin could not be the svabhāva of the bare particular as well - it would be superfluous. Even if you assume it also has charge or another property to do this distinction, then you can further split this bare particular into its constituent properties and come up with the notion of bare particular of a bare particular of an electron and go on ad infinitum.

So, a bare particular’s further characterization (by charge or anything else) would therefore be superfluous for establishing its status as a primary existent or leads to an infinite regress. Because for having independent existence, a bare particular should be able to exist independently without anything else. But here, the fact that the bare particular is that of an electron (and not that of a photon) comes from its association with its universal properties like charge, mass, or spin and hence lacks independent existence. Alternatively we could assume that the bare particular did not have its nature intrinsically and essentially but was dependent on something else. We could then ask again whether this something else has its property essentially, and so on. In this case we get into a regress which the opponent of Nāgārjuna has to terminate somewhere (no infinite regress), since he wants to establish that some objects (i.e., the true primary existents) exist by svabhāva and are therefore not dependent on anything else. We therefore end up with the first possibility again, as the various properties that make up the supposed svabhāva of the primary elements like the fundamental particles, and so forth are superfluous in characterizing these foundational objects as primary existents, since these objects are already existent as such.

But wait. We asked the question in the previous paragraph - how can we know that it is the bare-particular of an electron and not that of another particle like a photon? How about we go ahead and say that at the level of bare particulars, we cannot distinguish between the bare particular of an electron and the bare particular of a photon. How about we say that this bare particular only “exists” and is unflavoured by any property and that at this level, we cannot distinguish between any two bare particulars. Maybe, these bare particulars constitute the “pure stuff” of the world which is then “flavored” by such properties as charge so and so, mass so and so, and spin so and so, etc. in order to form electrons, photons and other particles (if this reminds you of advaita, you are thinking in the right direction - we will see explore these deeper connections as we progress through this series). Apart from the technicality that one cannot tell apart bare particulars of various particles, let us give this hypothesis a consideration - fundamentally, only these indistinguishable bare particulars exist independently. The properties flavor these bare particulars to give the world that we perceive. There is only one kind of bare particular at the fundamental level and they have only one property - that they exist independently (Does this remind you of the concept of sat सत् from the Upaniṣads - which is just bare existence? It is basically that). Nāgārjuna refutes this by using the third criterion of independent existence. Nobody sees these bare particulars. Everyone sees properties. Everyone sees anything with properties only. No one sees these bare particulars - their existence is inferred from observations by a mind and are hence concepts formulated in a conscious mind (like the concept of the number 9 or the concept of a triangle). So, these bare particulars depend on a mind for their conceptualization. Thus, they have no svabhāva (how the advaita responds to this - let’s wait patiently and see!).

So, to summarize all the arguments so far, we see that there is a fundamental problem when we assume that something exists independently (has svabhāva). Even if we take an indivisible atomic particle, then we can split it into its constituent properties and bare particulars (if assumed). Then, we see that the particle by itself cannot exist independently of its properties (existence of bare particulars led to paradoxes or mind dependence) and that properties are not perceived independent of the particles they are manifest in. In short, one can split any dharma into its constituent properties (like a chariot can be split into its constituent wheels, axles, seat etc…). But unlike the chariot whose parts have independent existence (the wheels can exist independently of the axles), the various properties and the bare particulars cannot exist independently of each other. The observed properties of a particle need to be either lumped together for constituting the individual identity to the particle (electron 1 here, 2 there, etc..) or should be attached to a bare particular. The bare particular that has just the individual identity of the particle needs to be lumped with its properties to give the particle its observed nature (so that we can say that it is a bare particular of an electron and not of a photon). So, nothing exists independently!

Recommended Reading: Chapter 5 of Mūlamadhyamakakārikā with the commentary of Candrakīrti.

Read next part.