Among the seven ways of knowing, some are valid ways of knowing and some are invalid. The valid ways of knowing are bare cognition and inferential understanding and all the rest are not valid. “Valid” means that it is fresh and non-fraudulent, and “non-fraudulent” means both accurate and decisive. An apprehending cognition is also non-fraudulent but doesn’t require the qualification of being fresh. Apprehending cognitions include not only bare cognition and inferential cognition but also subsequent cognition. Apprehension can be either explicit or implicit, depending on whether or not a mental hologram of the involved object arises and appears in the cognition. Explicit is when it appears in the cognition and implicit is when it does not appear.
Conceptual Cognition: Categories as Appearing Objects
The next division that we need to understand is between conceptual and non-conceptual cognition. First of all, conceptual cognition occurs only with mental consciousness, whereas non-conceptual cognition can occur with either mental or sensory consciousness. A conceptual cognition is a cognition of something through a mental category as the appearing object. The appearing object (snang-yul), if we imagine it graphically, is the one that is directly arising in the cognition, as if it were directly in front of the consciousness. In conceptual cognition, the appearing object is a mental category, which, according to the Sautrantika explanation, is a metaphysical entity (spyi-mtshan). It’s not something objective (rang-mtshan). In other words, it doesn’t exist outside of the conceptualizing process. There is no category sitting over there on the other side of the room and that we can see. There is a wall or a person over there, but there aren’t the mental categories of “walls” and “persons.”
A mental category is something that arises only in conceptual cognition and it is static. It is not something that changes. We can replace the composite defining characteristic of a mental category with another one, perhaps a more precise one, but the mental category itself didn’t grow or arise organically like a plant or a body or an understanding. It is just a fixed category. For instance, we might consider only human beings as belonging to the category “persons,” but then we learn that Buddhism considers all sentient beings as persons and so we expand the composite defining characteristic of the category “persons” so that it also includes non-human sentient beings as well, such as animals. There are also audio categories, where many sounds fit into the category of all being the sound of the same word. There are also object categories, as in all varieties of apples fit into the category of “apple.”
As humans, we have language associated with audio and object categories. Some animals also have language, but the cow doesn’t have to verbally think “barn” in order to know that whatever the angle it is viewing from, what it sees is the barn and it knows to go into it. Animals validly know such things because they too think conceptually. One of the problems is that we have the word “think” in our Western languages and we take it to include only verbal thinking and we restrict that to humans. There isn’t a commonly used word in our Western terminology that would also apply to the conceptual cognition of animals.
Conceptual Cognition: Conceptual Isolates and Mental Representations
The category in conceptual cognition is equivalent to a conceptual isolate of everything that does not share its composite defining characteristic (blo'i gzhan-sel). Like the category, the conceptual isolate is also a static phenomenon and is a type of specifier (ldog-pa, isolate), what I sometimes call a “nothing other than.” Although a category arises as a conceptual isolate in the mental activity of conceptual cognition, yet being static, it has no form. But, through it there also arises a nonstatic mental representation – again, a mental hologram – that represents the category. This is the conceptually implied object (zhen-yul) of the conceptual cognition and is a generic representation of all individual items that fit in the category.
For instance, when we think of, remember or imagine our mother, a representative picture may arise that represents her. The representation that we use to remember her might change each time we think of her, because what arises isn’t a stagnant photograph. What we imagine when we think of her also might be more or less in focus, or with different details.
We might also think of her by representing her merely with the mental sound of the word “mother.” In verbal thinking – which for most of us goes on a tremendous amount of the time – mental representations of the audio categories of words arise, with meaning categories, equivalent to object categories, associated with them. These representative mental sounds constitute the so-called “voice in our heads.” Of course, there is no separate “me” talking into a microphone at the desk in our heads, although it might feel like that. The mental sounds of “What should I say now? Everybody is looking at me” may arise, but there is no “me” separate from this mental activity that is the one talking in our head and worrying. To think that there is such a “me” is a total misconception of what actually is going on.
In the conceptual cognition of thinking of our mother, then, the appearing object is the category “my mother.” The involved object that the conceptual cognition engages with is the conceptually implied object, the generic mental representation of our mother. Such a representation arises from a habit built up, like a mental imprint, from having cognized her all her life. Conceptual cognitions do not have focal objects that cast an aspect of themselves on the mental consciousness and so are the external sources for the mental holograms that arise. The involved object is who we are thinking about, this mental representation of our mother. The representation could be a mental picture, or the mental sound of the word “mother” or the sound of her voice or whatever could represent “my mother” in our minds. The appearing object when we think about her or remember her is the category, “my mother,” which is a composite extending over every time we ever saw our mother or spoke with her.
Conceptual Cognition When the External Object Is Present
Conceptual cognition of something also occurs when the external object is present. When I look at Mary, for example, I see her. That is non-conceptual sensory bare cognition. Both the involved object and the appearing object is objectively Mary. In a separate conceptual cognition immediately following seeing her, I see her through the filter of the mental category I fit her into. That could be the category of remembering who she is, coming from the memory of an interaction with her in the past.
In fact, however, when I first saw her today, I didn’t remember who she was: I didn’t recognize her. Nevertheless, I was seeing a person with the name “Mary” and I had fit her conceptually in the category of “someone I don’t know.” Then, based on something embarrassing she said, such as “I am the one who invited you to give this seminar,” I recalled that this was the Mary who had invited me and whom I had seen yesterday at the airport when I arrived. Both before and after I recognized her, she remained the involved object of my sensory cognition, while a mental representation of her remained the involved object of a simultaneous conceptual cognition. But now I was perceiving her through the object category, “Mary, my host” rather than the object category “someone I don’t know.” Since she was right before my eyes, the mental hologram that arose in my conceptual cognition corresponded to the mental hologram that arose in my non-conceptual sensory cognition. But of course, that mental hologram changed every moment as she moved and spoke: I was not looking at a statue or listening to one.
Suppose I think about her later, after she has gone in the other room. I don’t actually see her any longer, but still there is a mental representation of her in my conceptual thought. If I have a really good memory, the mental image will look exactly like her posture and with the same facial expression during one of the moments when I saw her. This image will represent her over the entire period of time while I was seeing her, but of course she wasn't frozen in that position during the entire interaction. But, for most of us, we don’t have that kind of photographic memory. We can’t quite picture someone exactly the way that they looked in one specific moment. Still, there is a mental hologram representing what we saw.
Differences between Conceptual and Non-Conceptual Cognition
When we see Mary – or, more precisely, the colored shape of the body on which the person Mary is an imputation – we are also non-conceptually seeing what kind of person she objectively is: a human being, a woman, a young person and so on. We are, in fact, seeing a young female human. These facts are imputations on her body and, in turn, on her. Her body also objectively has properties (khyad-par), such as weight, height and so on. We are seeing a young female human who weighs something and is a certain height. But only when we conceptually cognize her immediately after seeing her do we fit what we see into the object categories of a human being, a woman, and so on, with whatever associations we might conceptually project onto these categories. Also, only in conceptual cognition, do we add qualities such as tall, short, kind and so on, and further categories such as “host.” When we simply see her, we don’t see her as tall or short, or as our host.
There is a big difference here between conceptual and non-conceptual cognition of Mary. With non-conceptual cognition, there are no categories. Even though there is no category of “Mary,” nevertheless, we are seeing Mary. That is an objective fact about this person: she is Mary, not Susan. Also objective is what kind of object we are seeing. We’re not seeing just colored shapes; we are seeing a commonsense object, a body – not a table – a body that can also be heard and known through the other senses. Further, we are not just seeing a body, but we are seeing a human being, a person, specifically a young female human, not a male giraffe. Which person are we seeing? Mary, an individual person with an uncommon defining characteristic and a name. We’re not seeing Susan. But we don’t cognize her as the young woman Mary unless we cognize her conceptually through the object category of this specific young woman designated with the name “Mary.”
Do you see the difference between non-conceptual and conceptual cognition? With non-conceptual cognition, we are seeing a person. We are seeing Mary. But, initially we don’t know that from the first moment of seeing a person. In order to know that it’s my host Mary, we need to fit her conceptually into the right category. The wrong category would be someone that I never met before.
Conceptual thought, then, is not just the voice going on in our heads. It includes the conceptions and pre-conceptions of the things that words refer to, which we need to make sense of what we are experiencing and communicating with others. Otherwise we couldn’t understand language.
Conceptions without Language
We can also have conceptions without language. A dog has the conception or category “my master.” Every time it smells a certain person, it fits it into this category and conceptually knows the person as “my master.” For the dog, “my master” is usually represented by a smell, but without a word for it. It doesn’t have to be represented by a visual image. Animals have conceptual cognition. After all, how does a baby lion know what is food? A baby lion learns just as a baby human learns.
So, how does a baby learn what is food and what is not food? Initially, a baby puts everything in its mouth, but eventually it learns to exclude what is not food and to differentiate food from not food. Otherwise, for the rest of its life that baby will need to stick everything in its mouth in order to know if it is eatable or not. But remember, categories, like the category “food,” are static phenomena; they do not grow organically. What the baby needs to learn is what items fit properly into the category “food” and what items to exclude. The same is the case with a baby lion. The difference is that a baby human eventually learns to designate the category “food” with the word “food”; the baby lion doesn’t learn a word for it.
Ascertaining the Items That Fit into a Mental Category
How do we know which items fit in which categories? We need to analyze carefully here. Some categories, like “food,” are naturally part of our biological make-up. No one needs to teach us that there is such a thing as food – we’re not talking about learning a word for it – and that certain substances are, in fact, food and we need to put them in our mouths, chew and swallow. We all need food to survive and instinctively we eat; and, as mammals, we have the instinctive need to suckle mother’s milk. The issue is what do we think fits into the category of food. For some species, like lions, they naturally include meat in the category of food, but not grass, for instance. For different species, different items are instinctively regarded as food. A cow instinctively eats grass, not meat. For humans, though, it’s not so straightforward.
The composite defining characteristic of the category “food” is something that performs the function of making the unpleasant sensation of hunger go away when ingested. Many things can do that, including both milk and rotten garbage. Although sucking on a rubber pacifier might satisfy a baby’s need for comfort, it does not make the sensation of hunger go away. So, through personal experience and trial and error, the baby learns to exclude the rubber pacifier, toys, and so on from the category “food.”
But how does it learn to exclude rotten garbage? Of course, if something tastes bad or burns the mouth, like chili peppers, the baby won’t eat it and will not consider it as food. But what about a cookie that has fallen in the dirt but still tastes good? The dirty cookie still has the individual defining characteristic feature that it will make the unpleasant sensation of hunger go away if eaten.
The baby needs to learn a subcategory of food – food that is safe to eat – which has the additional composite defining characteristic feature of something that does not make you sick when ingested. Determining whether something will make you sick or not is not so obvious. Some things will make you sick almost immediately and others, like junk food, will make you sick in the long run. There is also a distinction between what is safe to eat and what is healthy to eat. But let’s leave that aside for today.
Whether something that makes our hunger go away is safe to eat and will not make us sick requires recognition of the causal relation between eating something and making ourselves sick. How does a baby learn that? As an adult, we can learn that through trial and error from personal experience. We eat something that makes us sick and so we avoid eating it again. But for a baby it is difficult to understand the causal relation between eating the dirty cookie and getting sick. The baby needs to be taught the category “dirty” by its parent, that a cookie that has fallen in the dirt is an item that belongs to the category “dirty,” and that items in the category “dirty” cannot also be in the category “food safe to eat.” Learning all that might take time and require disciplining the baby, but the baby eventually learns. For adults to learn that junk food also needs to be excluded from the category “food that is safe to eat” is more challenging.
How Can We Look at a Flower and Have a Conceptual Thought about It at the Same Time?
That gets complicated. First, we look at the flower, which is non-conceptual bare visual cognition that takes a colored shape and a commonsense flower as its involved object. At this point, we don’t think “flower” or “how beautiful”; we have not yet fit what we see conceptually into the category “flowers” or “beautiful objects.” This non-conceptual bare cognition lasts just a tiny moment, but then, since our cognition of it is no longer fresh, our looking at it becomes subsequent cognition. It is still non-conceptual, but no longer bare cognition.
Then, before we think “flower,” we have a sixty-fourth of a second of non-determining cognition as our attention switches from the non-conceptual cognition to a conceptual. During the phases of bare cognition and subsequent cognition, our cognition of this object was decisive – we were definitely distinguishing this object and not something else from all the other objects around it. But now, our mental factor of distinguishing is turning toward which category to fit it in, and so we have this tiny fraction of a second of non-determining cognition of the flower.
When we now conceptually cognize the flower through the category “flowers” and possibly with the word “flower,” we are still looking at the flower, but with a separate non-conceptual visual cognition. The conceptual cognition is superimposed on the non-conceptual one, rendering the non-conceptual seeing as being slightly veiled. However, we wouldn’t say that we are no longer having non-conceptual cognition of the flower when we are fitting it into the category.
The simultaneous conceptual cognition of the mental representation of this object through the category “flowers” is what is called a semblance of bare cognition (mngon-sum ji-ltar-ba), not actual bare cognition. Although a mental hologram corresponding to the one that arises in the seeing arises in the conceptual cognition, the conceptual cognition is deceptive. It is deceptive because it appears as though all objects that fit in the category “flowers” looks like this. The mental hologram that looks like the external flower is taken as a generic mental representation of the category “flowers.”
The Role of Conceptual Cognition in Having Disturbing Emotions during Bare Sensory Cognition
After having a semblance of bare cognition of the flower, which is actually a conceptual cognition of it through the category “flower,” it is possible to have non-conceptual bare visual cognition of the flower once more without simultaneously conceptualizing about it. When we do, our understanding of it as a flower carries over to our bare visual cognition by the force of our previous conceptual cognition. This becomes very relevant for understanding how disturbing emotions come to accompany non-conceptual bare sensory cognition. They require conceptual cognition first.
Let’s analyze an example, greed for food. We look at the dessert we had for lunch and, initially, all we see are colored shapes and a commonsense object, the dessert. Our non-conceptual bare visual cognition of the dessert was not accompanied by any disturbing emotion, such as greed. Greed or attachment or desire only arises with an exaggeration of the positive qualities of an object and/or a projection of positive qualities that aren’t there.
Greed comes, then, only with the conceptual cognition in which we fit the dessert we see into the category of not only “dessert,” but also the category “the most fantastic thing in the world which, the more I eat, the happier it will make me.” This is an exaggeration of the good qualities of dessert and is an incorrect consideration (tshul-min yid-la byed-pa). We are fitting the dessert into an inappropriate mental box in which it doesn’t validly fit, “the most fantastic thing in the world.” That greed can then carry over into the non-conceptual bare sensory cognition of the taste of the dessert as we eat it.
Examples of Simultaneous Conceptual and Non-Conceptual Cognitions
We often have several cognitions occurring simultaneously. For example, when someone is speaking to us, we have non-conceptual bare audio cognition of hearing the sounds of the words they are saying, while we conceptually cognize them in the audio categories of words and the meaning categories of what the words mean. In this way, we understand what they’re saying. Actually, our conceptual cognition here is a type of inferential cognition. It is based on the line of reasoning: if I hear such and such sounds, they are the sounds of such and such words and have such and such meanings. This is called “inferential cognition based on conventions.”
Simultaneously with this inferential conceptual cognition, we are also seeing the person. Although the mental factor of attention is accompanying both our seeing and listening to them, the strength of our attention with each can vary. If the mental factor of distinguishing that accompanies our non-conceptual bare visual cognition distinguishes a certain look in their eyes and a certain expression on their face and a certain posture, we may also infer conceptually that they are tired. How do we know that? We know that by relying on the line of reasoning: “if someone looks like that, they are tired,” which we learned by personal experience. Relying on that inference, we continue to see them while also conceptualizing about them through the category “tired person.”
We might also conceptually infer they are tired from the aggressive, cranky tone of their voice. Because of that understanding, we will modify our behavior and not get into a big discussion or argument with them at that time because they are too cranky to be rational. Even more helpful is when we recognize in ourselves that we are too tired and have become cranky. Then, we follow the advice of Trijang Rinpoche to “put the baby to bed” when we are too cranky or annoyed to deal properly with something. We lay down, perhaps take a nap or go to sleep for the night, and undoubtedly, we will feel differently when we get up. We usually feel better and our mood has changed.
Other-Induced Cognition and Self-Induced Cognition
Before we went to bed we saw things very negatively and, in the morning, we felt better and saw things differently. How do we know which is correct? We use another valid way of knowing something: other-induced cognition (gzhan-la nges-kyi tshad-ma), or more fully: a valid cognition in which determination of its object (in this case, which one is correct) needs to be induced by another cognition. In other words, we know that to evaluate which is correct – and not just based on the reasoning: “It is correct because I think so” – we need to have another cognition in which we get further information. So, we check.
For example, suppose last night when we came home from work we thought, “I think I made a mistake at work today. I really messed things up in the office.” We conceptually fit ourselves into the mental category “horrible person” and became very worried, upset and in a foul mood. So, we went to sleep early, based on validly knowing that to determine whether or not we made a mistake, we would need to go back to the office in the morning and get more evidence. So, in the morning, when we have calmed down, we return to the office and check. In that way, we see, for instance, that we actually didn’t mess up at all. Now, with self-induced cognition (rang-las nges-kyi tshad-ma), we know that we do not need further information. Based on what we have learned at the office, we can determine that what we now conclude is correct. We conceptually cognize ourselves through the mental category “someone who did not mess things up.”
Similarly, if someone has said something to us that we thought was strange, we use other-induced cognition to know that before we come to a conclusion about what they meant, we need to ask the person for clarification.
Drawing the Correct Conclusion and the Emotional Response That Follows
When we rely on a line of reasoning and come to a conclusion, how do we know that our conclusion is correct? Again, that is not so simple. Faults in logic constitute a large topic studied in Buddhism. But there are other guidelines that we can use when we are not debating with formal logic. For example, in death and impermanence meditation, we consider that death will come for certain, but there is no certainty when it will come. We can draw two opposite conclusions from that. We can infer that death is the most terrible thing in the world. We then conceptually cognize death through the category “the most terrible thing in the world” and our response is that we get depressed. On the other hand, we can infer that we can never know when our time is up, so we better make best use of the opportunities we have now and not waste our precious human life. Then we conceptually cognize death through the category “an incentive not to waste my life, but to do something constructive.” Which is the correct inferential cognition?
Based on the general principle that everyone wants to be happy and nobody wants to be unhappy, if we just get depressed about death, then we did not draw the inference correctly. The incorrect conclusion of the inference was that it is hopeless, because the result of coming to this conclusion is that we are miserable. We become apathetic about doing anything positive. Correct inference leads to the conclusion that whatever we want to accomplish, it is best not to waste time but to do it now. Then we are not depressed about the uncertainty of the time of our death and we have a positive state of mind.
The emotional response is what goes with our way of knowing. We can’t say an emotional response is valid or invalid. The fact that we are going to die and are aware of it is the same regardless of which conclusion we conceptually come to and how we respond. The emotional response is a choice and it helps us to evaluate the conclusions we draw. One conclusion and response will cripple us and make us miserable, while the other will make us happy because we feel that we are making best use of our time. If we have well-developed discriminating awareness, we will be able to discriminate which is going to be beneficial and which is going to be self-destructive. Being in a depression, miserable and then we die: that’s not terribly helpful or fun.
These two opposite conclusions and emotional responses are both based on valid information. Regardless of how we conceptualize death based on this information, it is fact that death will come for sure and we can never know when it will come.
Emotional Responses and Valid Cognition
Our emotional response to any conclusion that we draw based on inferential conceptual cognition is going to be affected by the tendencies that we have, the strength of our disturbing emotions and other mental factors. For example, last night we were walking back from the restaurant and there was a person lying on the street. One response could be that we don’t want to get involved, for a variety of reasons, and so we could just walk by without caring. That is one type of emotional response, based perhaps on being in a hurry and not having time. Our cognition was accurate in terms of seeing that there was someone lying in the street and it was decisive: someone was really lying there. It wasn’t a mannequin from a store window; it was a human being. We didn’t know what was wrong with the person and, not caring, we could have not stopped to ask.
But since the tendency of compassion was strong with us, we validly knew that we needed to get further information in order to determine whether or not this person needed help. Any further response would need to be based on validly knowing whether or not they were hurt. The people that were with me, speaking Finnish, stopped and asked the person if they were alright. Their emotional response was concern. The person said that they were alright and the taxi they called for hadn’t come as yet and they were resting. They weren’t drunk either. In fact, they got up and walked away.
We then speculated that maybe it was some sort of an experiment to see if anyone will stop and offer help. That conclusion could have been completely incorrect, because we didn’t check. It seemed to us that it was self-evident that they were a journalist or something like that. We made up an entire story of why this person was lying on the ground and why they were able to get up and walk away after we asked if they needed help. Our inference could have been correct or incorrect. Now we will never know.
The question is: which emotional response was valid: being concerned and getting involved, or not caring and walking on past? After all, maybe the person was really just resting while waiting for the taxi, as he said, and so our asking if they needed help was totally unnecessary. They could have responded with telling us to mind our own business. But we concluded that showing concern and compassion was the better response, but we can’t say that it was a more valid emotional response than just walking on past. Certain emotional responses are appropriate, and others are not, depending on the situation. In many cases, however, we respond emotionally before we get enough information to be accurate and decisive about what we see or hear and what we infer. It was accurate and decisive that we saw someone lying on the ground. It was neither accurate nor decisive to infer that they were hurt or drunk.
Our emotional responses to what we see or hear, then, depend very much on what inferences we draw from the information we gain. Because some emotional responses cause us and others much suffering, being able to differentiate valid from invalid ways of knowing is very important.