Part III: Intrinsic Ontology of IIT >> Meaning & reference >> Intrinsicality of meaning
Part III: Intrinsic Ontology of IIT >> Meaning & reference >> Intrinsicality of meaning
I see a round, red apple on my kitchen table, in front of a fruit basket, with the window behind letting in an autumnal landscape. As a content of experience, ‘this apple’ is obviously meaningful. Where is its meaning to be found?
According to IIT, all meaning, including the meaning of ‘this apple,’ is always wholly intrinsic—it is always “in the head.” Dreams provide a vivid demonstration of the intrinsicality of meaning. I can certainly dream of a particular apple—‘this apple’—so convincingly at times that it seems real. I may see the apple on my kitchen table, in front of a fruit basket, with a backdrop of autumn colors through the window. Dreaming or not, I experience ‘this apple’ as a bundle composed of general concepts and particular features—a bundle that ‘hangs together’ well and tightly. I experience it as an apple (in fact, I cannot help but experience the name ‘apple’ in my mind), as a kind of fruit, as something edible, as an object occupying a region of space. I also experience the apple as occupying this particular location, with a particular contour that belongs to its particular roundish shape, and its particular shape bound to a particular color, to a smooth surface, and to the expectation of a firm texture.
The section on Modes of being argued that the feeling of space, time, and objects is due to the kinds of distinctions specified by the appropriate substrate and to the way they are related. Thus, a grid specifies reflexive distinctions corresponding to spots, bound by relations enforcing connection, fusion, and union, which yield the feeling of extendedness. From the intrinsic perspective, the feeling of extendedness is the meaning of the experience of space. Similarly, a directed grid specifies directed distinctions corresponding to moments, bound by relations enforcing directed connection, fusion, and union. These yield the feeling of flow, the meaning of the experience of time. Finally, a rooted tree specifies configurations and concepts, bound by relations enforcing horizontal and vertical (invariant and hierarchical) connection, fusion, and union. These yield the feeling of objects, which is also their meaning.
For all contents of consciousness, the meaning is the feeling—of spots and blots, of moments and occurrences, and of concepts and objects. If there is no feeling, there is no meaning. For a computer, the apple at the start of the chapter means nothing at all because it feels like nothing at all. The computer may well be able to recognize the apple and pick it up as well as I do. However, lacking the appropriate larger Φ-structure, the computer would see nothing, feel nothing, and understand nothing.
If the meaning is the feeling, then every feature of experience is meaningful. It follows that every meaning must be taken in the context of the overall experience. Of course we are well aware that context is critical. Even the letter ’A’ can mean very different things in different contexts. But the term context, as employed loosely in speech, is both vague and mysterious: vague because it is hard to know how far it extends and towards what, and mysterious because often we can only point to subtle associations, colorations, undertones, and the like.
In this respect, it may provide some precision and clarity to identify a meaning with a Φ-fold of high ΦR. Since a Φ-fold is necessarily part of a larger Φ-structure, the rest of the Φ-structure inevitably provides a global context to the local meaning of the Φ-fold. Moreover, the context of a Φ-fold can be defined precisely: it is given by all the distinctions that are related to it. The value of integrated information (Φ) could also be used to determine which Φ-folds are most related to the one of interest. Most of the time, for example, the strongest context would be confined within one modality. Also, the complex ways in which contents of consciousness are bound together—how an object such as an apple is bound with locations in space, moments in time, qualia, and other objects—can be characterized objectively by assessing causal distinctions and relations. While we are far from being able to do so, we can already have an idea of how even a humble spot on a canvas can have an immense context, and how it is organized. The other side of the coin, of course, is that even something as seemingly simple as a homogeneous, blank canvas is characterized by relations of extraordinary richness; hence the subtleties of meaning in thought, language, music, and art are beyond not just description but imagination.
Meaning holism in no way precludes the concentration of meaning in clusters or hubs, corresponding to Φ-folds that are well-defined local maxima of high ΦR, like ‘this apple’ I see in front of me. Indeed, it is only clusters of meaning—such as ‘this apple’ or ‘this face’—that we typically consider meaningful. Instead, we would consider as meaningless a jumble of unrelated features, such as a haphazard set of marks on the screen. Accordingly, ‘this apple’ should correspond to a Φ-fold of high ΦR—a proper content of experience that hangs together well—whereas a jumble of unrelated features would not correspond to a local maximum of integrated information.
As discussed in chapter 8, high-level concepts, such as ‘apple’ or ‘face’ have a special role in gathering together clusters of meaning corresponding to Φ-folds of high ΦR. They are pivots around which many levels of disjunction and conjunction converge, yielding dense Φ-folds corresponding to the experience of deep objects. High-level concepts are also ‘handles’ that allow us to manipulate entire meanings effectively through attention and introspection, and they are typically associated with words. If we see marks on a page arranged in an appropriate way, we may immediately recognize the letter “A”—we have a word and a high-level concept for it, ‘A,’ which helps lower-level concepts and relations to hang together well. By contrast, we have no way to manipulate together a haphazard jumble of marks strewn around the screen, nor do we have a name for it. The reason, of course, is that we do not have a concept for it.
In what follows, the discussion of meaning and meaningfulness will usually make reference to high-level meanings pivoting around high-level concepts, in accordance with common usage. However, we should not forget that the extendedness of space and the flow of time are just as meaningful, in accordance with the maxim that ‘the meaning is the feeling’ (or ‘the feeling is the meaning’).
In chapters 7 and 8, we also saw that the structure of experience, rich as it is, is hard to decompose through introspection. Even the experience of space, which is partially penetrable thanks to the flexibility of spatial attention, cannot be decomposed down to individual distinctions and relations. Qualia such as color seem to offer no handhold to introspection. Meanings pivoting around high-level concepts are also hard to ‘deconstruct’ from within. In general, the extreme compositionality of experience implies that meanings are to a large extent impenetrable, even when we have words for them.
If I try to decompose the meaning (feeling) of ‘this apple,’ I cannot proceed much beyond describing its shape, location, and color, and enumerating a few concepts that are bound to this particular apple, such as ‘apple,’ ‘fruit,’ ‘round,’ ‘ripe,’ ‘edible,’ ‘sweet,’ and so on. I can then try to enumerate associations, such as ‘autumn,’ ‘countryside,’ ‘health,’ and so on. I can think of its name, of how it sounds and how it is written; I can try and remember varieties of apples, how they taste and smell, and on and on, but the full extent of its meaning remains beyond introspection and description. A painter or a poet might still succeed in highlighting some relations that escaped my introspection, and that is a remarkable feat. But not even Proust could do justice to the “heady luxuriance” of an apple. Nonetheless, I experience its meaning in full simply by looking at it, even in a dream.
We saw in the section on sub-structures that inactivation, damage, or disconnection of a subset of units in the NSC leads to the ‘folding’ of the Φ-fold it would otherwise contribute to the Φ-structure. The loss of all the distinctions and relations associated with those units corresponds to the loss of contents of consciousness and therefore to the loss of their meaning. In the section on space, it was mentioned that lesions in visual cortical areas can lead to a restriction in the experience of space. For example, patients with left hemianopia due to a large lesion of right visual cortex only experience the left half of space. Intriguingly, however, they tend to be unaware of their problem: from their intrinsic perspective, space is definite and fully present, rather than missing the entire left half. Because all feeling and all meaning are gone for the left side of space, the patients have absolute agnosia for half of space: they absolutely do not know what one is talking about when one refers to the half of space they are “missing.”
As we also saw in the section on objects, patients with severe lesions within higher-level areas in posterior cortex can have absolute agnosia for objects. For example, lesions in fusiform cortex can lead to absolute prosopagnosia, in which patients lose not only the ability to recognize faces but even the very meaning of ‘face.’ Yet they can experience lower-level features of the face and even copy it faithfully without recognizing what they are seeing. Sometimes a transient agnosia of this kind may also occur in normal subjects upon awakening from sleep, presumably due to a transient inactivation of corresponding cortical areas. Thus, for a few seconds after waking up, I may see the picture in front of my bed as a mere jumble of colors, without realizing what it means—a still life with an apple. Another example is semantic fatigue (or satiation): if I repeat aloud a word such as ‘apple’ many times, it seems to lose its meaning, becoming an empty shell of sound. A possible explanation is that by exercising some neurons intensely, I may make them so “tired” that the intrinsic meaning to which they contribute is lost. In these cases, it is likely that units are deactivated in the high-level areas around which the Φ-fold triggered by a face, an apple, or a word was pivoting. This deactivation decapitates the top of the hierarchy, folding it into a smaller, less meaningful Φ-fold.
Just as we can lose certain intrinsic meanings, we can lack other meanings altogether. Consider what it is like to listen to a heated argument in a language that is completely foreign: I hear a meaningless jumble of sounds and understand nothing, even though I rightly infer that those sounds must be meaningful to other ears. The same is true if I attempt to read an indecipherable alphabet or an unfathomable equation. But after I have developed the appropriate concepts, the very same sounds, scripts, or symbols become full of meaning, like a familiar face. Indeed, it becomes impossible to divest them of meaning: there is no way not to understand my mother tongue or not to read an “A” printed on a page.
In other cases, we may lack some meanings so thoroughly that we cannot even realize what we are missing. For example, to a person endowed with synesthesia, certain colors may be inextricably related to certain sounds—say, the color red and middle G. But such relations among colors and sounds are something I can neither experience nor fully understand, likely because I lack a suitable neural substrate. A synesthete could then rightly consider me to be an absolute agnosic, an unfortunate being congenitally unable to experience and understand the color of sound.
Such realizations should prompt both tolerance and humility when considering the point of view of others, others who may lack some concepts we have, or have some concepts we lack. The intrinsicality of meaning implies that, in a fundamental sense, we are all tied to our NSC, limited by the borders of our Φ-structures—that we are all prisoners inside our vessels and cannot step out. But if we at least understand that our intrinsic view is limited, we can leave room inside for other views.