I discussed different models that conceive the brain mainly in an empirical context (chapters 1–3). That was complemented by developing a spatiotemporal model of consciousness (chapters 4–8). The spatiotemporal model of consciousness emphasized the central role of the world–brain relation. How the world–brain relation characterizes the existence and reality of brain and consciousness remains unclear though.
The brain is usually considered the subject of empirical observation in neuroscience. In contrast, the brain as such is not considered the subject of philosophy. For example, there is no well-established “philosophy of brain” (Northoff, 2004) as distinguished from, for instance, “philosophy of mind” (Searle, 2004). Unlike the mind, the brain has not yet been intensely scrutinized in epistemology and ontology. Instead of following the traditional path and starting with the mind, I here aim to do the opposite—namely, to develop an ontology of brain. Such an ontology of brain can, in a second step, serve as stepping-stone for an ontology of consciousness (chapter 10).
The main aim in this chapter is to develop an ontology of brain that is empirically plausible, that is, in accordance with the empirical data as discussed in Parts I and II of the book. I will argue for an ontological definition of the brain’s existence and reality by relation and structure as developed in structural realism (SR). It shall be noted that I use the concept of ontology in a certain sense as will be outlined briefly in the next section within this introduction.
I will develop the structural-realist characterization of the brain in the first part of this chapter. Hence I will discuss two arguments against such structural-realist view of the brain, the argument of individuation and the argument of time and space, and I will reject both. That allows me to maintain and support my claim of a structural-realist ontological characterization of the brain through world–brain relation. Such structural-realist characterization of the brain must be distinguished from its traditional ontological determination by elements such as physical or mental properties.
Definition of Ontology
What do I mean by ontology? Ontology is the study of being, and it deals with the categories of existence and reality. Ontology in this sense is often subsumed under or taken to be more or less equivalent with metaphysics, the question of Being (Tahko, 2015; van Inwagen, 2014). However, I do not take such a stance here. I carefully distinguish ontology from metaphysics—existence and reality, as understood here, are not mere instances of the more general “Being as such” as dealt with in metaphysics.
One distinguishing feature for now is that I use empirical data to support my ontological assumptions, something that is usually rejected in metaphysics as in analytic metaphysics (MacLaurin & Dyke, 2012) and metametaphysics (Tahko, 2015). Accordingly, unlike metaphysics, ontology is here not understood as operating on purely a priori, analytic, and conceptual grounds. Instead, my use of ontology includes a posteriori, synthetic, and empirical elements as being linked and coupled with the traditional a priori, analytic, and conceptual strategy. Importantly, that does not amount to a reductive methodological strategy though (as, e.g., in Anglo-American neurophilosophy; Churchland, 1986, 2002)—I focus on the linkages and transitions between, for instance, empirical and ontological domains (Northoff, 2014).
The clear distinction of ontology from metaphysics entails that I remain within the realm of the phenomenal as distinguished from the noumenal (as understood in a Kantian sense; Kant, 1781/1998; see chapter 13 for more details). Metaphysics targets the noumenal realm while ontology, as understood here, remains within the phenomenal domain. The distinction between metaphysics and ontology thus finds its analogue in the one between noumenal and phenomenal realms. All I am interested in is the phenomenal realm, that is, the world we live in and how the brain as part of that world is related to that very same world (i.e., the world–brain relation) and can thereby yield mental features such as consciousness.
What exactly do I mean by the concept of world? I determine the concept of world in a phenomenal sense through space and time. The world we live in is essentially spatiotemporal. That leaves open the exact nature of time and space as they determine our world. Therefore, in this chapter, I will put considerable effort into developing and outlining a proper concept of time and space—this is important in order to determine not only the concept of world itself but also the existence and reality of brain as a fundamental part of that very same world and its time and space. In contrast to the phenomenal world and its spatiotemporal features, I am happy to leave open the search for the noumenal realm (in a Kantian sense), that is, what is behind our world and its world–brain relation—this is the territory of metaphysics in general and the metaphysics of mind and mind–body relation in particular (see chapter 13 for more details).
My concept of ontology must also be distinguished from what is described as cognitive ontology (Poldrack & Yarkoni, 2016; Smith, 1995). In a nutshell, cognitive ontology takes features of human cognition (rather than language) as a starting point to characterize existence and reality. In contrast to such cognitive ontology, I here do not consider cognition as an ontological starting point—I therefore speak of spatiotemporal ontology.
Such spatiotemporal ontology integrates the brain within the world in spatiotemporal terms, that is, in terms of world–brain relation. Ontology of brain is consequently closely coupled to spatiotemporal ontology with both taking an intermediate position between metaphysics and cognitive ontology.
Finally, I shall briefly mention the notion of brain per se. The ontology of brain suggested here concerns the brain as a whole. The focus on the brain as a whole must be distinguished from the development of a taxonomy of specific mental and cognitive functions in cognitive ontology that focuses on specific parts and functions of the brain (Poldrack & Yarkoni, 2016). Accordingly, I detach my ontological determination of the brain from its cognitive functions and mental features (i.e., consciousness), as well as from specific parts of the brain. Hence, the ontological focus in this chapter is exclusively on the brain as a whole and its relation to the world, the world–brain relation, prior to and independent of its different parts or regions and their respective cognitive and mental features.
Part I: Ontology of Brain—Structural Realism
Structural Realism Ia: Ontological Priority of Relation over Elements
What is structural realism (SR)? SR highlights the central role of relations and structure. Either relata are included in conjunction with relations (moderate SR; Beni, 2017; Esfeld & Lam, 2008, 2011; Floridi, 2008, 2009, 2011) or relata are eliminated completely in favor of relations (eliminativist SR; French, 2014; French & Ladyman, 2003). SR has been discussed mainly in the context of physics (Esfeld & Lam, 2008, 2011; French, 2014; French & Ladyman, 1998) but has also recently been applied to information (Floridi, 2008, 2009, 2011; see also responses by Beni, 2017; Berto & Tagliabue, 2014; Fresco & Staines, 2014; Sdrolia & Bishop, 2014), cognitive science (Beni, 2016), the brain (Beni, 2016), and secondary qualities (Isaac, 2014). Finally, SR comes in an epistemic and ontological version. The epistemic version of structural realism (ESR) is the more modest one when claiming that all we can know are structure and relations. Importantly, this epistemic claim is not accompanied by ontological assumptions. ESR remains agnostic to the question of whether what we know really corresponds to ontological existence and reality independent of ourselves (i.e., ontic structural realism; OSR).
Structural realism highlights the notions of relation and structure. How can we define the concepts of relation and structure? Let us start with determining the concept of relation. One may want to distinguish two determinations of relation. In the first case, relation is supposedly constituted by the combination of and connection between different elements such as mental or physical properties.
Existence and reality are here ultimately traced to the basic elements rather than the relations themselves—this presupposes what I describe as element-based ontology with the supposition of, for instance, specific properties (or substances) such as mental or physical properties. Even if element-based ontology considers the notion of relation, it still claims ontological priority of elements, with relations remaining ontologically secondary at best. This is the most traditional form of ontology.
That is not the notion of relation as presupposed in SR, however. Here relations themselves are constitutive of reality and existence—there is ontological priority of relation over elements with the latter remaining ontologically secondary. SR claims that relations themselves are constitutive of existence and reality and are therefore ontologically most basic. Hence, the traditional element-based ontology is here replaced by what I describe as relation-based ontology. Relation-based ontology can be characterized by ontological priority of relation over elements—I will argue that such ontological characterization of relation is central for describing the existence and reality of brain.
Structural Realism Ib: Moderate OSR and Structure
How does OSR conceive the relationship between relation and elements or relata? Different versions of OSR have been distinguished with, for instance, noneliminativist or moderate versions on the one hand and eliminativist ones on the other (Esfeld & Lam, 2008, 2010). The eliminativist version of OSR claims that the relation itself is the sole basis and fundament of existence and reality with no role at all for the relata anymore (see, e.g., French & Ladyman, 1998; Ladyman, 2014).
In contrast, the noneliminativist or moderate version of OSR claims that relata still have a role but cannot be defined as such (and their properties) independent of their relation to each other—the relata thus no longer show any intrinsic properties (Esfeld & Lam, 2008, 2010). I here presuppose the moderate version of OSR, namely, that the relata themselves do not show intrinsic features, for example, elements or properties that, independent of their relations, define their existence and reality. However, at the same time, the relata still have a role besides the relation itself: the difference between relata (as between world and brain) is considered, which, unlike in traditional element-based ontology, is not traced to some intrinsic properties within the relata themselves (i.e., within world and brain) but to the relation holding between them (i.e., world–brain relation).
What about the concept of structure? The concept of structure can be determined as the combination and organization of different relations. There can be relation with distinct degrees of spatiotemporal extension. The concept of structure describes how the different relations, including their distinct spatiotemporal extensions, are related to each other and thus organized. Briefly, structure concerns the organization of relations (see chapter 11 for a more detailed definition of structure).
One empirical example of structure in this sense is the brain’s spontaneous activity and its spatiotemporal structure. For instance, one frequency (such as 10 Hz) is based on the relation (i.e., differences) between peaks and troughs across time. That very same frequency and its difference-based relation is now coupled to other frequencies (such as 0.01 Hz; i.e., cross-frequency coupling [CFC]; see chapter 1)—CFC organizes the different frequencies and thus provides structure in the brain’s spontaneous activity. Note that structure in the current context does not pertain to what is empirically described as anatomical structure and the different regions. Instead, structure is here understood in a strictly functional sense that determines the different parts (as empirically manifest in the determination of single regions’ neural activities by their functional connectivity).
Taken in this sense, the concept of structure needs to be distinguished from that of aggregate. The concept of aggregate refers to the mere addition or collection of different elements or relations without any organization between them. For example, one would then encounter a brain with different frequencies (i.e., difference-based relation) that are no longer coupled with each other (i.e., without CFC). There would still be relation but no structure anymore; that is, for instance, the case during the loss of consciousness as in a vegetative state, sleep, or anesthesia (see chapter 5).
Structural Realism IIa: Moderate SR and Brain—Difference-Based Coding
How can we apply SR to the brain? I will argue that the brain’s existence and reality can be defined by relation and, more specifically, the world–brain relation, rather than elements or properties within the brain itself. For that, we need to understand the brain’s coding strategy (i.e., difference-based coding) and its ontological implications.
I characterized the brain’s coding strategy by difference-based coding (see chapter 2). Briefly, difference-based coding refers to the encoding of neural activity in terms of statistically based differences between different stimuli. Taken in this sense, difference-based coding must be distinguished from stimulus-based coding, which refers to the encoding of neural activity in terms of single stimuli remaining independent of other stimuli. Empirical evidence, as discussed in chapter 2 (see also Northoff, 2014a, for more details), speaks in favor of difference-based coding rather than stimulus-based coding. Therefore, I consider difference-based coding as the brain’s coding strategy that constitutes and shapes its neural activity, including both stimulus-induced and spontaneous activity.
What does difference-based coding imply in an ontological regard for the brain’s existence and reality? I so far considered difference-based coding in a purely empirical sense. The concept of difference-based coding as stochastic coding strategy depends on observation (i.e., indirect observation) as related to our models of brain (see chapter 2). Taken in this sense, the concept of difference-based coding does not seem to carry any ontological implications, that is, concerning the brain’s existence and reality. That is not so as I will argue in the following.
The brain’s existence and reality can be defined by its neural activity. If there is no neural activity anymore, the brain is considered dead. That is empirically the case if one can observe a zero line in EEG, in which case one speaks of “brain death” (see chapter 5 and Northoff, 2016a,b). Accordingly, even if the brain as mere anatomical gray mass (i.e., as physical substance) is still present, the absence of the brain’s neural activity goes along with the absence of the brain. The brain’s existence and reality, as functionally meaningful, are thus determined by the presence of its neural activity rather than its presence as gray matter or physical substance.
How does such definition of the brain’s existence and reality stand in relation to difference-based coding? The brain’s existence and reality are defined by neural activity. That very same neural activity is based on and constituted by difference-based coding. Therefore, the brain’s existence and reality are determined by difference-based coding and, more generally, the differences as encoded into the brain’s neural activity during difference-based coding. Difference-based coding is thus not only empirically relevant in characterizing the brain’s coding strategy (Northoff, 2014a) but also ontologically relevant in that it determines the brain’s existence and reality.
Difference-based coding is based on the encoding of statistically based differences in terms of the relation between different stimuli; that is distinguished from the encoding of single stimuli as single elements as in stimulus-based coding. This presupposes ontological priority of relation over elements. Difference-based coding is thus quite compatible with the assumption of relation as a basic ontological feature as suggested in SR.
Accordingly, difference-based coding is relevant for the ontology of brain. This becomes even more clear when considering that difference-based coding allows us to encode the brain’s neural activity in relation (i.e., difference) to both body and world. The brain encodes the world’s events or objects in their relation to the brain itself (i.e., its spontaneous activity) into its neural activity. Difference-based coding thus allows us to establish a relation between world and brain, the world–brain relation, which, in turn, constitutes the brain’s existence and reality (i.e., its neural activity). This is compatible with the assumed ontological priority of relation over elements as claimed in SR.
Structural Realism IIb: Moderate SR and Brain—Threat of Logical Circularity?
One may now want to argue that the presumed ontological relevance of difference-based coding for the brain’s existence and reality amounts to logical circularity. Difference-based coding is a feature of the brain, and for that to hold the brain must already exist—difference-based coding presupposes the brain’s existence and reality. At the same time, I suppose that difference-based coding establishes the existence and reality of the brain by constituting its relation to the world, the world–brain relation. That is logically circular though: the brain’s existence and reality must be already presupposed (as the basis of the brain’s difference-based coding) to establish it at the same time (i.e., through difference-based coding in relation to the world).
To avoid such a threat of logical circularity, we need to define the brain by specific elements prior to and independent of its relation to the world as established by then merely empirical difference-based coding—we must thus revert to element-based ontology. Element-based ontology presupposes specific single elements such as physical or mental properties to underlie and determine existence and reality. Presupposing element-based ontology, the brain’s existence and reality would then need to be determined by specific elements rather than relation as in SR.
More specifically, the brain’s neural activity would need to be traced to single elements such as specific stimuli in body and world. That ultimately presupposes stimulus-based coding on the empirical level. Instead of encoding the differences between different stimuli and ultimately the difference between world, body, and brain into its neural activity, the brain would then encode single stimuli by themselves: it would encode the world independent of its relation to the brain into its neural activity. This amounts to what I described as stimulus-based coding as distinguished from difference-based coding (chapter 2). In a nutshell, element-based ontology of brain implies stimulus- rather than difference-based coding.
That is contrary to empirical evidence though. The brain shows difference-based coding rather than stimulus-based coding (chapter 2 and Northoff, 2014a). We therefore need to reconcile element-based ontology on the ontological level with difference-based coding holding on the empirical level. The relation between world and brain established by difference-based coding would then remain ontologically secondary at best while elements would still attain ontological priority by defining the existence and reality of both world and brain independent of their relation. This avoids logical circularity while at the same time considering empirical evidence.
However, we are then confronted with discrepancy between ontological presupposition (i.e., element-based ontology) and empirical characterization (i.e., difference-based coding). Though avoiding logical circularity and thus being logically plausible, the assumption of element-based ontology is not empirically plausible given that difference-based coding entails relation-based ontology (i.e., SR). I argue that we need to suppose SR rather than element-based ontology in order to allow for empirically plausible ontological assumptions that are in accordance with the empirical data (i.e., difference-based coding). However, this raises the threat of logical circularity.
Structural Realism IIc: Moderate SR and Brain—Relational View of the Brain
How can we avoid the threat of logical circularity? The threat of logical circularity is based on the assumption that difference-based coding cannot hold at the same time and co-occur with the constitution of the brain’s existence and reality by relating it to the world (i.e., world–brain relation). I suggest using the notion of difference in an ontological rather than empirical sense. By encoding its relation to the world in terms of differences (i.e., difference-based coding), the brain constitutes its existence and reality.
Difference in this ontological sense constitutes existence and reality rather than presupposing it (as is the case when using the notion of difference in a merely empirical sense). Such ontological determination of difference avoids the threat of logical circularity (which therefore is ultimately based on confusing empirical and ontological understandings of the concept of difference). Most important, the ontological notion of difference allows us to determine the brain’s existence and reality in a logically noncircular way by world–brain relation as constituted by difference-based coding. This amounts to what I describe as a relational view of the brain.
Such a relational view of the brain is quite compatible with moderate OSR. As with moderate OSR, I argue that the brain’s existence and reality depend on its relation to the world, the world–brain relation. The relation (i.e., the world–brain relation) is thus constitutive of the existence and reality of the functioning brain—this is made possible by difference-based coding that entails the concept of difference in an ontological (rather than merely empirical) sense.
At the same time, the concept of world–brain relation entails and acknowledges the distinction between world and brain: world and brain show distinct spatiotemporal scales, which, as I propose, is the very basis of their relation. Therefore, the ontological determination of the brain by world–brain relation and its relational view of brain is quite compatible with moderate SR. Moderate SR gives a role to both relation (i.e., world–brain relation) and relata (i.e., world and brain). In contrast, the relational view of the brain is not compatible with eliminativist SR that denies any role by the relata themselves (i.e., world and brain) and would therefore disregard the spatiotemporal distinction between world and brain.
Structural Realism IIIa: Conflation of the Notion of Difference
How can we determine the notion of “difference” in an ontological rather than merely empirical sense? This is even more important given that a possible counterargument about false inference from empirical to ontological levels may be raised. Let me detail this.
One may now want to argue that I so far did not really provide any argument for moderate OSR of the brain in terms of world–brain relation. Rather I merely stated my assumption and distinguished it from the alternative supposition, namely, the determination of the brain’s existence and reality by elements or properties. Even worse, the philosopher may want to accuse that I infer from the empirical level to the ontological determination of the brain.
This amounts to what can be called the empirical–ontological fallacy that historically can be traced to Kant and his characterization of Locke as a “physiologist of reason” (Kant, 1781/1998). Thereby, the concept of the empirical strictly conforms to observation as in science independent of whether any knowledge is acquired; hence the notion of the empirical is distinguished from that of the epistemic. The fallacy pointed out thus amounts to an empirical–ontological fallacy (rather than an epistemic–ontological fallacy; see chapter 14 for the latter).
Specifically, one may say that I infer from the empirical observation of difference as in the brain’s difference-based coding to the ontological level of relation that can also be determined by difference. I thus conflate two notions of difference: the empirical concept of difference as the difference between different stimuli in difference-based coding and the ontological notion of difference as inherent in relation. One and the same concept, that is, the concept of difference, is thus used and applied in both contexts, that is, empirical and ontological.
I suggested that the brain’s difference-based coding implies the ontological determination of its existence in terms of relation (i.e., world–brain relation). Those who take an opposing view may now want to argue that I inferred the ontological concept of difference as inherent in the notion of relation from the empirical one as in difference-based coding. Since the empirical level of observation and the ontological level of existence and reality are not identical, any inference from the brain’s difference-based coding to the brain’s existence and reality must be considered fallacious. That amounts to nothing less than an empirical–ontological fallacy (see figure 9.1).
In order to avoid such empirical–ontological fallacy, I must refrain from characterizing the brain’s existence and reality by difference and thus relation (i.e., world–brain relation). The ontological determination of the brain by world–brain relation must thus be rejected on conceptual or logical grounds. How can we escape the conflation between empirical and ontological notions of difference and consequently the empirical–ontological fallacy? I will argue that we need to distinguish two different concepts of difference, empirical and ontological; for that, I turn to Floridi (2008), who distinguishes between two distinct concepts of difference, the empirical notion of difference per se and the ontological concept of difference de re.
Structural Realism IIIb: Difference De Re versus Difference Per Se
What is meant by the concepts of difference per se and difference de re? Floridi determines the concept of difference per se in a purely empirical sense as the difference we can observe. For instance, we can observe a difference between two different brain regions and their neural activities—this amounts to difference per se. Yet another example of a difference per se would be that we can observe neural differences between spontaneous and task-evoked activity. Hence, the notion of difference in difference per se is understood in a purely empirical sense without any ontological connotations.
How can we determine the ontological meaning of difference (i.e., difference de re)? Difference de re means that existence and reality are based on difference rather than unity (or identity) of elements (i.e., nondifference). Taken in this sense, differences de re are the basic constituents of reality and existence and are thus ontological rather than empirical (or epistemic; Floridi, 2008). Therefore, the ontological concept of difference de re must be distinguished from its empirical counterpart, that is, difference per se.
How does Floridi describe his ontological concept of difference de re? He illustrates the ontological concept of difference de re using the example of marriage. Without the difference between two people (man and woman, woman and woman, or man and man) marriage would not exist—one cannot marry oneself. The existence and reality of marriage are thus based on difference (difference de re), that is, the difference between two people whose existence and reality as wife and husband are determined by their relation.
The ontological concept of difference de re is well reflected in the following quote by Floridi (2008):
However, the relation of difference is binary and symmetric. In the example, the white sheet of paper is not just the necessary background condition for the occurrence of a black dot as a datum; it is a constitutive part of the datum itself, together with the fundamental relation of inequality that couples it with the dot. In this specific sense, nothing is a datum per se, without its counterpart, just as nobody can be a wife without there being a husband. It takes two to make a datum. So, ontologically, data (as still unqualified, concrete points of lack of uniformity) are purely relational entities. (p. 220)
Note that Floridi’s ontological concept of difference is akin to the notion of differences as suggested by other philosophers in the European-continental tradition. This includes the Heidegger (1927/1962) (who introduces difference as distinguished from identity), Derrida (1978) (who speaks of “différance”), and Deleuze (1994, in his work Difference and Repetition). However, the exact details of such ontological notion of “difference,” including the differences between the different authors, are beyond the scope of this book.
Structural Realism IVa: Difference De Re and the Brain
How can we apply the ontological notion of difference de re to the brain? We can observe differences (i.e., difference per se) between two regions’ neural activities. If, now, the neural activity of each region is determined by its relation to the respective other, the regions’ neural activities are constitutively dependent on each other—this is indeed empirically supported by the data (see chapters 1 and 2). In that case the observable difference per se can be traced to and is based on an underlying difference de re in the spatial domain of the brain.
The same holds analogously in the temporal domain. The difference frequencies in the brain are determined in their power by their relationship to each other (i.e., CFC; chapter 1). If, for instance, their relation (CFC) is strong, the power of the single frequency is low while the power of the single frequency is high in the case of low CFC. Hence, the power of the single frequency is determined by and in dependence on its relation to others (i.e., CFC).
Yet another example is the relationship between spontaneous and task-evoked activity. As described in chapters 1 and 2, we can observe differences (i.e., difference per se) between both forms of neural activity. Moreover, the empirical data suggest that the task-evoked activity is constituted in dependence on and relation to the spontaneous activity that, conversely, is dependent on the former (see chapters 1 and 2 for details). Constitutively, spontaneous and task-evoked activity are thus mutually or reciprocally dependent upon each other—it is their difference (i.e., difference de re) that determines their respective neural activity levels.
One may now want to argue that I use the concept of difference de re in a merely empirical rather than ontological way. To use the concept of difference de re in a truly ontological sense, I would need to apply it to the brain as a whole, to its existence and reality, and its relation to the world, the world–brain relation. That is easily done. The measures cited above (i.e., functional connectivity, CFC, and the spectrum of relation between spontaneous and stimulus-induced activity) constitute not only the brain’s neural activity but, at the same time, its relation to the world, the world–brain relation (chapters 3 and 8). They can consequently be considered the empirical manifestations of the ontological existence and reality of brain in terms of difference de re.
If, in contrast, these measures were not reflecting the brain’s relation to the world, the world–brain relation, difference-based coding would no longer be ontologically relevant. The notion of difference de re would then no longer be distinguished from and thus collapse into that of difference per se. However, that is contrary to empirical evidence. That evidence shows the relevance of functional connectivity, CFC, and the spectrum of relation between spontaneous and stimulus-induced activity for establishing and constituting the brain’s relation to the world (as, e.g., in a rhythmic or continuous mode; chapter 8).
How can we illustrate the roles of world and brain in world–brain relation analogous to Floridi’s example of marriage? Marriage is defined by the difference between two people who, on the basis of their relation in marriage, are subsequently determined as wife and husband (or wife and wife or husband and husband). Analogously, world–brain relation is constituted by the difference between world and brain. The spatiotemporal difference between world and brain first and foremost makes possible their relation, the world–brain relation. That very same relation, the world–brain relation, determines, at the same time, the existence and reality of world and brain (in a noncircular way; see above), that is, as world as distinct from the brain as well as brain as distinct from the world. This is analogous to the way the two people are determined as wife and husband (or wife and wife or husband and husband) by their relation (i.e., the marriage).
Taken together, the ontological characterization of the brain in terms of SR presupposes difference de re as it features world–brain relation. The difference between world and brain, the world–brain relation, determines the brain’s existence and reality. Without that very same relation, the brain would not exist. The concept of relation in world–brain relation must thus be understood in terms of difference de re’s entailing an ontological rather than empirical meaning (as in difference per se).
Structural Realism IVb: Empirical–Ontological Plausibility versus Empirical–Ontological Fallacy
The introduction of two different concepts of difference, that is, empirical and ontological, makes any inference from the empirical to the ontological level futile and superfluous. We presuppose different independently generated concepts of difference in both empirical and ontological contexts (i.e., difference per se and difference de re). This allows us to feature the ontological level in an independent way by a specific concept of difference (i.e., difference de re) that distinguishes and makes it independent of the difference as understood on the empirical level (i.e., difference per se). The independence of the two concepts precludes a fallacious inference, that is, an empirical–ontological fallacy.
Those who take an opposing view may nevertheless want to argue that we still rely on empirical data to decide the ontological determination of the brain in favor of OSR. Though we no longer use the same concept, we nevertheless use empirical data to opt for OSR rather than element-based ontology in our ontological determination of the brain. The ontological determination of the brain thus remains empirically based. Such an empirical basis of the ontological determination of the brain must be rejected, though: the ontological determination of the brain must remain independent of its empirical characterization.
I argue that this argument needs to be rejected. The proponent of such argument presupposes that we cannot use any empirical data in our ontological determination of the brain. I reject this assumption: we can use empirical data in order to test whether our ontological determination of the brain is empirically plausible or not, amounting to what I describe as empirical–ontological plausibility (see figure 9.2).
The advocate of this argument confuses the concept of empirical–ontological fallacy with what I describe as empirical–ontological plausibility. She or he considers that any use of empirical data in ontological determination of the brain amounts to an inference from the empirical to the ontological level (i.e., empirical–ontological fallacy). However, such inference (empirical–ontological fallacy) occurs only when one uses one and the same concept on both the empirical and ontological levels in the same way—this is indeed the case when one does not distinguish empirical and ontological concepts of difference.
This, in contrast, is no longer the case if one uses different concepts and meanings of difference such as difference de re and difference per se to describe the ontological and empirical levels. In that case, one can investigate whether both concepts hold and are plausible on their respective levels—that is, empirical and ontological—independent from each other. If both concepts hold on their respective levels, one may consider the ontological concept of difference de re as empirically plausible while the empirical concept of difference per se can be regarded as ontologically plausible. Given the strong empirical evidence for difference per se in terms of difference-based coding, I argue that the ontological characterization of the brain by difference de re in terms of OSR is empirically plausible.
How about the opposite scenario with a discrepancy between ontological and empirical levels? In that case, the empirical data show stimulus- rather than difference-based coding entailing that there is no empirical evidence in favor of difference per se. One could nevertheless still characterize the brain ontologically by difference de re in terms of OSR, which, taken by itself (i.e., independent of the empirical data), may still be ontologically plausible. However, given the empirical data speaking against difference per se, such ontological characterization of the brain in terms of difference de re would no longer be empirically plausible. One would then rather revert to element-based ontology to account for the brain’s existence and reality in an empirically plausible way.
Part II: Ontology of Brain—Argument of Individuation and Argument of Time and Space
Argument of Individuation Ia: Individuation of Relata—Individuation by Spatiotemporal Structure
One of the arguments against OSR is that it fails to account for the individuation of objects or relata (Esfeld & Lam, 2008, 2010). A particular relatum must possess some intrinsic properties in order to allow for its ontological distinction from other relata. For instance, the brain must possess some specific properties that are intrinsic to the brain which allow us to distinguish its existence and reality from those of nonbrains within the world. More generally, this means that one must suppose element-based ontology to allow for individuation while that remains impossible in the case of relation and structure and thus OSR. I therefore speak of the “argument of individuation.”
The argument of individuation rests on the presupposition that individuation requires elements or properties, for example, element- and property-based individuation. In contrast, individuation on the basis of relation and structure as in OSR remains impossible. I reject the argument of individuation by showing that relation and structure and thus OSR can well account for individuation.
How can we counter the argument of individuation? Presupposing OSR, Esfeld and Lam (2008, 2010) reject this argument by postulating that the position of objects within the “web of structure” (Esfeld & Lam, 2006, p. 28; as analogous to Quine’s “web of beliefs,” Quine, 1969, p. 134) can identify their particular existence and reality, including their distinction from other objects. Individuation in this sense entails a central role for space and time: the individuation of the object by its position within the web of structure must be determined in spatial and temporal terms.
The conceptualization of “individual” may presuppose specific discrete points in time and space. However, these discrete points in time and space may be dependent upon spatiotemporal relations and thus the overall spatiotemporal structure. If so, any particular discrete point in time and space featuring a particular individual can only be individuated by its spatiotemporal relation to others and therefore by the overall spatiotemporal structure.
This individuation by spatiotemporal structure is reflected in the following quote about space and time in general relativity theory (GR) by Esfeld and Lam (2010):
On the other hand, the physical description of space-time within GR (and in particular the principle of active general covariance) makes meaningless any individuation of space-time points (with the help of intrinsic properties or of primitive thisness for instance) independently of the space-time relations they enter into or independently of the space-time structure they are part of—both being represented by the metric. (p. 22)
Argument of Individuation Ib: Spatiotemporal Individuation—Individuation of Brain by World–Brain Relation
How does the argument of individuation apply to world–brain relation? First and foremost, I am here considering the individuation of the brain as a whole and its distinction from the rest of the world, including nonbrains. Hence, I understand individuation in a strictly ontological sense. This must be distinguished from individuation of different parts within the brain such as its different regions and frequencies; that would concern individuation in a neuronal and thus empirical way rather than in an ontological sense as it is targeted here.
More generally, we must distinguish between two different forms of individuation with regard to the brain as a whole, empirical and ontological. “Empirical individuation” allows for individuation between different individual subjects, including their brains—the brain of subject A is individuated and distinguished from the brain of subject B within a particular species (such as the human species). In contrast, “ontological individuation” makes possible the individuation and distinction of brains from nonbrains within the world across different species. I now argue that OSR can account for both empirical and ontological individuation.
Let us start with empirical individuation. Empirical data suggest that the spontaneous activity’s spatiotemporal structure is highly individual for a particular subject and distinguishes it from other subjects (chapter 8). For instance, different individual subjects experiencing different degrees of early childhood trauma show different degrees of entropy, that is, dissimilarity or chaos, in their spontaneous activity’s spatiotemporal structure (see chapter 3 and Duncan et al., 2015, for details). Thus, entropy of the spontaneous activity’s spatiotemporal structure can individuate the individual subjects’ brains on the basis of their brains’ relation to the world and its (potentially traumatic) early life events, the world–brain relation. In a nutshell, the brain’s spontaneous activity and its spatiotemporal relation to the world, the world–brain relation, can account for empirical individuation.
How about ontological individuation? Let us take the example of brains and nonbrains. Brains and nonbrains have different spatiotemporal features. Both brains and nonbrains are part of one and the same world. This by itself makes impossible any ontological individuation and distinction of brains and nonbrains unless one wants to characterize them by different elements or properties. However, that is to neglect that both brains and nonbrains are related to one and the same world in different ways: the brain may be related to the world in a much broader spatiotemporal scale or range than nonbrains such as stones—world–brain relation and world–stone relation can thus be distinguished on spatiotemporal grounds. I therefore speak of spatiotemporal individuation.
How can we further characterize such spatiotemporal individuation on ontological grounds? We recall that relation and structure can ontologically be traced to and are based on difference (i.e., difference de re or differentiating de re). Brains and nonbrains such as stones are “differentiating de re” from the world and thereby from each other in different ways on mainly spatiotemporal grounds: the brain is individuated and distinguished from stones by including differences de re of a much larger spatiotemporal scale or range in its relation to the world (i.e., the world–brain relation) when compared to the world–stone relation. Accordingly, true to its name, spatiotemporal individuation is based on the spatiotemporal features of differences de re in the relation of brain and nonbrains to the world (i.e., world–brain relation and world–stone relation).
Ontological individuation in terms of spatiotemporal individuation is quite compatible with the account of individuation in OSR as suggested by Esfeld and Lam (2008, 2010). What Esfeld and Lam (2008, 2010) describe as “web of structure” can well be specified as “spatiotemporal structure,” as it features the world as a whole, including its relation to the parts, such as brains and nonbrains: in the same way the web of structure is determined by spatiotemporal features, the world, including its relation to brains and nonbrains, is signified by spatiotemporal features.
Moreover, Esfeld and Lam’s talk about a position within the web of structure corresponds to what I describe as relation in world–brain relation and world–nonbrain relation: the spatiotemporal scale or range of its relation to the world (i.e., the world–brain relation) situates the brain in a different position on the spatiotemporal trajectories within the world’s spatiotemporal structure when compared to the relation of nonbrains (i.e., stones). Accordingly, brains and nonbrains are distinguished in their existence and reality in an indirect way, namely, by means of their different relation to the world (i.e., world–brain relation and world–nonbrain relation). The relation to the world thus allows for ontological individuation.
In sum, individuation, including both empirical and ontological individuation, is not tied to the supposition of elements and element-based ontology. Structure and relation can well account for both empirical and ontological individuation—since such individuation occurs on spatiotemporal grounds, I speak of spatiotemporal individuation. Spatiotemporal individuation allows us to individuate the brain in both empirical and ontological regards. We can therefore reject the argument of individuation as it is based on the presupposition that individuation is possible only on the basis of elements or properties.
Most importantly, spatiotemporal individuation on both empirical and ontological levels is quite compatible with empirical evidence (see Parts I and II of this book). For that reason, I favor spatiotemporal individuation over individuation by elements or properties and therefore reject the argument of individuation as argument against the ontological determination of the brain in terms of OSR.
Argument of Time and Space Ia: Determination of Time and Space
How can we more clearly distinguish relation and OSR as relation-based ontology from element-based ontology? Proponents of OSR such as Esfeld and Lam (2008, 2010) make a strong case for characterizing the concept of relation in terms of spatial and temporal features—OSR consequently amounts to what I describe as spatiotemporal ontology (see also Northoff, 2016b). The concept of spatiotemporal ontology entails that time and space themselves are the basic units of existence and reality—this, as I will argue, is quite compatible with the focus on structure and relation in OSR.
Esfeld and Lam (2008) mainly draw on physics, for example, GR and quantum physics in particular, when featuring OSR in spatiotemporal terms. This is, for instance, reflected in the following quote:
Moreover, the space-time structure described by GR is such that the space-time relations and the objects that stand in the relations (the space-time points or events) are on the same (fundamental) ontological footing. On the one hand and in an analogous way to the general case discussed in the first section, it makes no sense to consider an actual (that is, instantiated in the physical world) space-time relation without relata standing in the relation—space-time points or events in the pure gravitational cases. (Esfeld & Lam, 2010, p. 22; emphasis added)
My overall argument is that, analogous to GR in physics, the brain must also be characterized by space-time relations rather than space-time points or events. “Space-time relation” specifies and lends further support to relation-based ontology of brain in terms of OSR. To understand the notion of the space-time relation, we need to distinguish between different concepts of time and space, namely, relational time and space and observational time and space.
The proponent of element-based ontology may want to put forward the following argument against OSR. Relation and structure are temporal and spatial and must therefore ultimately be traced to and are based on single discrete points in time and space (i.e., space-time points or events). This opens the door for supposing elements as basic units of existence and reality (i.e., element-based ontology): elements such as physical or mental properties are determined by discrete points in time and space (i.e., space-time points or events) rather than space-time relation. We consequently need to reject OSR as relation-based ontology while, at the same time, embracing element-based ontology. Because the notion of time and space is central in this line of reasoning, I describe this argument as the argument of time and space.
The rejection of OSR may be further aggravated by considering that the notions of time and space are rather trivial. The brain and its neural activity are spatial and temporal by default; this is empirically reflected and manifest in functional connectivity and the fluctuations in different frequencies. That makes any ontological characterization of the brain in terms of time and space superfluous at best and trivial at worst (see chapter 7 for the argument of triviality in the context of consciousness). The argument of time and space can thus also be understood as an argument against the trivial characterization of the brain by time and space in an ontological (rather than empirical; see chapter 7) sense.
Argument of Time and Space Ib: Observational Time and Space
The argument of time and space is primarily an argument against relation-based ontology. However, I will reject that argument by disputing its presupposition. In a nutshell, my rejection is as follows. Structure and relation as presupposed in relation-based ontology such as OSR are indeed temporal and spatial. However, structure and relation in an ontological sense cannot be traced to and are not based on single discrete points in time and space (i.e., space-time points or events) but presuppose a different notion of time and space as defined by space-time relation. Space-time points or events reflect what I will describe as observational time and space while space-time relation presupposes relational time and space.
Accordingly, to reject the argument of time and space, we need to describe and distinguish the notions of observational and relational time and space. That shall be the focus in the following. It should be noted that I cannot go into full detail about the metaphysics of time and space (see, e.g., Dainton, 2010), which would deserve a book by itself. Instead, I only focus on time and space as they are relevant in the current context of the brain. Let us first start with observational time and space.
We observe the brain and its relation to the world. For instance, we observe the brain and its neural activity when applying specific tasks or stimuli to probe the brain’s stimulus-induced or task-evoked activity at discrete and single points in time and space. The brain and its neural activity, for example, stimulus-induced activity, is then framed and put within the time and space of the observer and her or his discrete points in time and space she or he presupposes when applying specific stimuli or tasks. The time and space attributed to the brain and its neural activity are thus based on the time and space of the observer—time and space in this sense are dependent upon the observer. I therefore speak of observational time and space, which, epistemically, must be characterized as mind dependent or brain dependent.
Observational time and space can be characterized by space-time points or events. We observe the brain and its neural activity in terms of “here” and “now” and thus in terms of specific discrete points in time and space. For instance, neural activity is located in a certain region, that is, “here,” at a specific point in time, that is, “now,” which distinguishes it from neural activity in other regions, that is, another “here,” and other points in time, that is, another “now.” In contrast, we remain unable to directly observe the relationship between the different points in time and space. We cannot link and relate the different “now” points with each other nor the various “here” points in our observation. Observational time and space thus presuppose space-time points or events rather than space-time relation.
Argument of Time and Space Ic: Relational Time and Space
How about time and space as they remain independent of our observation? I argue that this leads us to a different notion of time and space, relational time and space as based on space-time relation rather than space-time points or events. Such relational time and space characterizes the brain itself, including its relation to the world, the world-brain relation, independent of our observation of brain and world (in terms of observational time and space with space-time points or events). Therefore, space-time relation with relational time and space must be conceived epistemically as mind independent or brain independent. Historically, the concept of relational time and space as advocated here stands in close relationship to specifically dynamic concepts of time as have been suggested by Leibniz and Clarke (2000), Whitehead (1929/1978), Bergson (1904), and more recently Dainton (2010) when claiming for presentism. Future investigation is required for detailed comparison of my structural-realist approach to time in terms of relational time with the ones discussed in the metaphysics of time in philosophy.
How can we characterize the time and space of the brain itself, including its relation to the world, the world–brain relation? I demonstrated in Part I (chapters 1–3) that the brain’s spontaneous activity shows spatiotemporal structure—this amounts to what Esfeld and Lam describe ontologically as space-time relation. I now argue that the concept of space-time relation can ontologically be specified by those of duration and extension. Let us start with duration.
The concept of duration refers to the time the brain and its neural activity construct by themselves, the brain’s inner time. Empirically, duration is related to the time it takes for neural activity, and thus a neural event, to occur by itself independent of any external stimuli, including their “here” and “now”—that is manifest in the brain’s spontaneous activity and its various frequency ranges that define the brain’s inner duration (see chapters 1–3). This is well reflected in the following quote by D. Griffin (1998) in his description of process philosophy: “Having an inside would mean that they [objects or events such as the brain] can have an inner duration, which is the time it takes each event to occur—the time between its reception of information and its transmission of this information into subsequent events” (p. 144).
Taken in an ontological sense, duration determines the existence and reality of the brain in terms of its relation to the world. The world–brain relation can then ontologically be characterized by time in the sense of duration, that is, inner duration, that describes its extension in time across different points in time and distinguishes from other durations, that is, outer durations, in the rest of the world.
Analogous to duration, I speak of extension to characterize the spatial features of the brain. The brain and its neural activity show a certain spatial extension as is empirically manifest in, for instance, its functional connectivity (chapters 1–3). Functional connectivity makes it possible that the single region’s neural activity is extended to others, which constitutes an inner extension that is specific to the brain and remains independent of the observer. Analogous to duration, I therefore suggest using the term extension in an ontological sense to characterize the existence and reality of brain in terms of space, that is, its inner extension. The world–brain relation can then be characterized by a certain spatial extension, an inner extension, that distinguishes it from the outer extension in the rest of the world.
How can we illustrate the concepts of inner duration and extension in further detail? Inner duration and extension are not defined by single discrete points in time and space (i.e., “here” and “now”). Instead, as based on empirical evidence, inner duration and extension of the brain’s spontaneous activity are based on the relation between different regions, that is, functional connectivity, and frequencies, that is, CFC with scale-free activity (chapters 1–3, 5, and 6). Both inner duration and extension can therefore be signified by space-time relation rather than space-time points or events. Moreover, as empirical evidence shows, the brain aligns to and integrates itself within the world on the basis of its own inner duration and extension when constituting space-time relation with the world (i.e., world–brain relation; chapters 4 and 8).
Argument of Time and Space IIa: Spatiotemporal Spectrum Model
One may now want to raise the question of the relationship between relational and observational time and space with regard to the brain. Relational time and space can be characterized by space-time relation while observational time and space are characterized by space-time points or events. Are both mutually incompatible and exclusive, or are they compatible with each other? This depends on the version of OSR one presupposes.
If one presupposes eliminativist OSR (see above), they are not compatible with each other since then space-time points or events do not exist at all, not even by themselves (not even as abstractions from the more concrete space-time relation). If, in contrast, one presupposes a moderate or noneliminativist stance in OSR, space-time points or events can exist but only in dependence on space-time relation. This is what Esfeld and Lam (2010) suggest:
Space-time points do not possess any independent existence (they are not atoms in the philosophical sense), but only exist in virtue of their standing in relation to other space-time points. There is no ontological priority, but rather a mutual ontological dependence between space-time relations and space-time points. (p. 22)
Presupposing moderate or noneliminativist OSR, observational time and space are a specific instance of relational time and space. That is central in explaining the paradox that observational time and space is based on the observer and her or his brain that by itself can be characterized by relational time and space. How can observational time and space be based on something, that is, the brain with its world–brain relation, that by itself shows a different notion of time and space, that is, relational time and space?
First and foremost, I suggest that, put into an empirical context, observational and relational time and space are related to different forms of neural activity, that is, spontaneous and stimulus-induced activity. The brain’s relational time and space is manifest in its spontaneous activity and its relation to the world (i.e., world–brain relation) with its space-time relation as described by inner duration and extension—while the observer’s observational time and space is based on her or his brain’s stimulus-induced or task-evoked activity and its various perceptual and cognitive functions that allow for observing space-time points or events.
We demonstrated that there is no sharp and clear-cut empirical distinction between spontaneous and stimulus-induced activity—this amounted to the spectrum model on the empirical level of the brain’s neural activity (chapter 1). Analogously, I now suppose that there is no sharp and clear-cut distinction between relational and observational time and space and hence between space-time relation and space-time points or events. This amounts to an analogous spectrum model on the ontological level of time and space, a spatiotemporal spectrum model.
How can I determine the spatiotemporal spectrum model? What we describe as spatiotemporal points or events in observational time and space may ontologically be an extreme instance of an extremely short extension and duration of space-time relation in relational time and space. That is, rather than being different in principle and mutually exclusive, relational and observational time and space can be characterized by a continuum or spectrum of different spatiotemporal scales or ranges: relational time and space entail a longer spatiotemporal scale or range which configures as space-time relation, while the one in observational time and space is extremely short and surfaces as space-time points or events. Relational and observational time and space can consequently be “positioned” or “located” on different ends of a commonly shared spatiotemporal spectrum.
Note that the concept of spectrum is here understood in a truly ontological sense concerning the spectrum of different notions of time and space within the world. That must be distinguished from the more empirical use of the notion of spectrum as in the “spectrum model of brain” as developed in the first chapter. Here time and space are limited to the brain as well as to our observation of time and space. That is different in the ontological notion where time and space are considered within the world rather than the brain as well as independent of observation.
Argument of Time and Space IIb: Rejection of the Argument of Time and Space
The spatiotemporal spectrum model also characterizes time and space of the brain. From the world over the brain’s spontaneous activity to the brain’s stimulus-induced activity, a spectrum of different temporal and spatial scales or ranges is recruited and implicated. When the brain and its spontaneous activity align themselves to and integrate within the world (i.e., world–brain relation), a large scale or range of time and space is involved—this is manifest in space-time relation and relational time and space.
If, in contrast, the brain’s stimulus-induced or task-induced activity is recruited during observation, the spatiotemporal scale or range becomes smaller and shifts toward space-time points or events as in observational time and space. Observational time and space with its space-time points or events may thus be conceived as an abstraction from the spatiotemporally more extended relational time and space.
However, stimulus-induced or task-evoked activity is dependent upon spontaneous activity (see chapter 2), which, in turn, is dependent upon its relation to the world, the world–brain relation. This also means that space-time points or events as in observational time and space are dependent and based upon space-time relation as in relational time and space.
This is obviously quite compatible with Esfeld and Lam’s noneliminativist version of OSR in which space-time points or events do not exist independently of space-time relation. Their stance can now be extended and complemented by the spatiotemporal spectrum model that, as an ontological model of time and space, suggests a continuum between different spatiotemporal extensions of different notions of time and space such as relational and observational time and space. The space-time points or events are then a continuum of space-time relations on an extremely small spatiotemporal scale—the former are thus an abstraction of the latter.
We are now ready to address the argument of time and space. The argument of time and space is based on the presupposition that relation and structure are spatiotemporal and must therefore be based on space-time points or events (see above). However, this presupposition is rendered wrong by the spatiotemporal spectrum model. The spatiotemporal spectrum model argues that space-time points or events are based and dependent upon space-time relation rather than the latter being dependent upon the former. Such spatiotemporal spectrum model is not only ontologically plausible but also empirically supported given that it rests on the spectrum model of brain (chapter 1). I am thus able to reject the argument of time and space as an argument against moderate OSR of the brain in terms of world–brain relation and space-time relation on both ontological and empirical grounds.
Moreover, the ontological characterization of the brain by time and space is not trivial at all. One can characterize the brain ontologically by observational time and space; in that case, there would be no distinction between empirical and ontological determination of the brain. Alternatively, one can ontologically describe the brain by relational time and space as distinguished from observational time and space; in that case, one would need to distinguish between ontological and empirical characterization of the brain.
We can then also address the argument of triviality. We already rejected the empirical version of that argument in chapter 7 by hinting at different spatiotemporal mechanisms (chapter 7). This is complemented now by rejecting its ontological version, namely, that the ontological characterization of world and brain including world–brain relation by time and space is trivial. Because we are confronted with two alternative ontological options, that is, relational versus observational time and space, the ontological determination of the brain by time and space cannot be considered trivial at all.
Most important, this carries major ontological implications. In the case of observational time and space, the brain’s existence and reality are determined independent of those of the world, entailing element-based ontology. In contrast, that is no longer the case with regard to relational time and space where the brain’s spatiotemporal determination entails world–brain relation, which presupposes relation- rather than element-based ontology. We will see in the next chapter that both world–brain relation and relation-based ontology are central for the ontological determination of consciousness.
Conclusion
I characterize the brain ontologically by relation and structure. This amounts to SR and, more specifically, moderate OSR of the brain. Moderate OSR determines the brain’s existence and reality by relation and thus world–brain relation. The brain is its relation to the world; in short, the brain is world–brain relation. Without its relation to the world, the brain does not exist. Such relational view of the brain must be distinguished from its ontological definition by elements such as mental or physical properties as in element-based ontology. The discussion of the ontological determination of brain can be regarded as a first step toward a “philosophy of brain” (Northoff, 2004).
How can the relation in the concept of world–brain relation be defined in more ontological detail? Following OSR, the relation can be defined in temporal and spatial terms, that is, by space-time relation rather than space-time points or events. This led me to distinguish relational time and space with duration and extension from observational time and space, which I then applied to the brain and world–brain relation. The world–brain relation remains independent of our observation and thus observational time and space. Instead, the world–brain relation can be characterized by relational time and space that determines the brain’s existence and reality. Most important, based on empirical evidence (see chapters 7 and 8), I propose that the world–brain relation, including its relational time and space, is a predisposition of consciousness. This shall be the focus in the next chapter.
