This paper introduces a framework for understanding what distinguishes humans from artificial communication at the level of individual human interaction. Every human act carries what the paper terms conscience tokens (weighted moral exchanges tracked subconsciously in an internal ledger). AI generates only “plain tokens” (unweighted information transfer). This categorical difference has profound implications for AI deployment in roles requiring moral accountability, particularly judicial authority. The framework emerges from a simple observation: when humans interact, we do not merely exchange information. We (...) exchange moral weight across all senses (speech, gesture, glance, touch, and subtler cues). Different tokens carry different moral weights, and everyone has a different measure. Your formed conscience weights these instinctively. Ledger balance brings peace, imbalance, and moral debt. And more besides, beyond full articulation or conscious grasp. (shrink)

The classical rule of Repetition says that if you take any sentence as a premise, and repeat it as a conclusion, you have a valid argument. It's a very basic rule of logic, and many other rules depend on the guarantee that repeating a sentence, or really, any expression, guarantees sameness of referent, or semantic value. However, Repetition fails for token-reflexive expressions. In this paper, I offer three ways that one might replace Repetition, and still keep an interesting notion (...) of validity. Each is a fine way to go for certain purposes, but I argue that one in particular is to be preferred by the semanticist who thinks that there are token-reflexive expressions in natural languages. (shrink)

Token physicalism is often viewed as a modest and unproblematic physicalist commitment, as contrasted with type physicalism. This paper argues that the prevalence of functional individuation in biology creates serious problems for token physicalism, because the latter requires that biological entities can be individuated physically and without reference to biological functioning. After characterizing the main philosophical roles for token physicalism, I describe the distinctive uses of functional individuation in models of biological processes. I then introduce some requirements (...) on token identity claims that arise from a position on individuation and identity known as sortalism. An examination of biological examples shows that these sortalist requirements cannot be plausibly met due to differences between individuation by functional biological criteria and by physical criteria. Even without assuming sortalism, token physicalism faces the more basic problem of excluding functionally irrelevant detail from the individuation of biological tokens. I close by suggesting that the philosophical roles for token identity are better fulfilled by a notion of token composition, which promotes a hierarchical picture of individuality. (shrink)

This paper introduces a framework for understanding what distinguishes humans from artificial communication at the level of individual human interaction. Every human act carries what the paper terms conscience tokens (weighted moral exchanges tracked subconsciously in an internal ledger). AI generates only “plain tokens” (unweighted information transfer). This categorical difference has profound implications for AI deployment in roles requiring moral accountability, particularly judicial authority. The framework emerges from a simple observation: when humans interact, we do not merely exchange information. We (...) exchange moral weight across all senses (speech, gesture, glance, touch, and subtler cues). Different tokens carry different moral weights, and everyone has a different measure. Your formed conscience weights these instinctively. Ledger balance brings peace, imbalance, and moral debt. And more besides, beyond full articulation or conscious grasp. (shrink)

The entry provides an overview of the type-token distinction, including a comparison to other nearby distinctions.

In this book, Linda Wetzel examines the distinction between types and tokens and argues that types exist (as abstract objects, since they lack a unique ...

Token-reflexive theories of indexicals – words like ‘I,’ ‘here,’ and ‘today’ – are widely thought to face a problem in account for intuitively valid arguments involving indexicals. Yet all discussions of the problem with which I am familiar focus on particular examples or on particular rules of inference. In this paper, I first state the problem in its full generality, and then argue that two recent attempts to reject the problem fail. Finally, I consider the proposal by García-Carpintero that (...) demonstratives – ‘this’ and ‘that’ – raise the same problem. I argue that while there are important similarities between the logic problem for token-reflexives and the role of demonstratives in logic, recent work on the logic of demonstratives can offer insights into logic that token-reflexive theories cannot. (shrink)

There are many grounds to object to tokenism, but that doesn’t mean we should always avoid being the token woman, argues Anca Gheaus.

In this article, I argue that Kant’s Categorical Imperative applies to maxim tokens rather than to maxim types. The article has three main parts. In the first, I explain my thesis. In the second, I argue for it. In the third, I argue, further, that, if my thesis is correct, then tokens of different maxim types can have different deontic statuses for different agents.

Reichenbachian approaches to indexicality contend that indexicals are 'token-reflexives': semantic rules associated with any given indexical-type determine the truth-conditional import of properly produced tokens of that type relative to certain relational properties of those tokens. Such a view may be understood as sharing the main tenets of Kaplan's well-known theory regarding content, or truth-conditions, but differs from it regarding the bearers of truth-conditional import and truth-conditions. Kaplan has criticized these approaches on different counts, the most damaging of which is (...) that they make impossible a 'logic of demonstratives'. The reason for this is that the token-reflective approach entails that not two tokens of the same sentential type including indexicals are guaranteed to have the same truth-conditions. In this paper I rebut this and other criticisms of the Reichenbachian approach. Additionally, I point out that Kaplan's original theory of 'true demonstratives' is empirically inadequate, and claim that any modification capable of accurately handling the linguistic data would have similar problems to those attributed to the Reichenbachian approach. This is intended to show that the difficulties, no matter how real, are not caused by idiosyncrasies of the 'token-reflexive' view, but by deep facts about indexicality. (shrink)

Token-reflexivity is commonly understood as reference of a token to a token of which it is a part, proper or not. It may be compared with its familiar formal kin – Gödelian reflexivity. In this paper the possibility of the latter type of construction in a formal setting provides a stark point of contrast with token-reflexivity understood as token self-reference, a purported species of natural phenomena, with the token-reflexives themselves understood as the bearers of (...) self-reference. I argue that there is no token-reflexivity thus understood, and so, no token-reflexives. The case provides a particularly useful background against which to discuss the centrality of conditions of production – as opposed to conditions of consumption – in the study of natural language. (shrink)

The distinction between a type and its tokens is a useful metaphysical distinction. In §1 it is explained what it is, and what it is not. Its importance and wide applicability in linguistics, philosophy, science and everyday life are briefly surveyed in §2. Whether types are universals is discussed in §3. §4 discusses some other suggestions for what types are, both generally and specifically. Is a type the sets of its tokens? What exactly is a word, a symphony, a species? (...) §5 asks what a token is. §6 considers the relation between types and their tokens. Do the type and all its tokens share the same properties? Must all the tokens be alike in some or all respects? §7 explains some problems for the view that types exist, and some problems for the view that they don't. §8 elucidates a distinction often confused with the type-token distinction, that between a type (or token) and an occurrence of it. It also discusses some problems that occurrences might be thought to give rise to, and one way to resolve them. (shrink)

We present a probabilistic extension to active path analyses of token causation. The extension uses the generalized notion of intervention presented in : we allow an intervention to set any probability distribution over the intervention variables, not just a single value. The resulting account can handle a wide range of examples. We do not claim the account is complete --- only that it fills an obvious gap in previous active-path approaches. It still succumbs to recent counterexamples by Hiddleston, because (...) it does not explicitly consider causal processes. We claim three benefits: a detailed comparison of three active-path approaches, a probabilistic extension for each, and an algorithmic formulation. (shrink)

There are two extant versions of the new tenseless theory of time: the date versionand the token-reflexive version. I ask whether they are equivalent, and if not, whichof them is to be preferred. I argue that they are not equivalent, that the date version isunsatisfactory, and that the token-reflexive version is correct. I defend the token-reflexive version against a string of objections from Quentin Smith. My defence involves a discussion of the ontological and semantic significance of truth (...) conditions, and of the connection between truth and reality on the one hand, and that between truth and meaning on the other. I argue that Smith's objections to the token-reflexive theory stem from his confusing these two aspects of the notion of truth. (shrink)

The purpose of this article is to assess Rosabeth Moss Kanter's work on tokenism in light of more than a decade of research and discussion. While Kanter argued that performance pressures, social isolation, and role encapsulation were the consequences of disproportionate numbers of women and men in a workplace, a review of empirical data concludes that these outcomes occur only for token women in gender-inappropriate occupations. Furthermore, Kanter's emphasis on number balancing as a social-change strategy failed to anticipate backlash (...) from dominants. Blalock's theory of intrusiveness suggests that surges in the number of lower-status members threaten dominants, thereby increasing gender discrimination in the forms of sexual harassment, wage inequities, and limited opportunities for promotion. Although Kanter's analysis of the individual consequences of tokenism was compelling to researchers and organizational change agents, continued reliance on numbers as the theoretical cause of, and as the solution to, gender discrimination in the workplace neglects the complexities of gender integration. (shrink)

It has recently been suggested that the type/token theorist concerning musical works cannot come up with an adequate semantic theory of those sentences in which we purport to talk about such works. Specifically, it has been claimed that, since types are abstract entities, a type/token theorist can only account for the truth of sentences such as “The 1812 Overture is very loud” and “Bach's Two Part Invention in C has an F-sharp in its fourth measure” by adopting an (...) untenable semantic claim: namely, that the predicates in such sentences, once applied to musical works, undergo a systematic shift in their meanings. This article is a sustained explanation of why our talk about musical works in fact provides no problem for the type/token theorist. First, we demonstrate that the aforementioned “meaning shift” approach to the sentences’ predicates is well motivated and very credible. Second, we explain how the type/token theorist can adopt the best available version of an alternative, generic quantificational approach to such sentences. Third, we establish that other semantic theories, presented as undermining the type/token theory by giving us a reason for adopting eliminativism about types, are much less theoretically virtuous than the two theories that a type/token theorist can freely adopt. (shrink)

A robust stakeholder analysis requires extensive conceptual and empirical work. Yet it is often unclear how to effectively do so. This paper introduces a new method—the Stakeholder Tokens—for designers to elicit a more inclusive set of stakeholders and gain better understanding of stakeholder interrelationships and dynamics. Stakeholder Tokens present a playful hands-on design approach to support value sensitive design stakeholder analyses by employing a style of role play.

Miranda Fricker maintains that testimonial injustice is a matter of credibility deficit, not excess. In this article, I argue that this restricted characterization of testimonial injustice is too narrow. I introduce a type of identity-prejudicial credibility excess that harms its targets qua knowers and transmitters of knowledge. I show how positive stereotyping and prejudicially inflated credibility assessments contribute to the continued epistemic oppression of marginalized knowers. In particular, I examine harms such as typecasting, compulsory representation, and epistemic exploitation and consider (...) what hearers are obligated to do in response to these injustices. I argue that because epistemic harms to marginalized knowers also arise from prejudicially inflated assessments of their credibility, the virtue of testimonial justice must be revised to remedy them. (shrink)

John Searle has argued that all perceptual experiences are token-reflexive, in the sense that they are constituents of their own veridicality conditions. Many philosophers have found the kind of token-reflexivity he attributes to experiences, which I will call _causal_ token-reflexivity, unfaithful to perceptual phenomenology. In this paper, I develop an argument for a different sort of token-reflexivity in perceptual (as well as some non- perceptual) experiences, which I will call _temporal_ token-reflexivity, and which ought to (...) be phenomenologically unobjectionable. (shrink)

The purpose of this paper is to uncover and correct several confusions about expressions, tokens and the relations between them that crop up in even highly sophisticated writing about language and logic.

According to the Multiple Realizability Argument, a higher-level property typically has many physical realizers, so it cannot be type-identical to any one of them. This enables the non-reductive physicalist to claim that some higher-level properties are type-distinct from physical properties. The reductive physicalist can counter with the Disjunctive Strategy: nothing prevents us from type-identifying the higher-level property with the disjunction of its realizers. Developing a powers-based ontology of properties, Shoemaker and Wilson present responses to the Disjunctive Strategy, wherein higher-level property (...) instances are token-distinct from their realizers, while instances of disjunctions are not. In this paper, I argue that such responses to the Disjunctive Strategy can be reasonably resisted, either by denying token-distinctness, or by insisting that exhaustively overlapping disjunctive properties are also token-distinct from the realizers. This secures the Disjunctive Strategy as a core component of the argument for reductive physicalism. (shrink)

(2007). Racial Tokenism in the School Workplace: An Exploratory Study of Black Teachers in Overwhelmingly White Schools. Educational Studies: Vol. 41, No. 3, pp. 230-254.

Causal Bayes nets (CBNs) provide one of the most powerful tools for modelling coarse-grained type-level causal structure. As in other fields (e.g., thermodynamics) the question arises how such coarse-grained characterisations are related to the characterisation of their underlying structure (in this case: token-level causal relations). Answering this question meets what is called a “coherence-requirement” in the reduction debate: How are different accounts of one and the same system (or kind of system) related to each other. We argue that CBNs (...) as tools for type-level causal inference are abstract enough to roughly fit any current token-level theory of causation as long as certain modelling assumptions are satisfied, but accounts of actual causation, i.e. accounts that attempt to infer token-causation based on CBNs, for the very same reason, face certain limitations. (shrink)

Slurs are derogatory, and theories of slurs aim at explaining their “derogatory force”. This paper draws a distinction between the type derogatory force and the token derogatory force of slurs. To explain the type derogatory force is to explain why a slur is a derogatory word. By contrast, to explain the token derogatory force is to explain why an utterance of a slur is derogatory. This distinction will be defended by examples in which the type and the (...) class='Hi'>token derogatory force come apart. Because of the distinction, an adequate theory of slurs must be plausible for both the type and the token derogatory force. However, I will argue that many theories fail to be plausible for both. In particular, Hom’s combinatorial externalism and the conventional implicature theory offer implausible accounts of the token derogatory force, whereas the prohibitionist theory is insufficient to explain the type derogatory force. (shrink)

Clearing algorithms are at the core of modern payment systems, facilitating the settling of multilateral credit messages with (near) minimum transfers of currency. Traditional clearing procedures use batch processing based on MILP - mixed-integer linear programming algorithms. The MILP approach demands intensive computational resources; moreover, it is also vulnerable to operational risks generated by possible defaults during the inter-batch period. This paper presents TORC3 - the Token-Ring Clearing Algorithm for Currency Circulation. In contrast to the MILP approach, TORC3 is (...) a real time heuristic procedure, demanding modest computational resources, and able to completely shield the clearing operation against the participating agents’ risk of default. (shrink)

The distinction between token and type physicalism is a familiar feature of discussion of psychophysical relations. Token physicalism, or ontological physicalism, is the view that every token, or particular, in the spatiotemporal world is a physical particular. It is contrasted with type physicalism, or property physicalism -- the view that every first-order type, or property, instantiated in the spatiotemporal world is a physical property. Token physicalism is commonly viewed as a clear thesis, strictly weaker than property (...) physicalism, strictly stronger than substance physicalism, and as a good statement on its own or in conjunction with other theses of minimal physicalism.[i] It is also generally simply assumed to be true, though Davidson has offered a famous argument for its truth, and some have argued against it. Many of those arguing against it are substance physicalists, indicating that they believe token physicalism to be a strictly stronger view.[ii]. (shrink)

Synonymy, at its most basic, is sameness of meaning. A token-reflexive expression is an expression whose meaning assigns a referent to its tokens by relating each particular token of that particular expression to its referent. In doing so, the formulation of its meaning mentions the particular expression whose meaning it is. This seems to entail that no two token-reflexive expressions are synonymous, which would constitute a strong objection against token-reflexive semantics. In this paper, I propose and (...) defend a notion of synonymy for token-reflexive expressions that allows such expressions to be synonymous, while being a fairly conservative extension of the customary notion of synonymy. (shrink)

This paper disputes a common definition of token identity theory. It also observes that within the philosophical literature there are two significantly different definitions of token identity theory that are commonly used.

According to a Reichenbachian treatment, indexicals are token-reflexive. That is, a truth-conditional contribution is assigned to tokens relative to relational properties which they instantiate. By thinking of the relevant expressions occurring in “ordinary contexts” along these lines, I argue that we can give a more accurate account of their semantic behavior when they occur in indirect contexts. The argument involves the following: (1) A defense of theories of indirect discourse which allows that a reference to modes of presentation associated (...) with expressions occurring in indirect contexts can be made depending on contextual aspects. (2) A defense of the “doubleindexing” theories proposed by Stalnaker and others in order to account for the difference between metaphysical and epistemological modalities. (3) The claim that a Reichenbachian view improves upon the theories defended in (1) and (2). (shrink)

Contextualist scholars working on the rhetoric of corporeal presence in seventeenth-century English religious lyrics have naturally focused their attention on sacramental discourse of the Reformation era. As part of the Common Knowledge symposium on the future of contextualism, this full-length monograph, serialized in installments, argues that the contextualist focus on a single and time-limited “epistemic field” has resulted in a less than adequately ramified understanding of the poetry of John Donne, George Herbert, Aemilia Lanyer, and John Milton. What the contextualist (...) approach misses is that even the religious discourses of the period were tied to a long and in no way local epistemological debate about signs and their meaning, whose roots are to be found in Greek and Latin rhetorical theory. This first installment of “Tokens of Love” commences a discussion of the role of classical pagan sign-theory in the development of Reformation sacramental discourse. (shrink)

Some philosophers have argued that so long as two neural events, within a subject, are both of the same type and both carry the same content, then these events may jointly constitute a single mental token, regardless of the sort of causal relation to each other that they bear. These philosophers have used this claim—which I call the “singularity-through-redundancy” position—in order to argue that a split-brain subject normally has a single stream of consciousness, disjunctively realized across the two hemispheres. (...) This paper argues, against this position, that the kind of causal relations multiple neural events bear to each other constrains the mental tokens with which functionalists who are realists can identify them. (shrink)

Concerns about the under-representation of female academic philosophers and about the stereotype that philosophy is best done by men have recently led to efforts to make academic philosophy a more inclusive discipline. An example is the Gendered Conference Campaign, encouraging event organisers and volume editors to include women amongst invited speakers and authors. Initiatives such as the GCC raise worries about tokenism. Potential invitees may be concerned about unfairness towards whose who would have been invited in their place in the (...) absence of affirmative action and about the way in which affirmative action can affect the quality of the conference or volume in question. And women philosophers often worry that, if formal rules or significant social pressures towards gender inclusiveness play a role in selection processes, their achievements will be discounted. I argue there is no good reason for these fears: there is no pure meritocracy in academia, nor is the ideal of pure meritocracy either feasible or desirable. There are several legitimate grounds — independent of professional competence — for including people in positions of visibility and prestige; gender is such a legitimate reason. (shrink)

The argument from causal overdetermination is considered to be the shortest route to token monism. It only assumes that:1.Efficacy: Mental events are causes of physical events.2.Closure: Every physical event has a sufficient physical cause.3.Exclusion: Systematic Causal Overdetermination is impossible: if an event x is a sufficient cause of an event y then no event x* distinct from x is a cause of y.4.Identity: Therefore, mental events are physical events.Exclusion does not deny the possibility of two gunmen that fi re (...) at a victim at the same time. But event-dualism is like a systematic fi ringsquad case — whenever I want to raise my arm, my arm is raised, and that is intolerable. (shrink)

In his (1977) "Anomalous Monism and Kripke's Cartesian Intuitions," Colin McGinn argues that Donald Davidson's anomalous monism is untouched by Kripke's (1980) argument against the identity theory. The type-identity of the physical with the mental may very well fall at the feet of Kripke's powerful arguments, but a token identification, argues McGinn, is left standing due to the simple fact that token physicalism countenances a kind of imagined separation of token mental states with their corresponding token (...) physical states. If McGinn is correct, a full-blooded physicalism is consistent with Kripke's Cartesian intuitions regarding the non-identity of the mental and the physical. But I think McGinn is mistaken. In particular, McGinn misunderstands the nature of an "epistemic counterpart" of a token pain. So contrary to McGinn, token physicalism does not seem to be able to defend against Kripke after all. (shrink)

The two-fold ontological character of linguistic objects revealed due to the distinction between “type” and “token” introduced by Ch. S. Peirce can be a base of the two-fold, both theoretical and axiomatic, approach to the language. Referring to some ideas included in A. A. Markov’s work [1954] (in Russian) on Theory of Algorithms and in some earlier papers of the author, the problem of formalization of the concrete and abstract words theories raised by J. Słupecki was solved. The construction (...) of the theories presented here has two levels. The axiomatic theory of label-tokens: material, physical linguistic objects, constitutes the first one. Label-types, according to the literature of the subject, are defined on the other level as equivalence classes of equiform label-tokens. Assuming the opposite point of view, one can accept that theory of label-types: abstract labels, formalized on the first level, in which it is possible to define the notion of label-token as well as the derivative notions on the second level, should become the basis of formalization of the theory of linguistic expressions and the theory of language in general. The axioms and definitions of both theories of labels: T k and T p representing the other approach to the ontology of language are included in the sequel of the abstract. The foundations of the theory of labels T k in which the primary assumption as to the label-types existence is superfluous have been referred on the basis of the author's monography "Teorie Języków Syntaktycznie Kategorialnych" ( "The Theories of Syntactically Categorial Languages"), PWN, Warszawa-Wrocław 1985. The basis of the theory of labels T p which takes into account the other position has to be presented here for the first time. Some extended ideas of the paper will also be presented in author's paper "Logiczne podstawy ontologii składni Języka" ("Logical foundations of language syntax ontology), Studia Filozoficzne 6-7 (271-272), (1988), pp. 263-284. (shrink)

This paper proposes that the relation between property instances and token causal powers is akin to the relation between primary substances and property instances on the Aristotelian account of property instantiation. This view permits an individual to have two tokens of the same type of causal power. Paul Audi has argued that this cannot be: two tokens of the same power type are discernible, he claims, only if they are borne by discernible individuals. In the context of this criticism, (...) he proposes that power tokens are individuated in part by the individuals bearing them. This paper responds to Audi’s criticism and argues against his preferred view. This paper thus establishes a theory of token causal powers where previously the literature contained only a presumed view and Audi’s objections. (shrink)

The subject of this paper is the problem of knowledge revealed by the emergence of advanced natural language processing technology, called large language models (LLMs). It proposes an interpretation of the appearance of this phenomenon as an emergent effect of a complex system, such as LLMs. The meaning of this interpretation is based on the observation of the computational trajectory of computational units (vectors) which, on the other hand, represent discrete semantic units, tokens. To justify the autonomy and relevance of (...) this interpretation of knowledge, the paper invokes the concept of discursive space, which allows, in particular, for the reliance on language as a container of knowledge and for the generalization of the idea of knowledge to other semantic environments. To describe more general units of knowledge, the strong version of the theory proposes the introduction of the institution of gnosemes, of which discourse is a special case. Gnosemes find application in the case of LLMs. Due to the different ontical contexts of the entities described, the paper proposes a transdisciplinary approach, the axis of which is the field of social sciences. (shrink)

In his famous argument for the unreality of time, McTaggart claims that i) being past, being present, and being future are incompatible properties of an event, yet ii) every event admits all these three properties. In this paper, I examine two key concepts involved in the formulation of i) and ii), namely that of “validity” and that of “contradiction”, and for each concept I distinguish a static version and a dynamic version of it. I then arrive at three different ways (...) of formulating McTaggart’s claims that avoid the notorious McTaggart’s Paradox. So long as we demand that McTaggart make clear use/mention and token/type distinctions in his claims, we shall find that it is indeed very difficult for him to get a genuine contradiction from i) and ii). (shrink)

Inspired by Castañeda (1966, 1968), Perry (1979) and Lewis (1979) showed a specific variety of singular thoughts, thoughts about oneself “as oneself”—_de se_ thoughts, Lewis called them—that raise special issues, and they advanced rival accounts. Their suggestive examples raise the problem of _de se_ thought—how to characterize it so as to give an accurate account of the data, tracing its relations to singular thoughts in general. The main tenets of the two contrasting accounts—Lewisian and Perrian—are rehearsed in the first section (...) of this chapter, together with some weaknesses of Perry’s presentation of his view. In the second, a proposal is presented, a specific elaboration of the Perrian account which overcomes the weaknesses. The chapter concludes with a brief discussion of reasons for preferring one or another account, in particular regarding the issue of the communication of _de se_ thoughts. (shrink)

The distinction between institution types and institution tokens plays an important role in Francesco Guala’s philosophy of institutions. In this commentary, I argue that this distinction faces a number of difficulties that are not sufficiently addressed in Understanding Institutions. In particular, I critically discuss Guala’s comparison between the taxonomy of organisms and the taxonomy of institutions, consider the semantics of institution terms on different levels in this taxonomy, and argue for an alternative solution to the problem of how to reconcile (...) reformism and realism about institutions like marriage. (shrink)

Propertarian justifications of intellectual property postulate the appropriation of various entities, often called patterns, designs, or technologies. These must be immaterial and should not be confused with material structures that embody them. Hence two classes of objects are distinguished. It is convenient to refer to them as types and tokens. The type must involve a condition defining which material structures should be considered its tokens. For an IP regime to be economically meaningful one must necessarily appropriate types in a way (...) which restricts access to wide classes of similar, non-identical material structures. Therefore, type conditions must be general. They must define, with a margin of tolerance, only certain aspects of the structure, leaving others unspecified. Consequently, the relationship between such types and tokens is of many-to-many variety. The recognition of these facts leads to four problems in justifying intellectual property on propertarian grounds. The first problem is to demonstrate possession of the type in the pre-legal situation. The second one is to explain, why boundaries of an appropriated type should be placed in a particular location. The third problem is to avoid claim deadlocks resulting from conflicts of rights generated by separate overlapping types. The fourth problem is to justify why ownership of the type should entail the control over other separate entities—that is tokens. In the propertarian framework grounded in the pre-legal state of nature, satisfactory solutions do not seem to have been proposed for any of these four problems. (shrink)

This study analyzes tokenization in Natural Language Processing (NLP) at the intersection of language, technology, and political economy. Recent studies show that subword tokenization in multilingual Large Language Models (LLMs) distributes processing costs unevenly across languages, as non-Latin scripts are disproportionately fragmented by statistical segmentation. Taking this disparity as a point of departure, the study reconceptualizes tokenization beyond a purely linguistic or computational problem. The analysis approaches tokenization through the epistemological problems of translation and situates it within a longer history (...) of socio-technical infrastructures from telegraphy to machine learning that mediate between linguistic abstraction and regimes of social productivity. Rather than treating tokens as neutral technical units, it shows that they function as operational measures of linguistic value beyond meaning, enabling the quantification and automation of linguistic labor. Drawing on Marxist and structuralist accounts of linguistic value, and postcolonial critiques of semiotic standardization, the study frames tokenization as both a technical abstraction and a metric in the social division of labor. (shrink)

Judea Pearl (2000) has recently advanced a theory of token causation using his structural equations approach. This paper examines some counterexamples to Pearl's theory, and argues that the theory can be modified in a natural way to overcome them.

The literature on causation distinguishes between causal claims relating properties or types and causal claims relating individuals or tokens. Many authors maintain that corresponding to these two kinds of causal claims are two different kinds of causal relations. Whether to regard causal relations among variables as yet another variety of causation is also controversial. This essay maintains that causal relations obtain among tokens and that type causal claims are generalizations concerning causal relations among these tokens.