This paper critiques a number of standard ways of understanding the role of the metalanguage in a semantic theory for natural language, including the idea that disquotation plays a nontrivial role in any explanatory natural language semantics. It then proposes that the best way to understand the role of a semantic metalanguage involves recognizing that semantics is a model-based science. The metalanguage of semantics is language for articulating features of the theorist's model. Models are understood as mediating instruments---idealized structures used (...) to represent select aspects of the world, aspects the theorist is seeking some theoretical understanding of. The aspect of reality we are seeking some understanding of in semantics is a dimension of human linguistic competence---informally, knowledge of meaning. (shrink)

V. 1. Structures and processes -- v. 2. Domains and categories.

Frege begins Die Grundlagen der Arithmetik, the work that introduces the project which was to occupy him for most of his professional career, with the question, 'What is the number one?' It is a question to which even mathematicians, he says, have no satisfactory answer. And given this scandalous situation, he adds, there is small hope that we shall be able to say what number is. Frege intends to rectify the situation by providing definitions of the number one and the (...) concept of number. But what, exactly, is required of a definition? Surely it will not do to stipulate that the number one is Julius Caesar - that would change the subject. It seems reasonable to suppose that an acceptable definition must be a true statement containing a description that picks out the object to which the numeral '1' already refers. And, similarly, that an acceptable definition of the concept of number must contain a description that picks out precisely those objects that are numbers - those objects to which our numerals refer. Yet, while Frege writes a great deal about what criteria his definitions must satisfy, the above criteria are not among those he mentions. Nor does he attempt to convince us that his definitions of '1' and the other numerals are correct by arguing that these definitions pick out objects to which these numerals have always referred. Yet, while Frege writes a great deal about what criteria his definitions must satisfy, the above criteria are not among those he mentions. Nor does he attempt to convince us that his definitions of ‘1’ and the other numerals are correct by arguing that these definitions pick out objects to which these numerals have always referred. There is, as we shall see shortly, a great deal of evidence that Frege’s definitions are not intended to pick out objects to which our numerals already refer. But if this is so, how can these definitions teach us anything about our science of arithmetic? And what criteria must these definitions satisfy? To answer these questions, we need to understand what it is that Frege thinks we need to learn about the science of arithmetic. (shrink)

Suppose we consider knowledge to be valuable because of the role known propositions play in practical reasoning. This, I argue, does not provide a reason to think that knowledge is valuable in itself. Rather, it provides a reason to think that true belief is valuable from one standpoint, and that justified belief is valuable from another standpoint, and similarly for other epistemic concepts. The value of the concept of knowledge is that it provides an economical way of talking about many (...) epistemic concepts that are valuable in itself from one standpoint or another. In this way knowledge is like a Swiss Army Knife; a Swiss Army Knife is not useful as such on any occasion, but it provides an efficient way of carrying around different tools that are useful in themselves on different occasions. (shrink)

I propose that testimony is subject to a norm that is backed by a credibility sanction: whenever the norm is violated, it is appropriate for the testifier to lose some credibility for their future testimony. This is one of a family of sanction-based norms, where violation of the norm makes it appropriate to lose some power; in this case, the power to induce belief through testimony. The applicability of the credibility norm to testimony follows from the epistemology of testimony, in (...) that false or unjustified testimony weakens the reason for belief that is provided by the speaker's future testimony. (shrink)

Normative ethical theories owe us an account of how to evaluate decisions under risk and uncertainty. Deontologists seem at a disadvantage here: our best decision theories seem tailor-made for consequentialism. For example, decision theory enjoins us to always perform our best option; deontology is more permissive. In this paper, we discuss and defend the idea that, when some pro-tanto wrongful act is all-things considered permissible, because it is a ‘lesser evil’, it is often merely permissible, by the lights of deontology. (...) We show that this raises new problems for deontological decision theory, and we show that to resolve them, we need to take a more innovative approach to morally evaluating decision-making under risk and uncertainty. (shrink)

The House We Live In explores the commonalities underlying three classical approaches to virtue ethics-Aristotelean, Buddhist, and Confucian-to develop a flourishing-based ethics capable of addressing the problems of liberal democracies. This book will appeal to scholars and to general readers.

Epistemic modal operators give rise to something very like, but also very unlike, Moore's paradox. I set out the puzzling phenomena, explain why a standard relational semantics for these operators cannot handle them, and recommend an alternative semantics. A pragmatics appropriate to the semantics is developed and interactions between the semantics, the pragmatics, and the definition of consequence are investigated. The semantics is then extended to probability operators. Some problems and prospects for probabilistic representations of content and context are explored.

In this paper, we suggest that affect meets the traditional definition of “cognition” such that the affect–cognition distinction is phenomenological, rather than ontological. We review how the affect–cognition distinction is not respected in the human brain, and discuss the neural mechanisms by which affect influences sensory processing. As a result of this sensory modulation, affect performs several basic “cognitive” functions. Affect appears to be necessary for normal conscious experience, language fluency, and memory. Finally, we suggest that understanding the differences between (...) affect and cognition will require systematic study of how the phenomenological distinction characterising the two comes about, and why such a distinction is functional. (shrink)

This article considers the implications for film analysis of the presence or absence of a manual crank. More specifically, it looks at the 16 mm Time and Motion Study Projector as used in behavioral research in the 1960s and 1970s. The controversial concept of ‘interactional synchrony’, or the dance-like coordination of people in conversation, emerged from the use of this hand-turned projector. William S. Condon developed the concept along with the technique of microanalysis. Starting with the projector manufactured by Bell (...) & Howell, he made numerous improvements to facilitate observation—‘sweeping’ over segments of very short duration to discover the rhythmic synchrony of all filmed participants. It led him to a theory of ‘process’ in communication, and in the reception of speech in particular. People always ‘danced’ to the tune of their own voice, and their listeners ‘danced’ to the tune of the speaker—at intervals of one-sixth or one-eighth of a second. This also led Condon to an epistemology of discovery derived partly from philosophy but mostly from his machinery. The universe, he said, is a ‘continuum of order’ whose structures are preserved through translations of order: of thought into speech, speech into vibrations, vibrations into neurons, and back into behavior. The only exceptions are people with disabilities, like the autistics Condon studied from the 1970s onward. But the very distinction of normal and pathological was epiphenomenal to his scanning technique; it was rooted in material and formal qualities of film and of the projector whose crank he turned often. (shrink)

When I tell you that it’s raining, I describe a way the world is—viz., rainy. I say something whose truth turns on how things are with the weather in the world. Likewise when I tell you that the weatherman thinks that it’s raining. Here the truth of what I say turns on how things are with the weatherman’s state of mind in the world. Likewise when I tell you that I think that it’s raining. Here the truth of what I (...) say turns on how things are with my state of mind in the world. (shrink)

Understanding human behaviour, thoughts, and emotional expressions can be challenging in the global context. Due to cultural differences, the study of psychology cannot be de-contextualised. This calls for unearthing of the explanatory systems that exist in Africa to understand and account for behaviour, emotions, and cognition of Africans. This call is addressed through the emergence of African Psychology (AP) or Indigenous Psychology in Africa (IPA) as a legitimate science of human experience. This Element discusses the motivations for AP, centrality of (...) culture, demarcations of AP, and the different strands within AP. It highlights issues related to African philosophy, African cultural anthropology, African philosophy of science, and suitable methodological approaches for AP research. It also discusses some selected theoretical contributions and applications of AP. The Element concludes that AP researchers and practitioners need to pursue interdisciplinarity and avoid meaningless rejection of good ideas from other cultural settings. (shrink)

I argue that there is a wide variety of epistemic norms, distributed along two different spectra. One spectrum runs from the ideal to the practical and concerns the extent to which it is possible to follow the norm given our cognitive and epistemic limitations. The other spectrum runs from thin to thick and concerns the extent to which the norm concerns facts about our beliefs over and above the content of the belief. Many putative epistemic norms, such as truth and (...) various conceptions of justification, can be found at different points on the spectra. There is no single obvious privileged point from which to say any of these norms is more fundamental than the others, though there may be some reason to doubt that some of the norms are intrinsically interesting. (shrink)

I develop a conception of expressivism according to which it is chiefly a pragmatic thesis about some fragment of discourse, one imposing certain constraints on semantics. The first half of the paper uses credal expressivism about the language of probability as a stalking-horse for this purpose. The second half turns to the question of how one might frame an analogous form of expressivism about the language of deontic modality. Here I offer a preliminary comparison of two expressivist lines. The first, (...) expectation expressivism, looks again to Bayesian modelling for inspiration: it glosses deontically modal language as characteristically serving to express decision-theoretic expectation (expected utility). The second, plan expressivism, develops the idea (due to Gibbard 2003) that this language serves to express 'plan-laden' states of belief. In the process of comparing the views, I show how to incorporate Gibbard's modelling ideas into a compositional semantics for attitudes and modals, filling a lacuna in the account. I close with the question whether and how plan expressivism might be developed with expectation-like structure. (shrink)

Cancer biology features the ascription of normal functions to parts of cancers. At least some ascriptions of function in cancer biology track local normality of parts within the global abnormality of the aberration to which those parts belong. That is, cancer biologists identify as functions activities that, in some sense, parts of cancers are supposed to perform, despite cancers themselves having no purpose. The present paper provides a theory to accommodate these normal function ascriptions—I call it the Modeling Account of (...) Normal Function (MA). MA comprises two claims. First, normal functions are activities whose performance by the function-bearing part contributes to the self-maintenance of the whole system and, thereby, results in the continued presence of that part. Second, MA holds that models of system-level activities that are (partly) constitutive of self-maintenance are improved by including a representation of the relevant function-bearing part and by making reference to the activity/activities that part performs which contribute(s) to those system-level activities. I contrast MA with two other accounts that seek to explicate the ascription of normal functions in biology, namely, the organizational account and the selected effects account. Both struggle to extend to cancer biology. However, I offer ecumenical readings which allow them to recover some ascriptions of normal function to parts of cancers. So, though I contend that MA excels in this respect, the purpose of this paper is served if it provides materials for bridging the gap between cancer biology, philosophy of cancer, and the literature on function. (shrink)

The temporal structure for motivating, monitoring, and making sense of agency depends on encoding, maintaining, and accessing the right contents at the right times. These functions are facilitated by memory. Moreover, in informing action, memory is itself often active. That remembering is essential to and an expression of agency and is often active suggests that it is a type of action. Despite this, Galen Strawson (Proceedings of the Aristotelian Society, 103, 227–257, 2003) and Alfred Mele (2009) deny that remembering is (...) an action. They claim that memory fails to admit of control. Remembering is automatic—once remembering starts, the process can neither be stopped nor intervened on. Moreover, the agent does not initiate remembering. An agent has control over an event or process if and only if she has the capacity and opportunity to initiate and intervene on that event or process. Actions are events over which an agent has control. Since it is automatic, we fail to have control over remembering. Thus, remembering is not an action. In this paper, I draw out an assumption of Strawson’s and Mele’s accounts: an event-type whose tokens exhibit automaticity cannot, for that reason, be an action (§2). Against this assumption, I draw parallels between skilled bodily action and memory. I show that memory exhibits two defining features of skill: it can be learned with practice and it admits of attributions of excellence (§3). These features reveal how intelligent control is exerted in the exercise of skill despite apparent automaticity—control is gained over time (§4). Since exercises of skill are by definition actions and since memory exemplifies the defining features of skill, memory is a skill and instances of remembering are actions too. (shrink)

This paper defends a counterexample to Modus Tollens, and uses it to draw some conclusions about the logic and semantics of indicative conditionals and probability operators in natural language. Along the way we investigate some of the interactions of these expressions with 'knows', and we call into question the thesis that all knowledge ascriptions have truth-conditions.

This is a study in the meaning of natural language probability operators, sentential operators such as probably and likely. We ask what sort of formal structure is required to model the logic and semantics of these operators. Along the way we investigate their deep connections to indicative conditionals and epistemic modals, probe their scalar structure, observe their sensitivity to contex- tually salient contrasts, and explore some of their scopal idiosyncrasies.

Cancer biologists ascribe normal functions to parts of cancer. Normal functions are activities that parts of systems are in some minimal sense supposed to perform. Cancer biologists’ finding normality within the abnormality of cancer pose difficulties for two main approaches to normal function. One approach claims that normal functions are activities that parts are selected for. However, some parts of cancers that have normal functions aren’t selected to perform them. The other approach claims that normal functions are part-activities typical for (...) the system and that contribute to survival/reproduction. However, cancers are too heterogeneous to establish what’s typical across a type. (shrink)

We investigate a basic probabilistic dynamic semantics for a fragment containing conditionals, probability operators, modals, and attitude verbs, with the aim of shedding light on the prospects for adding probabilistic structure to models of the conversational common ground.

Despite its enduring insights, Durkheim’s theory of suicide fails to account for a significant set of cases because of its overreliance on structural forces to the detriment of other possible factors. In this paper, we develop a new theoretical framework for thinking about the role of culture in vulnerability to suicide. We argue that by focusing on the cultural dynamics of excessive regulation, particularly at the meso level, a more robust sociological model for suicide could be offered that supplements structure-heavy (...) Durkheimian theory. In essence, we argue that the relevance of cultural regulation to suicide rests on the (1) degree to which culture is coherent in sociocultural places, (2) existence of directives related to prescribing or proscribing suicide, (3) degree to which these directives translate into internalized meanings affecting social psychological processes, and (4) degree to which the social space is bounded. We then illustrate how our new theory provides useful insights into three cases of suicide largely neglected within sociology: specifically, suicide clusters in high schools, suicide in the military, and suicides of “despair” among middle-aged white men. We conclude with implications for future sociological research on suicide and suicide prevention. (shrink)

A major approach to the ethics of artificial intelligence is to use social choice, in which the AI is designed to act according to the aggregate views of society. This is found in the AI ethics of “coherent extrapolated volition” and “bottom–up ethics”. This paper shows that the normative basis of AI social choice ethics is weak due to the fact that there is no one single aggregate ethical view of society. Instead, the design of social choice AI faces three (...) sets of decisions: standing, concerning whose ethics views are included; measurement, concerning how their views are identified; and aggregation, concerning how individual views are combined to a single view that will guide AI behavior. These decisions must be made up front in the initial AI design—designers cannot “let the AI figure it out”. Each set of decisions poses difficult ethical dilemmas with major consequences for AI behavior, with some decision options yielding pathological or even catastrophic results. Furthermore, non-social choice ethics face similar issues, such as whether to count future generations or the AI itself. These issues can be more important than the question of whether or not to use social choice ethics. Attention should focus on these issues, not on social choice. (shrink)

Spoken word recognition involves a perceptual tradeoff between the reliance on the incoming acoustic signal and knowledge about likely sound categories and their co-occurrences as words. This study examined how adult second language (L2) learners navigate between acoustic-based and knowledge-based spoken word recognition when listening to highly variable, multi-talker truncated speech, and whether this perceptual tradeoff changes as L2 listeners gradually become more proficient in their L2 after multiple months of structured classroom learning. First language (L1) Mandarin Chinese listeners and (...) L1 English-L2 Mandarin adult listeners took part in a gating experiment. The L2 listeners were tested twice—once at the start of their intermediate/advanced L2 language class and again two months later. L1 listeners were only tested once. Participants were asked to identify syllable-tone words that varied in syllable token frequency (high/low according to a spoken word corpus) and syllable-conditioned tonal probability (most probable/least probable in speech given the syllable). The stimuli were recorded by 16 different talkers and presented at eight gates ranging from onset-only (gate 1) through onset+40 ms increments (gates 2 through 7) to the full word (gate 8). Mixed-effects regression modeling was used to compare performance to our previous study which used single-talker stimuli (Wiener, Lee, & Tao, 2019). The results indicated that multi-talker speech caused both L1 and L2 listeners to rely greater on knowledge-based processing of tone. L1 listeners were able to draw on distributional knowledge of syllable-tone probabilities in early gates and switch to predominantly acoustic-based processing when more of the signal was available. In contrast, L2 listeners, with their limited experience with talker range normalization, were less able to effectively transition from probability-based to acoustic-based processing. Moreover, for the L2 listeners, the reliance on such distributional information for spoken word recognition appeared to be conditioned by the nature of the acoustic signal. Single-talker speech did not result in the same pattern of probability-based tone processing, suggesting that knowledge-based processing of L2 speech may only occur under certain acoustic conditions, such as multi-talker speech. (shrink)

Focusing on cases which involve binding into epistemic modals with definite descriptions and quantifiers, I raise some new problems for standard approaches to all of these expressions. The difficulties are resolved in a semantic framework that is dynamic in character. I close with a new class of problems about de re readings within the scope of modals.

Purpose This paper aims to formalize long-term trajectories of human civilization as a scientific and ethical field of study. The long-term trajectory of human civilization can be defined as the path that human civilization takes during the entire future time period in which human civilization could continue to exist. -/- Design/methodology/approach This paper focuses on four types of trajectories: status quo trajectories, in which human civilization persists in a state broadly similar to its current state into the distant future; catastrophe (...) trajectories, in which one or more events cause significant harm to human civilization; technological transformation trajectories, in which radical technological breakthroughs put human civilization on a fundamentally different course; and astronomical trajectories, in which human civilization expands beyond its home planet and into the accessible portions of the cosmos. -/- Findings Status quo trajectories appear unlikely to persist into the distant future, especially in light of long-term astronomical processes. Several catastrophe, technological transformation and astronomical trajectories appear possible. -/- Originality/value Some current actions may be able to affect the long-term trajectory. Whether these actions should be pursued depends on a mix of empirical and ethical factors. For some ethical frameworks, these actions may be especially important to pursue. (shrink)

This paper discusses means for promoting artificial intelligence that is designed to be safe and beneficial for society. The promotion of beneficial AI is a social challenge because it seeks to motivate AI developers to choose beneficial AI designs. Currently, the AI field is focused mainly on building AIs that are more capable, with little regard to social impacts. Two types of measures are available for encouraging the AI field to shift more toward building beneficial AI. Extrinsic measures impose constraints (...) or incentives on AI researchers to induce them to pursue beneficial AI even if they do not want to. Intrinsic measures encourage AI researchers to want to pursue beneficial AI. Prior research focuses on extrinsic measures, but intrinsic measures are at least as important. Indeed, intrinsic factors can determine the success of extrinsic measures. Efforts to promote beneficial AI must consider intrinsic factors by studying the social psychology of AI research communities. (shrink)

Public discourse implicitly appeals to what we call the “Traumatic Untrustworthiness Argument” (TUA). To motivate, articulate, and assess the TUA, we appeal to Hawley’s (2019) commitment account of trust and trustworthiness. On Hawley’s account, being trustworthy consists in the successful avoidance of unfulfilled commitments and involves three components: the actual avoidance of unfulfilled commitments, sincerity in one’s taking on elective commitments, and competence in fulfilling commitments one has incurred. In contexts of testimony, what’s at issue is the speaker’s competence and (...) sincere intention to speak truthfully. The TUA targets trauma victims’ competence rather than their sincerity. According to the TUA, empirical evidence shows that trauma undermines victims’ trustworthiness with regard to speaking truthfully about their trauma by undermining their competence to remember the relevant event. We argue that what the evidence shows is rather that remembering traumatic events involves a distinct “mode of manifesting” the competence to remember particular events from the personal past. Trauma victims are competent to speak truthfully about their trauma and ought to be trusted at least with regard to the central details of the event. By suggesting otherwise, the TUA threatens an insidious form of epistemic injustice which Hawley’s account helps us locate. (shrink)

The view that actually has a reading on which it is a two-dimensional indexical modal operator has some problems.

Gottlob Frege (1848–1925) was unquestionably one of the most important philosophers of all time. He trained as a mathematician, and his work in philosophy started as an attempt to provide an explanation of the truths of arithmetic, but in the course of this attempt he not only founded modern logic but also had to address fundamental questions in the philosophy of language and philosophical logic. Frege is generally seen (along with Russell and Wittgenstein) as one of the fathers of the (...) analytic method, which dominated philosophy in English-speaking countries for most of the twentieth century. His work is studied today not just for its historical importance but also because many of his ideas are still seen as relevant to current debates in the philosophies of logic, language, mathematics and the mind. The Cambridge Companion to Frege provides a route into this lively area of research. (shrink)

As Quine observed, the following sentence has a reading which, if true, would be of special interest to the authorities: Ralph believes that someone is a spy. This is the reading where the quantifier is naturally understood as taking wide scope relative to the attitude verb and as binding a variable within the scope of the attitude verb. This essay is interested in addressing the question what the semantic analysis of this kind of reading should look like from a Fregean (...) perspective—a perspective according to which attitude states are generally relations to structured Fregean thoughts composed of senses. The Fregean view faces a challenge of compositionality here. This essay describes the challenge and offers a response on the Fregean's behalf. (shrink)

A recent report by Persaud et al. [Persaud, N., McLeod, P. & Cowey, A. . Post-decision wagering objectively measures awareness. Nature Neuroscience 10, 257–261] addresses a fundamental issue in consciousness science: the experimental measurement of conscious content. The authors propose a novel technique, ‘post-decision wagering’, in which subjects place bets on the correctness of decisions or discriminations. In this note, I critique the authors’ claim that their method “measures awareness directly”.

Materialist metaphysicians want to side with physics, but not to take sides within physics.If we took literally the claim of a materialist that his position is simply belief in the claim that all is matter, as currently conceived, we would be faced with an insoluble mystery. For how would such a materialist know how to retrench when his favorite scientific hypotheses fail? How did the 18th century materialist know that gravity, or forces in general, were material? How did they know (...) in the 19th century that the electromagnetic field was material, and persisted in this conviction after the aether had been sent packing?The doctrine of physicalism casts a long shadow in contemporary philosophy, configuring all kinds of philosophical issues and projects. Unsurprisingly, its proponents argue that physicalism has all the obvious features necessary for a scientific hypothesis to be in what we will call ‘good standing,’ i.e. being worthy of serious scientific investigation. In fact, many claim much more, arguing that physicalism is a well-confirmed hypothesis and possibly amongst the best of our theories. But, as our second opening passage makes clear, a persistent worry has been that physicalism, or ‘materialism’ as van Fraassen terms it, is an edifice built on sand. For many philosophers question whether the ‘physical’ can be specified at all, or at least in a manner that will produce a physicalism that would be in good standing. (shrink)

By leveraging the case of Hindu sati, this paper elucidates the ways in which structure and culture condition suicidal behavior by way of social psychological and emotional dynamics. Conventionally, sati falls under Durkheim's discussion of altruistic suicides, or the self-sacrifice of underindividuated or excessively integrated peoples like widows in traditional societies. In light of the fact that Durkheim's interpretation was based on uneven data, nineteenth century Eurocentric beliefs, and a theoretical framework that can no longer resist modification and elaboration, by (...) reconsidering sati it is possible to sketch a new model that strengthens Durkheim's theory by making it more robust and generalizable. The following model is built on five principles. First, integration and regulation are not distinct causal forces, but overlapping contextual conditions. Second, to better explain the variation in suicidality across time and space, we must also pay attention to culture as it provides the underlying meanings of suicide that can increase the odds a person or class of persons become suicidal or are protected against suicidality. Third, structure still matters, but in many cases, the role power and power-differentials play must be considered. Fourth, understanding why and how people choose suicide depends on incorporating identity and status processes. Fifth, because the expression of social emotions like shame are patterned by structural and cultural conditions, to understand how suicidality is socioculturally patterned we must further explore the link between identity/status, social emotions, and structure and culture. (shrink)

Materialist metaphysicians want to side with physics, but not to take sides within physics.If we took literally the claim of a materialist that his position is simply belief in the claim that all is matter, as currently conceived, we would be faced with an insoluble mystery. For how would such a materialist know how to retrench when his favorite scientific hypotheses fail? How did the 18th century materialist know that gravity, or forces in general, were material? How did they know (...) in the 19th century that the electromagnetic field was material, and persisted in this conviction after the aether had been sent packing?The doctrine of physicalism casts a long shadow in contemporary philosophy, configuring all kinds of philosophical issues and projects. Unsurprisingly, its proponents argue that physicalism has all the obvious features necessary for a scientific hypothesis to be in what we will call ‘good standing,’ i.e. being worthy of serious scientific investigation. In fact, many claim much more, arguing that physicalism is a well-confirmed hypothesis and possibly amongst the best of our theories. But, as our second opening passage makes clear, a persistent worry has been that physicalism, or ‘materialism’ as van Fraassen terms it, is an edifice built on sand. For many philosophers question whether the ‘physical’ can be specified at all, or at least in a manner that will produce a physicalism that would be in good standing. (shrink)

Institutional differentiation has been one of the central concerns of sociology since the days of Auguste Comte. However, the overarching tendency among institutionalists such as Durkheim or Spencer has been to treat the process of differentiation from a macro, "outside in" perspective. Missing from this analysis is how institutional differentiation occurs from the "inside out, "or through the efforts and struggles of individual and corporate actors. Despite the recent efforts of the "new institutionalism" to fill in this gap, a closer (...) look at the literature will uncover the fact that (1) it has tended to conflate macro-level institutions and meso-level organizations and (2) this has led to a taken for granted approach to institutional dynamics. This article seeks to develop a general theory of institutional autonomy; autonomy is a function of the degree to which specialized corporate units are structurally and symbolically independent of other corporate units. It is argued herein that the process by which these "institutional entrepreneurs" become independent can explain how institutions become differentiated from the "inside out." Moreover, this article offers five dimensions that can be operationalized, measuring the degree to which institutions are autonomous. (shrink)