Robustness concepts are often invoked to manage two obstacles confronting models of ecological systems: complexity and uncertainty. The intuitive idea is that any result derived from many idealized but credible models is thereby made more reliable or is better confirmed. An appropriate basis for this inference has proven elusive. Here, several representations of robustness analysis are vetted, paying particular attention to complex models of ecosystems and the global climate. The claim that robustness is itself confirmatory because robustness analysis employs a Bayesian variety-of-evidence argument is criticized, but recent overwhelming pessimism about robustness may have a silver lining.

1. Introduction. David Malament (1995) has described a natural and satisfying resolution of the traditional problems of Newtonian cosmology—natural in the sense that it effects the escape by altering Newtonian gravitation theory in a way that leaves its observational predictions completely unaffected. I am in full agreement with his approach. There is one part of his account, however, over which Malament has been excessively modest. The resolution requires a modification to Newtonian gravitation theory. Malament presents the modification as so straightforward as to be automatic. This trivializes the crucial postulate, which I shall call the “relativity of acceleration.” It is a significant physical statement in its own right and requires careful justification. Moreover the postulate proved easy to overlook for decades of discussion of the paradox. It really only becomes natural from the perspective of the newer geometric methods Malament exploits. There the postulate has become a commonplace. My purpose here is to develop the following:

• While Newtonian cosmology can be repaired satisfactorily, in its traditional form it remains deeply troubled. These troubles can be expressed most vividly as the paradoxical contradictions indicated below. They persist in both the integral and differential formulations of Newtonian gravitation theory. (Section 2)

• Malament's careful geometric treatment is necessarily dense. By taking some liberties with precision, his core result can be expressed in a far simpler form. (Section 4)

• Attempts to avoid the resolution Malament describes do lead to disaster. Therefore this episode can be inverted and used as the strongest extant argument for the relativity of acceleration in Newtonian gravitation theory. (Section 5)

This article describes one style of functional analysis commonly used in the neurosciences called “task-bound functional analysis.” The concept of function invoked by this style of analysis is distinctive in virtue of the dependence relations it bears to transient environmental properties. It is argued that task-bound functional analysis cannot explain the presence of structural properties in nervous systems. An alternative concept of neural function is introduced that draws on the theoretical neuroscience literature, and an argument is given to show that this alternative concept may help to overcome the explanatory limitations of task-bound functional analysis.

In order to judge whether a theory is empirically adequate one must have epistemic access to reliable records of past measurement results that can be compared against the predictions of the theory. Some formulations of quantum mechanics fail to satisfy this condition. The standard theory without the collapse postulate is an example. Bell's reading of Everett's relative-state formulation is another. Furthermore, there are formulations of quantum mechanics that only satisfy this condition for a special class of observers, formulations whose empirical adequacy could only be judged by an observer who records her measurement results in a special way. Bohm's theory is an example. It is possible to formulate hidden-variable theories that do not suffer from such a restriction, but these encounter other problems.

A number of formal models, including a highly influential model from Hong and Page, purport to show that functionally diverse groups often beat groups of individually high-performing agents in solving problems. Thompson argues that in Hong and Page’s model, that the diverse groups are created by a random process explains their success, not the diversity. Here, I defend the diversity interpretation of the Hong and Page result. The failure of Thompson’s argument shows that to understand the value of functional diversity, we should be clearer about how we conceive of and measure that diversity.

While philosophical discussions of first-person methods often turn on the veridicality of first-person reports, I argue that more attention should be paid to the circumstances and aims of their experimental production in the science of perception. After pointing to the ‘constructedness’ of first-person reports, I raise questions about the criteria by which to judge whether they illuminate something about the nature of perception. I illustrate this point with a historical debate between Gestalt psychologists and atomists, both of whom used first-person methods to investigate perception, but who disagreed deeply over the epistemic value of their respective first-person data.

Murray Gell-Mann has proposed the concept of effective complexity as a measure of information content. The effective complexity of a string of digits is defined as the algorithmic complexity of the regular component of the string. This paper argues that the effective complexity of a given string is not uniquely determined. The effective complexity of a string admitting a physical interpretation, such as an empirical data set, depends on the cognitive and practical interests of investigators. The effective complexity of a string as a purely formal construct, lacking a physical interpretation, is either close to zero, or equal to the string's algorithmic complexity, or arbitrary, depending on the auxiliary criterion chosen to pick out the regular component of the string. Because of this flaw, the concept of effective complexity is unsuitable as a measure of information content.

The movement for the Unity of Science has been making headway for a number of years, and each year more workers in the various scientific domains have expressed their interest in it by participating in the discussion of its aims and its specific problems. A characteristic feature of this movement which explains this growing interest is that it does not propose a “super-science” which is to legislate to the special sciences. Its proposal of a Unity of Science is of a different kind, not based upon some grand metaphysical view of any sort. Those who are active within this movement are emphatic in their insistence that instead of aiming at a synthesis of the different sciences on the basis of a prior and independent philosophy, the special sciences will themselves supply their own synthesizing glue. For appreciating the significance of this movement it is therefore necessary to keep in mind that its tendency is toward a unified science, departmentalized into special sciences, and not toward an artificial and speculative juxtaposition of an autonomous philosophy and an autonomous group of sciences.

It is a common assumption in contemporary cognitive neuroscience that discovering a putative realized kind to be dissociably realized (i.e., to be realized in each instance by two or more distinct realizers) mandates splitting that kind. Here I explore some limits on this inference using two deceptively similar examples: the dissociation of declarative and procedural memory and Ramachandran's argument that the self is an illusion.

It has been argued, most trenchantly by Nancy Cartwright, that the diversity of the concepts and regularities we actually use to describe nature and predict and explain its behavior leaves us with no reason to believe that our foundational physical theories actually “apply” outside of delicately contrived systems within the laboratory. This paper argues that, diversity of method notwithstanding, there is indeed good reason to think that the foundational laws of physics are universal in their scope.

In this article I analyze process tracing, a causal mechanism-based technique for testing causal claims in the social sciences that requires one to specify a chain of intervening causes between any putative cause and effect. I argue that one should not only give evidence that the intervening causes are present in a suitable case study, as process tracing methodologists recommend, but also provide counterfactual evidence to show that each link in the chain is genuinely causal. I detail what that counterfactual evidence should consist of, using Woodward’s manipulability theory, and argue that this evidence relies on tentative comparisons to other case studies.

The large number of writers who have in recent years attacked the problem of the logical nature of commands appear generally in agreement in accepting the distinction of common grammar between imperative and declarative sentences as representing, albeit in no clear one-to-one manner, some real difference in the logical character of the two types of expression, and possibly in the psychological sign-functioning mechanism itself. The crucial logical difference adduced is that commands can apparently rot be classified as true or false. One is then, however, confronted with the problem of interpreting arguments involving imperatives which appear syllogistic in form, such as:

Keep your promises!

You promised to pay on Wednesday.

Pay on Wednesday!

Fine has recently proved the surprising result that satisfaction of the Bell inequality in a Clauser-Horne experiment implies the existence of joint probabilities for pairs of noncommuting observables in the experiment. In this paper we show that if probabilities are interpreted in the von Mises-Church sense of relative frequencies on random sequences, a proof of the Bell inequality is nonetheless possible in which such joint probabilities are assumed not to exist. We also argue that Fine's theorem and related results do not impugn the common view that local realists are committed to the Bell inequality.

The first use of the term “information” to describe the content of nervous impulse occurs in Edgar Adrian's The Basis of Sensation (1928). What concept of information does Adrian appeal to, and how can it be situated in relation to contemporary philosophical accounts of the notion of information in biology? The answer requires an explication of Adrian's use and an evaluation of its situation in relation to contemporary accounts of semantic information. I suggest that Adrian's concept of information can be to derive a concept of arbitrariness or semioticity in representation. This in turn provides one way of resolving some of the challenges that confront recent attempts in the philosophy of biology to restrict the notion of information to those causal connections that can in some sense be referred to as arbitrary or semiotic.

Bell's Theorem is proved for locality and conservation formulated in terms of subjunctive conditionals with chance consequents, rather than the usual conditional probability formulation. This brings into sharp focus the minimal counterfactual assumptions needed for Bell's theorem.

Philosophers frequently cite Dalton's chemical atomism, and its nineteenth century developments, as a prime example of inference to the best explanation. This was a controversial issue in its time. But the critics are dismissed as positivist-inspired antirealists with no interest in explanation. Is this a reasonable assessment?

I challenge the usual approach of defining emergence in terms of properties of wholes “emerging” upon properties of parts. This approach indeed fails to meet the requirement of nontriviality, since it renders a bunch of ordinary properties emergent; however, by defining emergence as the incompressibility of a simulation process, we have an objective meaning of emergence because the difference between the processes satisfying the incompressibility criterion and the other processes does not depend on our cognitive abilities. Finally, this definition fulfills the nontriviality and the scientific-adequacy requirements better than the combinatorial approach, emergence here being a predicate of processes rather than of properties.

A hallmark of ecological research is dealing with complexity in the systems under investigation. One strategy is to diminish this complexity by constructing models and theories that are general. Alternatively, ecologists can constrain the scope of their generalizations to particular phenomena or types of systems. However, research employing the second strategy is often met with scathing criticism. I offer a theoretical argument in support of moderate generalizations in ecological research, based on the notions of interdependence and causal heterogeneity and their effect on the trade-off between generality and realism.

Arguments about the evolutionary function of phenomenal consciousness are beset by the problem of epiphenomenalism. For if it is not clear whether phenomenal consciousness has a causal role, then it is difficult to begin an argument for the evolutionary role of phenomenal consciousness. We argue that complexity arguments offer a way around this problem. According to evolutionary biology, the structural complexity of a given organ can provide evidence that the organ is an adaptation, even if nothing is known about the causal role of the organ. Evidence from cognitive neuropsychology suggests that phenomenal consciousness is structurally complex in the relevant way, and this provides prima facie evidence that phenomenal consciousness is an adaptation. Furthermore, we argue that the complexity of phenomenal consciousness might also provide clues about the causal role of phenomenal consciousness.

This paper discusses a philosophical issue in taxonomy. At least one philosopher has suggested the taxonomic principle that scientific kinds are disjoint. An opposing position is defended here by marshalling examples of nondisjoint categories which belong to different, coexisting classification schemes. This denial of the disjointness principle can be recast as the claim that scientific classification is “interest-relative”. But why would anyone have held that scientific categories are disjoint in the first place? It is argued that this assumption is needed in one attempt to derive essentialism. This shows why the essentialist and interest-relative approaches to classification are in conflict.