This monograph looks at causal nets from a philosophical point of view. The author shows that one can build a general philosophical theory of causation on the basis of the causal nets framework that can be fruitfully used to shed new light on philosophical issues. Coverage includes both a theoretical as well as application-oriented approach to the subject. The author first counters David Hume’s challenge about whether causation is something ontologically real. The idea behind this is (...) that good metaphysical concepts should behave analogously to good theoretical concepts in scientific theories. In the process, the author offers support for the theory of causal nets as indeed being a correct theory of causation. Next, the book offers an application-oriented approach to the subject. The author shows that causal nets can investigate philosophical issues related to causation. He does this by means of two exemplary applications. The first consists of an evaluation of Jim Woodward’s interventionist theory of causation. The second offers a contribution to the new mechanist debate. Introductory chapters outline all the formal basics required. This helps make the book useful for those who are not familiar with causal nets, but interested in causation or in tools for the investigation of philosophical issues related to causation. (shrink)

We start this paper by arguing that causality should, in analogy with force in Newtonian physics, be understood as a theoretical concept that is not explicated by a single definition, but by the axioms of a theory. Such an understanding of causality implicitly underlies the well-known theory of causal nets and has been explicitly promoted by Glymour. In this paper we investigate the explanatory warrant and empirical content of TCN. We sketch how the assumption of directed cause–effect relations (...) can be philosophically justified by an inference to the best explanation. We then ask whether the explanations provided by TCN are merely post-facto or have independently testable empirical content. To answer this question we develop a fine-grained axiomatization of TCN, including a distinction of different kinds of faithfulness. A number of theorems show that although the core axioms of TCN are empirically empty, extended versions of TCN have successively increasing empirical content. (shrink)

In this paper I reconstruct and evaluate the validity of two versions of causal exclusion arguments within the theory of causal Bayes nets. I argue that supervenience relations formally behave like causal relations. If this is correct, then it turns out that both versions of the exclusion argument are valid when assuming the causal Markov condition and the causal minimality condition. I also investigate some consequences for the recent discussion of causal exclusion arguments (...) in the light of an interventionist theory of causation such as Woodward's (2003) and discuss a possible objection to my causal Bayes net reconstruction. (shrink)

Bayesian nets are widely used in artificial intelligence as a calculus for causal reasoning, enabling machines to make predictions, perform diagnoses, take decisions and even to discover causal relationships. This book, aimed at researchers and graduate students in computer science, mathematics and philosophy, brings together two important research topics: how to automate reasoning in artificial intelligence, and the nature of causality and probability in philosophy.

In this paper we develop an analysis of dispositions by means of causal Bayes nets. In particular, we analyze dispositions as cause-effect structures that increase the probability of the manifestation when the stimulus is brought about by intervention in certain circumstances. We then highlight several advantages of our analysis and how it can handle problems arising for classical analyses of dispositions such as masks, mimickers, and finks.

Causal Bayes nets have been developed in philosophy, statistics, and computer sciences to provide a formalism to represent causal structures, to induce causal structure from data and to derive predictions. Causal Bayes nets have been used as psychological theories in at least two ways. They were used as rational, computational models of causal reasoning and they were used as formal models of mental causal models. A crucial assumption made by them is the (...) Markov condition, which informally states that variables are independent of other variables that are not their direct or indirect effects conditional on their immediate causes. Whether people’s inferences conform to the causal Markov and the faithfulness condition has recently been investigated empirically. A review of respective research indicates that inferences frequently violate these conditions. This finding challenges some uses of causal Bayes nets in psychology. They entail that causal Bayes nets may not be appropriate to derive predictions for causal model theories of causal reasoning. They also question whether causal Bayes nets as a rational model are empirically descriptive. They do not challenge, however, causal Bayes nets as normative models and their usage as formal models of causal reasoning. (shrink)

One of Stuart Glennan's most prominent contributions to the new mechanist debate consists in his reductive analysis of higher-level causation in terms of mechanisms (Glennan, 1996). In this paper I employ the causal Bayes net framework to reconstruct his analysis. This allows for specifying general assumptions which have to be satis ed to get Glennan's approach working. I show that once these assumptions are in place, they imply (against the background of the causal Bayes net machinery) that higher-level (...) causation indeed reduces to interactions between component parts of mechanisms. I also briefly discuss the plausibility of these assumptions and some consequences for the mechanism debate. (shrink)

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)

How should we reason with causal relationships? Much recent work on this question has been devoted to the theses (i) that Bayesian nets provide a calculus for causal reasoning and (ii) that we can learn causal relationships by the automated learning of Bayesian nets from observational data. The aim of this book is to..

The paper displays the similarity between the theory of probabilistic causation developed by Glymour et al. since 1983 and mine developed since 1976: the core of both is that causal graphs are Bayesian nets. The similarity extends to the treatment of actions or interventions in the two theories. But there is also a crucial difference. Glymour et al. take causal dependencies as primitive and argue them to behave like Bayesian nets under wide circumstances. By contrast, I (...) argue the behavior of Bayesian nets to be ultimately the defining characteristic of causal dependence. (shrink)

Causal Bayes nets (CBNs) can be used to model causal relationships up to whole mechanisms. Though modelling mechanisms with CBNs comes with many advantages, CBNs might fail to adequately represent some biological mechanisms because—as Kaiser (2016) pointed out—they have problems with capturing relevant spatial and structural information. In this paper we propose a hybrid approach for modelling mechanisms that combines CBNs and cellular automata. Our approach can incorporate spatial and structural information while, at the same time, it (...) comes with all the merits of a CBN representation of mechanisms. (shrink)

This comment discusses the suggestions made in Mahoney’s “Constructivist Set-Theoretic Analysis: An Alternative to Essentialist Social Science” (2023). Mahoney presents an approach to cases of intersectionality or confounding which produce causal results unlike those that result from traditional net effects causal modeling. He presents it as an alternative to “essentialism,” which he describes as a cognitive error. These alternatives have the same problems as those he attributes to net effects analysis, with one exception: the method does allow for (...) the identification of subpopulations with different causal mechanisms producing different outcomes for this subpopulation. (shrink)

We outline a cognitive and computational account of causal learning in children. We propose that children employ specialized cognitive systems that allow them to recover an accurate “causal map” of the world: an abstract, coherent representation of the causal relations among events. This kind of knowledge can be perspicuously represented by the formalism of directed graphical causal models, or “Bayes nets”. Human causal learning and inference may involve computations similar to those for learnig (...) class='Hi'>causal Bayes nets and for predicting with them. Preliminary experimental results suggest that 2- to 4-year-old children spontaneously construct new causal maps and that their learning is consistent with the Bayes-Net formalism. (shrink)

We propose that children employ specialized cognitive systems that allow them to recover an accurate “causal map” of the world: an abstract, coherent, learned representation of the causal relations among events. This kind of knowledge can be perspicuously understood in terms of the formalism of directed graphical causal models, or “Bayes nets”. Children’s causal learning and inference may involve computations similar to those for learning causal Bayes nets and for predicting with them. Experimental (...) results suggest that 2- to 4-year-old children construct new causal maps and that their learning is consistent with the Bayes net formalism. (shrink)

These are chapters from a book forthcoming from MIT Press. Comments to the author at [email protected] would be most welcome. Still time for changes.

Bayesian networks are computer programs which represent probabilitistic relationships graphically as directed acyclic graphs, and which can use those graphs to reason probabilistically , often at relatively low computational cost. Almost every expert system in the past tried to support probabilistic reasoning, but because of the computational difficulties they took approximating short-cuts, such as those afforded by MYCIN's certainty factors. That all changed with the publication of Judea Pearl's Probabilistic Reasoning in Intelligent Systems, in 1988, which synthesized a decade of (...) research making accurate graphical probabilistic reasoning computationally achievable.Bayesian network technology is now one of the fastest growing fields of research in artificial intelligence. That it has become a publication industry in its own right is shown by a search on Google scholar :This development, together with a parallel related growth in the use of causal discovery algorithms which automate the learning of Bayesian networks from sample data, has generated considerable interest, and controversy, within the philosophy-of-science community.Three central questions bringing together AI researchers and philosophers of science are: Are Bayesian networks Bayesian? What is the relation between probability and causality? Are the assumptions behind causal discovery of Bayesian networks realistic or fantastical?Jon Williamson, as a philosopher of science with a keen interest in the technology, asks and answers these questions in his new book. Although it is self-contained, his book is not very likely as an introduction to the technology , nor is it optimal even as an introduction to the philosophical problems in interpreting Bayesian networks . Rather Williamson's book is an attempt to move the debate forward by solving the central problems of the …. (shrink)

Many of our decisions pertain to causal systems. Nevertheless, only recently has it been claimed that people use causal models when making judgments, decisions and predictions, and that causal Bayes nets allow us to formally describe these inferences. Experimental research has been limited to simple, artificial problems, which are unrepresentative of the complex dynamic systems we successfully deal with in everyday life. For instance, in social interactions, we can explain the actions of other's on the fly (...) and we can generalize from limited observations to predict future actions and their consequences. Our main argument is that none of these inferences (i.e., induction, generalization, explanation, and prediction) can be achieved without causal reasoning. As a case in point we use the popular television series desperate housewives and show how causal Bayes nets are able to explain the inferences made in social contexts. Crucially, causal Bayes nets also allow us to understand why we can infer so much from so little when making sense of a protagonist's behavior. (shrink)

This chapter explains the screening‐off rule in the psychological laboratory. The Markov assumption states that any variable in a set is independent in probability of all its ancestors in the set conditional on its own parents. The screening‐off rule is also critical to allow Bayes nets to make an inference of the state of an unknown variable in a causal structure from the states of other variables in that structure. The chapter examines which causal representations people use (...) to make predictions and whether people conform to the screening‐off rule with respect to a causal model they have in mind. It analyzes what leads them to modify their causal model and whether or not people are causally oriented when making predictions. Making probability judgments is hard and requires careful deliberation. People are capable of such deliberation although they avoid it until they cannot, until the facts require more careful thought. (shrink)

Bayes nets are formal representations of causal systems that many psychologists have claimed as plausible mental representations. One purported advantage of Bayes nets is that they may provide a theory of counterfactual conditionals, such as If Calvin had been at the party, Miriam would have left early. This article compares two proposed Bayes net theories as models of people's understanding of counterfactuals. Experiments 1-3 show that neither theory makes correct predictions about backtracking counterfactuals (in which the event (...) of the if-clause occurs after the event of the then-clause), and Experiment 4 shows the same is true of forward counterfactuals. An amended version of one of the approaches, however, can provide a more accurate account of these data. (shrink)

Hitchcock demonstrated that the validity of causal exclusion arguments as well as the plausibility of several of their premises hinges on the specific theory of causation endorsed. In this paper I show that the validity of causal exclusion arguments—if represented within the theory of causal Bayes nets the way Gebharter suggests—actually requires much weaker premises than the ones which are typically assumed. In particular, neither completeness of the physical domain nor the no overdetermination assumption are required.

This paper expounds that besides the well-known spatio-temporal problem there is a causal problem of entanglement: even when one neglects spatio-temporal constraints, the peculiar statistics of EPR/B experiment is inconsistent with usual principles of causal explanation as stated by the theory of causal Bayes nets. The conflict amounts to a dilemma that either there are uncaused correlations or there are caused independences . I argue that the central ideas of causal explanations can be saved if (...) one accepts the latter horn and explains the unfaithful independences by a stable fine-tuning of the causal parameters. (shrink)

The idea that causation can be reduced to transmission of an amount of some conserved quantity between events is spelled out and defended against important objections. Transmission is understood as a symmetrical relation of copresence in two distinct events. The actual asymmetry of causality has its origin in the asymmetrical character of certain irreversible physical processes and then spreads through the causal net. This conception is compatible with the possibility of backwards causation and with a causal theory of (...) time. Genidentity, the persistence of concrete objects, can be given an explanation in causal terms. The transmission theory is shown to escape difficulties faced by two important alternative theories of causation: Salmon's (1984) Mark Transmission Theory and Dowe's (1992a) Conserved Quantities Theory. (shrink)

An accessible and contemporary introduction to the methods for determining cause and effect in the social sciences Causal inference encompasses the tools that allow social scientists to determine what causes what. Economists--who generally can't run controlled experiments to test and validate their hypotheses--apply these tools to observational data to make connections. In a messy world, causal inference is what helps establish the causes and effects of the actions being studied, whether the impact (or lack thereof) of increases in (...) the minimum wage on employment, the effects of early childhood education on incarceration later in life, or the introduction of malaria nets in developing regions on economic growth. Scott Cunningham introduces students and practitioners to the methods necessary to arrive at meaningful answers to the questions of causation, using a range of modeling techniques and coding instructions for both the R and Stata programming languages. - -. (shrink)

Two key ideas of scientific explanation−explanation as causal information and explanation as unification-have frequently been set into mutual opposition. This paper proposes a “dialectical solution” to this conflict, by arguing that causal explanations are preferable to non-causal ones, because they lead to a higherdegree of unification at the level of explaining statistical regularities. The core axioms of the theory of causal nets (TC) are justified because they offer the best if not the only unifying explanation (...) of two statistical phenomena: screening off and linking up. Alternative explanations of the two phenomena are discussed and it isshown why they don’t work. It is demonstrated that although the core axioms of TC are empirically vacuous, extended versions of TC have empirical content by means of which they can generate independently testable predictions.Con frecuencia se han planteado como contrapuestas dos ideas clave en la explicacion cientifica (explicacion como informacion causal y explicacion como unificacion). El presente articulo propone una “solucion dialectica” argumentando que las explicaciones causales son preferibles a las no-causales porque aquellas comportan un mayor grado de unificacion en la explicacion de regularidades estadisticas. Los axiomas centrales de la teoria de redes causales (TC) estan justificados porque ofrecen la mejor, si no la unica, explicacion unificada de dos fenomenos estadisticos: neutralizacion (screening off) y vinculacion (linking up). Se discuten las explicaciones alternativas de estos dos fenomenos y se razona por que no funcionan. Se demuestra ademas que aunque los axiomas centrales de TC son empiricamente vacuos, las versiones extendidas de TC tienen un contenido empirico gracias al cual pueden generar predicciones independientemente contrastables. (shrink)

Two key ideas of scientific explanation - explanations as causal information and explanation as unification - have frequently been set into mutual opposition. This paper proposes a "dialectical solution" to this conflict, by arguing that causal explanations are preferable to non-causal explanations because they lead to a higher degree of unification at the level of the explanation of statistical regularities. The core axioms of the theory of causal nets are justified because they give the best (...) if not the only unifying explanation of two statistical phenomena: screening off and linking up. Alternative explanation attempts are discussed and it is shown why they don't work. It is demonstrated that not the core of TC but extended versions of TC have empirical content, by means of which they can generate independently testable predictions. (shrink)

Mechanistic philosophy of science views a large part of scientific activity as engaged in modelling mechanisms. While science textbooks tend to offer qualitative models of mechanisms, there is increasing demand for models from which one can draw quantitative predictions and explanations. Casini et al. (Theoria 26(1):5–33, 2011) put forward the Recursive Bayesian Networks (RBN) formalism as well suited to this end. The RBN formalism is an extension of the standard Bayesian net formalism, an extension that allows for modelling the hierarchical (...) nature of mechanisms. Like the standard Bayesian net formalism, it models causal relationships using directed acyclic graphs. Given this appeal to acyclicity, causal cycles pose a prima facie problem for the RBN approach. This paper argues that the problem is a significant one given the ubiquity of causal cycles in mechanisms, but that the problem can be solved by combining two sorts of solution strategy in a judicious way. (shrink)

The causal Bayes net framework specifies a set of axioms for causal discovery. This article explores the set of causal variables that function as relata in these axioms. Spirtes showed how a causal system can be equivalently described by two different sets of variables that stand in a non-trivial translation-relation to each other, suggesting that there is no “correct” set of causal variables. I extend Spirtes’ result to the general framework of linear structural equation models (...) and then explore to what extent the possibility to intervene or a preference for simpler causal systems may help in selecting among sets of causal variables. (shrink)

I highlight a metaphysical concern that stands in the way of more widespread adoption of causal modeling techniques such as causal Bayes nets. Researchers in some fields may resist adoption due to concerns that they don't 'really' understand what they are saying about a system when they apply such techniques. Students in these fields are repeated exhorted to be cautious about application of statistical techniques to their data without a clear understanding of the conditions required for those (...) techniques to yield genuine insight into the data. They are acutely aware that anyone can chuck some data into a software package and get what looks like an answer, even though these tests may not be well-defined for the data on which they can apparently be run. This is thus a healthy skepticism for uptake of causal modeling methods, which points directly to the need for a metaphysical understanding of causation in order to successfully use the modeling methods. Without a clear understanding of what the methods are committing to, including what it means to say that there exists a causal relationship, researchers have limited ability to identify potentially bad output, or to independently verify results. (shrink)

Though common sense says that causes must temporally precede their effects, the hugely influential interventionist account of causation makes no reference to temporal precedence. Does common sense lead us astray? In this paper, I evaluate the power of the commonsense assumption from within the interventionist approach to causal modeling. I first argue that if causes temporally precede their effects, then one need not consider the outcomes of interventions in order to infer causal relevance, and that one can instead (...) use temporal and probabilistic information to infer exactly when X is causally relevant to Y in each of the senses captured by Woodward’s interventionist treatment. Then, I consider the upshot of these findings for causal decision theory, and argue that the commonsense assumption is especially powerful when an agent seeks to determine whether so-called “dominance reasoning” is applicable. (shrink)

Nancy Cartwright offers an account of causal powers, and argues that it explains some important general features of scientific method. Patricia Cheng argues that this theory is superior as a psychological theory of learning to standard models of conditioning. I extend and develop the theory, and argue that it provides the best explanation of a number of problem cases for philosophical theories of causation, including preemption, overdetermination and puzzles about transitivity. Hitchcock and Halpern & Pearl on ‘actual causes’ Problems (...) and morals 2.1 Puzzles about prevention 2.2 Counterfactuals Causal powers 3.1 Generative causal power 3.2 Preventative causal power Net and component powers ‘Actual’ or ‘successful’ causes Solutions to puzzle cases Conclusion. (shrink)

In this paper, I want to substantiate three related claims regarding causal discovery from non-experimental data. Firstly, in scientific practice, the problem of ignorance is ubiquitous, persistent, and far-reaching. Intuitively, the problem of ignorance bears upon the following situation. A set of random variables V is studied but only partly tested for (conditional) independencies; i.e. for some variables A and B it is not known whether they are (conditionally) independent. Secondly, Judea Pearl’s most meritorious and influential algorithm for (...) class='Hi'>causal discovery (the IC algorithm) cannot be applied in cases of ignorance. It presupposes that a full list of (conditional) independence relations is on hand and it would lead to unsatisfactory results when applied to partial lists. Finally, the problem of ignorance is successfully treated by means of ALIC, the adaptive logic for causal discovery presented in this paper. (shrink)

Current psychological theories of human causal learning and judgment focus primarily on long-run predictions: two by estimating parameters of a causal Bayes nets, and a third through structural learning. This paper focuses on people’s short-run behavior by examining dynamical versions of these three theories, and comparing their predictions to a real-world dataset.