Thought experiments are basically devices of the imagination. They areemployed for various purposes such an entertainment, education,conceptual analysis, exploration, hypothesizing, theory selection,theory implementation, etc. Some applications are more controversialthan others. Few would object to thought experiments that serve toillustrate complex states of affairs, or those that are used ineducational contexts. The situation is different, however, withrespect to the appropriation of imagined scenarios to investigatereality (very broadly conceived to include things like electrons,tables, rain, beliefs, morals, people, numbers, universes, and evendivine beings). It is this use of thought experiments that attractsmost of the attention inside and outside of philosophical discourse.Significant is the overlap here with many other central philosophicaltopics, such as the nature of the imagination, the importance ofunderstanding in contrast to explanation, the role of intuition inhuman cognition, and the relationship between fiction and truth.Moreover, thought experiments are interdisciplinary in two importantrespects. Firstly, not only philosophers study them as a researchtopic, but also historians, cognitive scientists, psychologists, etc.Secondly, they are used in many disciplines, including biology,economics, history, mathematics, philosophy, and physics (although,interestingly, not with the same frequency in each).

Most often thought experiments are communicated in narrative form,frequently with diagrams. It is important to distinguish between theimagined scenarios that are featured in thought experiments, on theone hand, and the narratives that establish those scenarios inpeople’s mind, on the other. Once a scenario is imagined it mayassume a life on its own, and this explains partly the creative powerof a good thought experiment. Experimental results may obtain thatactually run counter to the narrative that initiated the discussion ofan imagined scenario. Besides, thought experiments should bedistinguished from thinking about experiments, from merely imaginingany experiments to be conducted outside the imagination, and frompsychological experiments with thoughts, though there may be someoverlap. They should also be distinguished from counterfactualreasoning in general, as they seem to require a palpable element,which explains the impression that something is experienced in athought experiment (i.e., being seen, felt, heard, etc.; notliterally, of course). In other words, though many call anycounterfactual or hypothetical situation a thought experiment (see,e.g., Rescher 1991), this appears too encompassing.

It is a quite different matter as to whether there is a logicalstructure common to all of thought experiments. Based on suchconsiderations of logical structure, a taxonomy has been proposedaccording to which all thought experiments fall into two classes:“Necessity Refuters” and “PossibilityRefuters” (see Sorensen 1992, 132–160). Such proposalsespecially fuel the debate about identity conditions of thoughtexperiments. What modifications to logical structure does a thoughtexperiment tolerate before it ceases to exist and a new one is born?In other words, how much emphasis on propositional characteristics isappropriate in the analysis of thought experiments?

Looking at the development of the discussion about thought experimentsover the past thirty years, it is fair to say that thought experimentswere primarily an important topic in the philosophy of science and thephilosophy of philosophy (“metaphilosophy”), before thescope widened up at a later point. There is a simple reason for thatpath. At the core of the discussion sits a relatively simpleepistemological challenge that is presented in a particularly powerfulmanner by numerous thought experiments that the history of science hasto offer. They suggest that we can learn about the real world byvirtue of merely thinking about imagined scenarios. But how can welearn about reality (if we can at all), just by thinking? This is thecentral question. Are there really thought experiments that enable usto acquire new knowledge about nature without new empirical data? Ifso, where does the new information come from, assuming that it takesnew information to learn anything new about the world by means ofthought experiments? Finally, how can we distinguish good from badinstances of thought experiments? These questions seem urgent withrespect to scientific thought experiments, because many“recognize them as an occasionally potent tool for increasingour understanding of nature” (Kuhn 1977, p. 241).“Historically their role is very close to the double one playedby actual laboratory experiments and observations. First, thoughtexperiments can disclose nature’s failure to conform to apreviously held set of expectations. Second, they can suggestparticular ways in which both expectation and theory must henceforthbe revised” (Kuhn 1977, p. 261). Yet, questions surrounding theepistemological challenge that certain scientific thought experimentspose, are equally urgent with respect to thought experiments outsideof the natural sciences. This is especially true with respect tophilosophy itself. Philosophy offers numerous examples of thoughtexperiments that play a role similar in importance to some scientificthought experiments. And this fact provokes in turn further inquiriesinto the relationship between the natural sciences and philosophy,especially with respect to phenomena that implicate both the naturalsciences and philosophy, such as the mind and free will (see, e.g.,Wilkes 1988; Young 2013).

If scientific practice has room for thought experiments, then thequestion arises as to why we would want philosophical methodology tobe more discriminatory in this respect. One reason that is oftenoffered is that results of scientific thought experiments may besubjected to further empirical testing. Obviously, this can’t bedone for philosophical thought experiments. But, it seems difficult toaccept a categorical separation of science and philosophy along theselines. The 17th century saw some of the most brilliant practitionersof thought experimentation in Galileo, Descartes, Newton, and Leibniz,all of whom pursued the project of “natural philosophy.”And in our own time, the creation of quantum mechanics and relativityare almost unthinkable without the crucial role played by thoughtexperiments, most of which relate to important philosophical issuesthat arise from these scientific theories. Besides, much of ethics,philosophy of language, and philosophy of mind is based on the resultsof thought experiments in a way that seems very similar to scientificthought experiments (though some might contest this), includingSearle’s Chinese room, Putnam’s twin earth, andJackson’s Mary the colour scientist. Philosophy, even more thanthe sciences, would be severely impoverished without thoughtexperiments. These observations partly explain why it has been arguedthat a more “unified” account of thought experiments isdesirable (see Boniolo 1997; Cooper 2005, pp. 329–330;Gähde 2000). Of course, it is important not to downplay thesignificant differences between the sciences and philosophy. But anaccount of thought experiments seems more powerful if it can dojustice to the fact that not only in the sciences we find many ofthem.

There have been several attempts to define “thoughtexperiment” along the lines of traditional conceptual analysis(see, e.g., Picha 2011; McComb 2013), but likely it will be better toleave the term loosely characterized, so as not to prejudice theongoing investigation. Of course, we need to have some idea as to whatthought experiments are to guide a proper philosophical analysis (seeHäggqvist 2009), but this does not mean that we need to beginwith a technical definition, specifying necessary and sufficientconditions. In fact, many of the most important concepts we deal withremain rather loosely defined when philosophical inquiry begins, e.g.,religion or democracy. Luckily, there are plenty of examples to referto in order to circumscribe our subject matter well enough. As well asthose already mentioned, there are Newton’s bucket,Heisenberg’s gamma-ray microscope, Einstein’s elevator,Leibniz’s mill, Parfit’s people who split like amoebas,and Thomson’s violinist. Everyone is probably familiar with someof these. Less familiar thought experiments include the mouse thatbreaks into the tabernacle of a medieval Roman Catholic Churchbuilding to feed on the consecrated wafers kept in there (see Fehige2018). Roman Catholic Christians believe that a consecrated wafer isthe “body of Christ”. The “substance” of thewafer, understood in terms of Aristotelian categories, is believed tobe replaced. In its place is the “substance” ofChrist’s body after consecration by a priest. Only theAristotelian “accidents” of the wafer remain intact(smell, colour, texture, etc.). Does the mouse eat the “body ofChrist” (if any human actually does)? If not, then the“body of Christ” seems to be less than an objectivereality; if yes, the “body of Christ” must be able to dogood in the absence of a believing human soul. Another example lessknown is “the dome” thought experiment, which is to proveindeterminism in Newtonian physics. Imagine a mass sitting on aradially symmetric surface in a gravitational field. Guided byNewton’s laws of motion one comes to realize that the mass caneither remain at rest for all times, or spontaneously move in anarbitrary direction (see Norton 2008). This thought experimenttriggers a number of very interesting questions concerning the natureof Newtonian theory, the meaning of “physical”, and therole of idealizations in physics. And, of course, does it show what itclaims? (see Malament 2008).

This entry continues with an overview of the characteristics ofthought experiments in light of examples in Section 1. Section 2reviews several taxonomies for classfying thought experiments andSection 3 sketches a history of philosophical inquiry into the natureof thought experiments. Section 4 covers several views representingthe current state of the debate. The entry concludes by highlightingsome trends in discussions surrounding the so-called laboratory of themind.

• 1. Important Characteristics of Thought Experiments
• 2. Taxonomies of Thought Experiments
• 3. The History of Thought Experiments
• 4. Current Views on Thought Experiments
  • 4.1 The Skeptical Objection
  • 4.2 The Intuition–Based Account
  • 4.3 The Argument View
  • 4.4 Conceptual Constructivism
  • 4.5 Experimentalism
  • 4.6 The Mental–Model Account
• 5. Going Forward
• Bibliography
• Academic Tools
• Other Internet Resources
• Related Entries

1. Important Characteristics of Thought Experiments

Theorizing about thought experiments usually turns on the details orthe patterns of specific cases. Familiarity with a wide range ofexamples is crucial for commentators, and the list is very long (see,e.g., Stuart et al. 2018, pp. 558–560) We will provide a fewhere. One of the most beautiful early instances (found in Lucretius,De Rerum Natura 1.951–987; see Bailey 1950, pp.58–59) attempts to show that space is infinite: if there is apurported boundary to the universe, we can toss a spear at it. If thespear flies through, it isn’t a boundary after all; if the spearbounces back, then there must be something beyond the supposed edge ofspace, a cosmic wall that stopped the spear, a wall that is itself inspace. Either way, there is no edge of the universe; thus, space mustbe infinite.

This example nicely illustrates many of the most common features ofwhat it means to engage in the conduct of thought experiments: wevisualize some situation that we have set up in the imagination; welet it run or we carry out an operation; we see what happens; finally,we draw a conclusion. The example also illustrates the fallibility ofthought experiments. Since the time of Lucretius, we’ve learnedhow to conceptualize space so that it could be both finite andunbounded. Imagine a circle, which is a one-dimensional space. As wemove around, there is no edge, but it is nevertheless finite. Theuniverse might be a three-dimensional version of this topology. It is,therefore, true that we must try to be mindful of unexpectedlimitations due to “physical scale effects” (Klee 2008),or other such things, when imagining counterfactual scenarios.

Figure 1. “Welcome to the edge ofthe Universe”

Often a real experiment that is meant to be the analogue of a thoughtexperiment is impossible to be carried out as such due to physical,technological, ethical, or financial limitations (see, e.g., Sorensen1992, pp. 200–202); but physical unrealizability needn’tbe a defining condition of thought experiments. Rather, the main pointis that we seem able to get a grip on nature just by thinking, andtherein lies the great interest for philosophy. That was the positionof Ernst Mach (see Mach, 1897 and 1905; for a most instructiveassessments of his views see Kühne 2006, pp. 165–202, andSorensen 1992, pp. 51–75). Thought experiments are on a spectrumof different kinds of experiments. They allow us to tap into a greatstore of “instinctive knowledge” picked up from pastexperience. We will get back to Mach’s theory further down. Hisaccount of thought experiments remains one of the major theories ofhow thought experiments work. One of Mach’s favourite examplesis due to Simon Stevin (see Mach, 1883, pp. 48–58). When a chainis draped over a double frictionless plane, as in Fig. 2a, how will itmove? Add some links as in Fig. 2b. Now it is obvious. The initialsetup must have been in static equilibrium. Otherwise, we would have aperpetual motion machine; and according to our experience-based“instinctive knowledge,” says Mach, this is impossible. Wedo not have to perform the experiment in the real world, which wecould not do, anyway, since it would require a perfectly frictionlessplane. Nevertheless the outcome seems compelling.

(a)

(b)

Figure 2. “How will itmove?”

Judith Thomson provided one of the most striking and effective thoughtexperiments in the moral realm (see Thomson 1971). Her example isaimed at a popular anti-abortion argument that goes something likethis: A fetus is an innocent person. All innocent persons have a rightto life. Abortion results in the death of a fetus. Therefore, abortionis morally wrong. In her thought experiment, Thomson asks you toimagine a famous violinist falling into a coma. The society of musiclovers determines from medical records that you and you alone can savethe violinist’s life by being hooked up to him for nine months.The music lovers break into your home while you are asleep and hookthe unconscious (and unknowing, hence innocent) violinist to you. Youmay want to unhook him, but you are then faced with the followingargument put forward by the music lovers: The violinist is an innocentperson. All innocent persons have a right to life. Unhooking him willresult in his death. Therefore, unhooking him is morally wrong.However, the argument, even though it has the same structure as theanti-abortion argument, does not seem convincing in this case. Youwould be very generous to remain attached for nine months, but you arenot morally obligated to do so. The parallel with the abortion case isevident. Thomson’s thought experiment is effective indistinguishing two concepts that had previously been run together:“right to life” and “right to what is needed tosustain life.” The fetus and the violinist might each have theformer, but it is not evident that either has the latter. The upshotis that even if the fetus has a right to life (which Thomson does notbelieve but allows for the sake of the argument), it may still bemorally permissible to abort. Those opposed to Thomson’s viewhave two options. They can either dismiss her thought experiment as auseless fiction. In fact, thought experiments as a method in ethicshave their critics (see, e.g., Dancy 1985). Alternatively, they canprovide a different version of the same scenario to challenge theconclusion. It is a very intriguing feature of thought experimentsthat they can be “rethought” (see Bokulich 2001). Realexperiments are frequently open to reinterpretation, too. In thisrespect there does not seem to be a principled difference between thetwo classes of experiments.

Like arguments, thought experiments can be criticized in differentways. Perhaps the set up is faulty; perhaps the conclusions drawn fromthe thought experiment are not justified. Similar criticisms can arisein real experiments. Counter thought experiments are perhaps anotherform of criticism. They do not target the premises or conclusionsinvolved in a particular thought experiment but question thephenomenon, i.e. the non-propositional heart of an imagined scenario(see Brown 2007). For example, Daniel Dennett is convinced that FrankJackson’s Mary thought experiment is poor evidence to opposephysicalism in philosophy of mind. In Jackson’s version, Mary,who knows everything physics and the neurosciences can possibly knowabout colours but grew up in a colourless environment (seeing onlyblack, white and grey things), allegedly learns something new when shesees a red tomato for the first time. Now she knows what it is like toexperience red. This is an argument for qualia as something over andabove the physical. Instead of a red tomato, Dennett, in his versionof the thought experiment, presents Mary with a bright blue banana. Inhis version of the story (which seems just as plausible asJackson’s), Mary balks and says she is being tricked, since sheknows that bananas are yellow, and this, says Mary, is a consequenceof knowing everything physical about colour perception. Mary does notlearn anything new when she sees coloured objects for the first time,so there is no case against physicalism after all. Jackson’sinitial thought experiment was very persuasive, but Dennett’sseems equally so, thus, undermining Jackson’s argument, althoughthere is greater resistence to the conclusion of the latter than theformer! Dennett complains a great deal about the ongoing“Mariology”, as he calls the continuing acceptance ofJackson’s thought experiment as a poweful case againstphysicalism.

Clearly, thought experiments are characterized by an intriguingplasticity, and this raises the interesting question of what it isthat preserves the identity of a thought experiment. Replacing a redtomato with a blue banana might still leave us with the same thoughtexperiment––slightly revised. But, at what point do we geta new thought experiment? This is not merely a question aboutconceptual vagueness. It helps to facilitate a discussion of theintuitively most plausible view about the cognitive efficacy ofthought experiments, according to which this power depends on theirbeing arguments, in a fairly strict sense of argument. John D. Nortonholds such a view, which will be discussed below. In light of caseswhere the discussion of one and the same thought experiment played animportant role in settling a dispute, the following problem arises:how can one and the same thought experiment support opposing viewsabout a particular matter if the arguments that correspond to thedifferent versions of the thought experiment that were entertained bythe disputing parties are significantly different? The dilemma is: wecould say that if there is more than one argument then there is morethan one thought experiment involved in the dispute. But if that istrue then the disputing parties simply talked past each other. Oneparty presented an argument that the other party ignored whilepresenting their own. Alternatively, we can say that one thoughtexperiment can correspond to many different arguments. But, if that istrue then it becomes unclear in what non-trivial sense thoughtexperiments are supposed to be identical with arguments (see Bishop1999, and the response by Norton 2004, 63–64).

The plasticity of thought experiments coheres with another feature ofthought experiments, namely that they seem to have “evidentialsignificance only historically and locally, i.e., when and wherepremises that attribute evidential significance to it […] areendorsed” (McAllister 1996, p. 248).

2. Taxonomies of Thought Experiments

Many taxonomies can be found in the literature. They are not mutuallyexclusive. We will present three of them. The first follows the typeof purpose thought experiments serve. A very rudimentary version of itcan be found in Mach (1897 and 1905). Such a classification makessense, because an “imaginary experiment should be judged on itsspecific purpose” (Krimsky 1973, p. 331). Thought experimentsare conducted for diverse reasons (see, e.g., DeMay 2006; Sorensen1992, pp. 7–15), and this in a variety of areas, includingeconomics (see, e.g., Herfeld 2019; Thoma 2016), education (Helm andGilbert 1985; Helm et al. 1985, Klassen 2006; Sriraman 2006; Stonier1990), history (see, e.g., Maar 2014; Reiss 2009), literature (see,e.g., Davies 2007; Elgin 2004), mathematics (see, e.g. Brown 1991[2011], pp. 90–97; Glas 1999), morality (see, e.g., Hauerwas1996; Wilson 2016), as well as the natural sciences (see Krimsky1973), the socio-political realm (see, e.g. Roberts 1993: Thaler2016), and theology (see, e.g., Gregersen 2014; Fehige 2024). Thoughtexperiments may be used to entertain. This is probably true of shortstories or novels which some argue qualify as thought experiments ifcertain conditions apply (see, e.g., Davenport 1983). Some thoughtexperiments fulfil a specific function within a theory (see Borsboomet al. 2002). Others are executed because it is impossible to run theexperimental scenario in the real world (see, e.g., Sorensen 1992, pp.200–202). Sometimes thought experiments help to illustrate andclarify very abstract states of affairs, thereby accelerating theprocess of understanding (see Behmel 2001). Again others serve asexamples in conceptual analysis (see Cohnitz 2006). And, then thereare those that matter in the process of theory discovery (Praem andSteglich–Peterson 2015). The thought experiments that havereceived most of the attention are taken to provide evidence for oragainst a theory, putting them on a par with real-world experiments(see, e.g., Gendler 2004). The different ways to use thoughtexperiments, of course, do not exclude one another. Most obviously,for example, a thought experiment can both entertain and make a caseagainst a theory.

A second taxonomy classifies thought experiments in terms of theirlogical structure (see Sorensen 1992, pp. 132–166). The idea isto divide all thought experiments into two types of “alethicrefuters”: “Although there are a number of ways toclassify thought experiments, a refutation format scores the mostpoints when judged by familiarity, specificity, and simplicity.According to this scheme, thought experiments aim at overturningstatements by disproving one of their modal consequences. Modalitiesare operators that are applied to propositions to yield newpropositions. There are deontic modalities (permissible,forbidden), epistemic modalities (know, believe), andalethic modalities (possible, necessary). The alethicmodalities are the best–known and more–basic modality.Hence, we won’t miss anything by concentrating on them”(Sorensen 1992, p. 135). One type of thought experiment “isdesigned to refute a statement by showing that something ruled out asimpossible by that statement is really possible after all”(Sorensen 1992, p. 135). The most discussed examples in themetaphilosophical discussion on thought experiments is of such a type,namely the Gettier scenarios (see Grundmann & Horvarth 2014;Saint-Germier 2019). They are designed to refute the claim that allknowledge is justified, true belief. They serve as a “necessityrefuter.” The other type collects examples of “possibilityrefuters”. They don’t affirm “the possibility of thethought experiment’s content”. Instead, they establish“copossibilities”. A wonderful example is the scenario ofan omnipotent God who faces the task of creating a stone too heavy forthat God to lift. It seems God cannot succeed. The notion of divineomnipotence causes some headache here.

A third taxonomy (see Brown 1991, chapter 2), which has not goneunchallenged (see Norton 1993b), is more limited than the first twoinsofar as it focuses largely on the class of those thoughtexperiments that are taken to function in theory choice, which is theuse of thought experiments that has been receiving most of theattention. According to this taxonomy, the main division isconstructive vs. destructive and resembles KarlPopper’s distinction betweenapologetic andcritical thought experiments. Popper actually distinguishesbetween three types of thought experiments:heuristic (toillustrate a theory),critical (against a theory) andapologetic (in favour of a theory) (see Popper 1959). Hiscase in favour of a critical and against an apologetic use of thoughtexperiments is very limited. He focuses exclusively on quantum physicsand doesn’t really say much to address the primaryepistemological challenge presented by the success of critical thoughtexperiments.

Amongdestructive thought experiments, the following subtypescan be identified: the simplest of these is to draw out acontradiction in a theory, thereby refuting it. The first part ofGalileo’s famous falling bodies example does this. It shows thatin Aristotle’s account, a composite body (cannon ball and musketball attached) would have to fall both faster and slower than thecannon ball alone. A second subtype is constituted by those thoughtexperiments that aim to show that the theory in question is inconflict with other beliefs that we hold. Schrödinger’swell-known cat paradox, for instance, does not show that quantumtheory (at least on some interpretations) is internally inconsistent(see Schrödinger 1935, p. 812; translation: Trimmer 1980, p.328): “A cat is penned up in a steel chamber, along with thefollowing diabolical device (which must be secured against directinterference by the cat): in a Geiger counter there is a tiny bit ofradioactive substance, so small, that perhaps in the course of onehour one of the atoms decays, but also, with equal probability,perhaps none; if it happens, the counter tube discharges and through arelay releases a hammer which shatters a small flask of hydrocyanicacid. If one has left this entire system to itself for an hour, onewould say that the cat still lives if meanwhile no atom has decayed.The first atomic decay would have poisoned it. The q-function of theentire system would express this by having in it the living and thedead cat (pardon the expression) mixed or smeared out in equalparts.” This thought experiment shows that quantum theory (asinterpreted by Bohr) is in conflict with some very powerful commonsense beliefs we have about macro-sized objects such ascats––they cannot be both dead and alive in any sensewhatsoever. The bizarreness of superpositions in the atomic world isworrisome enough, says Schrödinger, but when it implies that samebizarreness at an everyday level, it is intolerable. There is a thirdsubtype of negative thought experiments, namely when, in effect, acentral assumption or premise of the thought experiment itself isundermined. For example, as we have seen above, Thomson showed withher thought experiment that “right to life” and“right to what is needed to sustain life” had been runtogether. When distinguished, the argument against abortion isnegatively affected.

A fourth sub-type of negative thought experiments are “counterthought experiments” (see Brown 2007). Norton very usefullyintroduces a related idea:“thought-experiment/anti-thought-experiment pairs” (seeNorton 2004, pp. 45–49). Above, we have already encountered thissubtype in our discussion of Lucretius’ spear-thoughtexperiment, and with Dennett’s reply to Jackson’s muchdiscussed Mary the colour scientist thought experiment. Here we wouldlike to add one more example, namely Mach’s counter thoughtexperiment against absolute space. In hisPrincipiaMathematica, Newton offers a pair of thought experiments asevidence for absolute space. One is the bucket thought experiment withwater climbing the wall (see Fig. 3), the other is about a pair ofspheres joined by a cord that maintained its tension in otherwiseempty space (see Fig. 4). The explanation for these phenomena, arguesNewton, is absolute space: the bucket and the joined spheres arerotating with respect to space itself. In response, Mach modifies thescenario and argues, contra Newton, that the two spheres would movetoward one another thanks to the tension in the cord, and if werotated a very thick, massive ring around a stationary bucket, wewould see the water climb the bucket wall. (For further discussion ofMach’s counter thought experiment to Newton’s seeKühne 2006, pp. 191–202). In short, the point ofMach’s counter thought experiments is to describe the phenomenaof the thought experiments’ scenarios differently, that is, todeclare that different things would happen. Mach’s counterthought experiment undermines our confidence in Newton’s thoughtexperiments. Absolute space might be a plausible explanation of thephenomena in Newton’s thought experiments, but now, in light ofMach’s counter thought experiment, we’re not so sure ofthe phenomena itself and thus of the idea of absolute space.

Figure 3. Stages in the bucketexperiment

Figure 4. Two spheres held by a cord inotherwise empty space

To be effective, counter thought experiments needn’t be veryplausible at all. In a court of law a jury would convict providedguilt is established “beyond a reasonable doubt.” A commondefence strategy is to provide an alternative account of the evidencethat has just enough plausibility to put the prosecution’s caseinto some measure of doubt. That is sufficient to undermine it. Acounter thought experiment need only do that much to be effective, andin this sense it operates like a “necessity refuter” inSorensen’s sense.

In addition to destructive ones, there is a second type, theconstructive thought experiments. Unsurprisingly, there aremany ways they could provide positive support for a theory. One ofthese is to provide a kind of illustration that makes a theory’sclaims clear and evident. In such cases thought experiments serve as akind of heuristic aid. A result may already be well established, butthe thought experiment can lead to a very satisfying sense ofunderstanding. In hisPrincipia Mathematica, Newton providesa wonderful example showing how the moon is kept in its orbit in justthe same way as an object falls to the earth (see Ducheyne 2006, pp.435–437). He illustrates this by means of a cannon shooting acannon ball further and further (see Fig. 5). In the limit, the earthcurves away as fast as the ball falls, with the eventual result beingthat the cannon ball will return to the spot where it was fired, and,if not impeded, will go around again and again. This is what the moonis doing. We could arrive at the same conclusion through calculation.But Newton’s thought experiment provides that elusive sense ofunderstanding. It’s a wonderful example of the “ahaeffect” that is typical of many powerful thoughtexperiments.

Figure 5. “The shot heard aroundthe world”

Thomson’s violinist showed that abortion could be morallypermissible even when the fetus has a right to life. Similarly,Einstein’s elevator showed that light will bend in agravitational field, because according to the principle ofequivalence, there is no difference between such a frame of referenceand one that is accelerating in free space; the laws of physics arethe same in all. Suppose then, an observer is inside an elevatorsealed off from the outside so that the observer cannot tell whetherhe is in a gravitational field or accelerating. If it wereaccelerating, and if a light beam were to enter one side, then, due tothe elevator’s motion, the beam would appear to drop or curvedown as it crossed the elevator. Consequently, it would have to do thesame thing if the elevator was in a gravitational field. Therefore,gravity ‘bends’ light.

Maxwell’s demon showed that entropy could be decreased: Thesecond law of thermodynamics implies that heat won’t pass from acold body to a hot one. In classical thermodynamics this law is quitestrict; but in Maxwell’s kinetic theory of heat there is aprobability, though extremely small, of such an event happening. Somethought this a reductio ad absurdum of Maxwell’s theory. To showhow it is possible to violate the second law, Maxwell imagined a tinycreature who controls a door between two chambers. Fast molecules fromthe cold box are let into the hot box, and slow molecules from the hotare allowed into the cold. Thus, there will be an increase in theaverage speed in the hot box and a decrease in the average speed ofmolecules in the cold. Since, on Maxwell’s theory, heat is justthe average speed of the molecules, there has been a flow of heat froma cold body to a hot one.

Parfit’s splitting persons shows that survival is a moreimportant notion than identity when considering personhood (for acritical discussion see Gendler 2002a). We say they “show”such and such, but, “purport to show” might be better,since some of these thought experiments are quite contentious. Whatthey have in common is that they aim to establish something positive.Unlike destructive thought experiments, they are not trying todemolish an existing theory, though they may do that in passing. Torepeat an important point: in principle, given the fact that thoughtexperiments can be rethought (see Bokulich 2001), and that theevidential significance is dependent on historical and localaccomplishments (see McAllister 1996), it cannot be irrelevant toidentify the intention of the thought experimenter, if one wants todetermine the type of a thought experiment: “An imaginaryexperiment should be judged on its specific purpose” (Krimsky1973, p. 331).

3. The History of Thought Experiments

The practice of thought experiments is not an invention of modernscience. That fact may be obscured by the dominance of scientificexamples in the lively discussions about thouht experiments today. ThePre-Socratics “invented thought experimentation as a cognitiveprocedure and practiced it with great dedication andversatility” (Rescher 2005, p. 2). “There is no ancientGreek term corresponding to what we nowadays refer to as a thoughtexperiment, and presumably ancient philosophers did not have ourmodern notion of a thought experiment. But there is no doubt that theydid use thought experiments. In fact, they often employed them in wayssimilar to those of contemporary philosophers, that is, both fordefending their own theories as well as for refuting the theories oftheir opponents ” (Ierodiakonou 2018, p. 31). (See also Becker2018; Diamond 2002, pp. 229–232; Fuhrer 2009; Glas 1999;Ierodiakonou 2005; Ierodiakonou and Roux (eds.) 2011; Irvine 1991;Rescher 1991 and 2005, pp. 61–72). The situation is similar withrespect to medieval natural philosophy, although there are furthernuances to be considered (see King 1991). According to Edward Grant,during the late Middle Ages “the imagination became a formidableinstrument in natural philosophy and theology in ways that would haveastonished ancient Greek natural philosophers, especiallyAristotle” (Grant 2007, p. 201). But this doesn’t meanthat we have reason to think of Aristotle as an opponent of theconduct of thought experimentstout court. On the contrary,“Aristotle uses thought experiments for argumentative persuasionand in places where, due to the obscure nature of the subject matteror the counterintuitive nature of the thesis they are meant tosupport, insight cannot be readily communicated by appeal toobservational facts” (Corcilius 2018, p. 73). With a fewexceptions that involved problems of motion, “thescholastics” of the medieval period made no meaningful effort totransform their hypothetical conclusions into specific knowledge aboutthe physical world. They did, however, assume that although thesehypothetical conclusions were naturally impossible, God could producethem supernaturally if he wished. Special attention received also aclass of medieval thought experiments that does not rely oncounterfactuals but depends on theological assumptions to studymatters non-theological, namely those thought experiments involvingangels, whose existence were affirmed at that time (see Perler 2008).Angels are gone by now (see Clark 1992), but not thought experiments.While most thought experiments involving angels have Christianity astheir context, there is evidence of the practice of thoughtexperiments also in the context of Islam and Judaism (see McGinnis2018; Fisch 2019). In fact, the case has been made “that IbnSina is the first philosopher in the Aristotelian tradition, and thusperhaps the first in Western philosophy overall, to try to identifythe psychological processes that go into postulating a hypotheticalscenario. Ibn Sina also exhibits an interest in accounting for why,and to what extent, such psychological acts are thought to carryweight in our study of nature” (Kukkonen 2014, p. 434).

Ernst Mach is commonly credited with introducing the word“thought experiment” (Gadankenexperiment) andthereby coining a term for philosophical discussion (recently done,for instance, by Krauthausen 2015, p. 15). “ This view isincorrect, however! […] it can be substantiated that it wasused […] already in 1811” (Witt-Hansen 1976, p. 48; seealso Buzzoni 2008, pp. 14–15; 61–65; Kühne 2005, pp.92–224; Moue et al. 2006, p. 63). The conceptual history of“thought experiment” goes back at least to the Danish“Tankeexperiment,” as it was used by Hans-ChristianØrsted. We can go back even further and find in the work of theGerman philosopher-scientist Georg Lichtenberg (1742–1799) atacit theory of “experiments with thoughts and ideas.”These experiments help to overcome habits of thought that can inhibitscientific progress, and make possible an enlightened philosophy (seeSchildknecht 1990, pp. 21; 123–169; Schöne 1982).Lichtenberg’s “aphoristic experiments” (see Stern1963, pp. 112–126) reflect “that Lichtenberg’sscientific preoccupations are the formal and thematic prolegomena tohis work as a literary artist” (Stern 1963, p. 126).Lichtenberg’s reflections on thought experimentation resemblethose of Popper and Thomas S. Kuhn, and it is plausible to think ofhim as one important figure of the very first period in the history ofphilosophical inquiry into thought experiments (see Fehige and Stuart2014).

Accordingly, the modern history of the philosophical investigationinto thought experiments can be divided into four stages: in the 18thand 19th century the awareness of the importance of thoughtexperiments in philosophy and science emerges. In addition toLichtenberg and Hans-Christian Ørsted, special mention shouldbe made of Novalis (see Daiber 2001). The topic reemerges in a moresystematic manner at the beginning of the 20th century with littlerelation to the attempts made at the first stage. The stakeholders ofthe second stage were Pierre Duhem, Mach, and Alexius Meinong (seeDuhem 1913, pp. 304–311; Mach, 1883, pp. 48–58, 1897 and1905; Meinong 1907). A third stage, probably due to the rediscovery ofthe importance of scientific practice for a proper understanding ofscience, followed in the first part of the second half of the 20thcentury. Again, the contributions of this stage bear little relationto the two previous stages. While the third period has seen a numberof noteworthy contributions (Cole 1983; Dancy 1985; Dennett 1985;Fodor 1964; Helm and Gilbert 1985; Helm et al. 1985; Krimsky 1973;McMullin 1985; Myers 1986; Poser 1984; Prudovsky 1989; Rehder 1980a,b;Yourgrau 1962 and 1967), the protagonists of this period wereAlexandre Koyré, Kuhn and Popper. The ongoing philosophicalexploration of thought experiments began in the 1980s, and marks thefourth stage. Arguably, it has been the most prolific one of all fourstages. With some very important sign-postings in place (Horowitz andMassey (eds.) 1991; Sorensen 1992; Wilkes 1988), the ongoingdiscussion took off in light of a debate between James Robert Brownand John D. Norton (see for a concise statement of each position Brown2004 and Norton 2004), which many have found useful to establish acontrast with their own alternative accounts of thought experiments.These views “represent the extremes of platonic rationalism andclassic empiricism, respectively” (Moue et al. 2006, p. 69).They will be described below.

4. Current Views on Thought Experiments

At this point it is important to recall the key epistemologicalchallenge described in the introduction: how can we learn about thereal world through merely thinking about imagined scenarios? Thischallenge sits at the center of the discussion about thoughtexperiments even though we must note that not all of the workdiscussed below focuses on it directly. Still, this section describessix views that can be seen as responding in some way to thischallenge: The Skeptical Objection, The Intuition-Based Account, TheArgument View, Conceptual Constructivism, Experimentalism, and TheMental-Model Account.

4.1 The Skeptical Objection

Of course, particular thought experiments have been contested. But forthe most part, the practice of thought experiments in the sciences hasbeen cheerfully accepted. Pierre Duhem, the great historian ofphysics, is almost alone in what has been understood as an outrightcondemnation of scientific thought experiments (see Duhem 1913, pp.304–311). A thought experiment is no substitute for a realexperiment, he claimed, and should be forbidden in science, includingscience education. However, in view of the important role of actualthought experiments in the history of physics — fromGalileo’s falling bodies, to Newton’s bucket, toEinstein’s elevator — it is unlikely that anyone will feelor should feel much sympathy for Duhem’s strictures. We hastento add that Buzzoni (2018) questions the validity of this reading ofDuhem, and argues that already Mach’s reception of Duhem’sviews suggests a more nuanced reading of Duhem’s position.

Philosophers can be as critical as Duhem when it comes to thoughtexperimenting in their own field (see Peijnenburg; Atkinson 2003;Thagard 2014; Wilson 2016). At least thought experiments in science,the skeptic claims, can be tested by physical experiment. However,this is clearly false, since frictionless planes and universes emptyof all material bodies cannot be produced in any laboratory. True, theresults of philosophical thought experiments cannot be evenapproximately tested. But, skeptics say little about why thoughtexperiments enjoy such popularity in philosophy. We are inclined tosay that skeptics underestimate the importance of thought experimentsfor the creative mind in any field. Also, one mustn’t forgetthat the cognitive power of real world experiments isn’t aself-evident matter either.

Few are outright skeptics, however. Many take a more ambiguous stance.Sören Häggqvist, for example, has developed a normativemodel for philosophical thought experiments (see Häggqvist 1996and 2009). Surprisingly, none of the commonly accepted philosophicalthought experiments satisfies his model. And the process ofidentifying successful thought experiments is only the first step inaddressing the central epistemological challenge posed by thoughtexperiments. It gets much messier once we begin to ask exactly howreliable “successful” thought experiments are. Granted,there is some justice in worrying about the reliability ofphilosophical thought experiments (see, e.g., Klee 2008). This mightbe true for ethics (see Dancy 1985, Jackson 1992; Wilson 2016),conceptual analysis (see Fodor 1964), and the philosophy of mind:“A popular strategy in philosophy is to construct a certain sortof thought experiment I call intuition pump. […] Intuitionpumps are often abused, though seldom deliberately” (Dennett1985, p. 12). The claim by Dennett and others is that thoughtexperiments too often rest on prejudice and faulty common sense; theyare inherently conservative, while real science will likely result inhighly-counterintuitive outcomes. Dennett believes that thoughtexperiments rest on naive “folk concepts,” which is whythey can be so misguided. It is far from clear that this is a faircharge. Everything involved in Galileo’s thought experiment thatproduced the principle of relativity could be called “folkconcepts.” If we are inside a ship and perform a number ofexperiments, such as walking about, tossing a ball, watching birds flyabout, we could not tell whether we are at rest in port or sailingover a smooth sea. The upshot is that nature behaves the same eitherway; the laws of nature are the same in any inertial frame. Thisresult is profound and is still with us in Einstein’srelativity, whether it is folk physics or not.

Frequently discussed is the skeptical challenge raised by KathleenWilkes. She expresses a deep suspicion of scenarios such as DerekParfit’s people splitting like an amoeba (see Parfit 1987;Gendler 2002a). Wilkes wants philosophy “to use science factrather than science fiction or fantasy” (Wilkes 1988, p. 1), andtherefore to refrain from using thought experiments because they are“both problematic and positively misleading” (Wilkes 1988,p. 2). She claims that thought experiments about personal identity inparticular often fail to provide the background conditions againstwhich the experiment is set (see Wilkes 1988, p. 7). She thinks wewould not know what to say if we encountered someone who split like anamoeba. She insists that a legitimate thought experiment must notviolate the known laws of nature. We do agree with Wilkes thatunderdetermination can be a problem. But instead of dismissing thoughtexperiments in philosophy we should consider it a crucial factor inassessing the quality of a thought experiment (see Rescher 2005, pp.9–14). That is to say that the more detailed the imaginaryscenario in the relevant aspects is, the better the thought experiment(see Brendel 2004, pp. 97–99; Häggqvist 1996, p. 28).

We also agree that the inferences drawn in thought experimenting arehighly problematic if the hypothetical scenario “is inadequatelydescribed” (Wilkes 1988, p. 8). But Wilkes seems to think thatthe lack of description is unavoidable, which supposedly amounts to areason against philosophical thought experiments on personal identitybecause persons are not natural kinds. This makes it impossible tofill in necessary information to make the thought experiment workgiven its unavoidable underdetermination. Wilkes thinks that“whenever we are examining the ranges of concepts that do notpick on natural kinds, the problem of deciding what is or what is not‘relevant’ to the success of the thought experiment is yetmore problematic than the same question as it arises in science; and,unlike the scientific problem, it may not even have an answer inprinciple” (Wilkes 1988, p. 15). She adds that scientific laws— especially those describing biological kinds like human beings— “are not disjoint and independent, detachable from oneanother […]. They are interrelated, to varying degrees ofcourse” (Wilkes 1988, p. 29). This implies, for example, that“a full psychophysiological account of the processes of humanperception must at some stage link up with part at least of linguisticability; for we typically see things under a certain description, andthat description may be a very sophisticated one” (Wilkes 1988,p. 29). These considerations have her rule out experiments thatchallenge the human monopoly of personhood. No thought experiment,claims Wilkes, is well conceived if it involves non-human animals orcomputers as persons. But also those thought experiments can be ruledout which involve the “fission or fusion of humans”because it is not theoretically possible. “The total impact ofthe sum of laws that group us together as human beings (a natural kindcategory) precludes our splitting into two […] or fusing withsomeone else” (Wilkes 1988, p. 36).

One can ascertain here all too well the inherent difficulties inthinking about personal identity and the limited benefit some thoughtexperiments might have for what is deemed the proper metaphysics ofpersonal identity. Nevertheless, good reasons have been given infavour of the use of thought experiments about personal identity (seeBeck 2006; Kolak 1993; Hershenov 2008). We also feel that the problemsabout thought experiments on personal identity reveal more about theintricate nature of the subject than about the usefulness ofphilosophical thought experiments. And, disregarding othershortcomings in Wilkes’ skepticism (for further discussion ofWilkes’ views see Beck 1992; Brooks 1994; Focquaert 2003;Häggqvist 1996, pp. 27–34), her suggestion that thoughtexperimental scenarios would have to satisfy current scientificknowledge about the relevant entities featured in a thought experimentis highly implausible. We learn a great deal about the world and ourtheories when we wonder, for instance, what would have happened afterthe big bang if the law of gravity had been an inverse cube lawinstead of an inverse square. Would stars have failed to form?Reasoning about such a scenario is perfectly coherent and veryinstructive, even though it violates a law of nature.

To some extent we should share Wilkes’s concern that thoughtexperimenting seems to be constrained only by relevant logicalimpossibilities and what seems intuitively acceptable. This is indeedproblematic because intuitions can be highly misleading and relevantlogical impossibilities are fairly ungrounded if they cannot besupplemented by relevant theoretical impossibilities based on currentscience in order to avoid the jump into futile fantasy. But in orderto dismiss thought experimenting as a useful philosophical tool onehas to show that intuition cannot be a source of knowledge and that anepistemic tool should be useless because there is a serious chance itcan fail. Timothy Williamson has argued that we should forget aboutintuition as a cushion in the philosophical armchair (see Williamson2004a,b, 2008, pp. 179–207, and 2009; see also Schaffer 2017).The importance of intuitions in philosophy has been neglected in thepast (see Williamson 2004b, p. 109–110), and for too longintuition didn’t receive the attention it deserves (see, e.g.,DePaul and Ramsey (eds.) 1998). Besides the traditional divide betweenempiricists, rationalists and skeptics, it is not only a verynon-uniform use of the word “intuition” that makes itdifficult to assess the progress of the last years of philosophicalinquiry about intuitions. The situation has been complicated by thecontributions of experimental philosophers on intuitions who adddifferent reasons to question their reliability (see for a carefulcritique of those reasons: Ludwig 2007; see also Ludwig 2018).Generally speaking, the reliability of intuitions has been challengedon two grounds. One stems from an evolutionary explanation of thecapacity to intuit; another is due to experiments which supposedlyshow the cultural relativity or racial and gender sensitivity ofintuitions (see, e.g., Buckwalter and Stich 2010): “…asubstantial list of philosophical intuitions vary across demographicgroups and…they are influenced by a number ofprimafacie irrelevant factors…Some writers…have urgedthat these findings justify a thoroughgoing skepticism about the useof intuitions as evidence in philosophy…But we think thisconclusion is much too strong…” (Stich & Toba 2018,p. 379). After all, knowledge without intuitions (if only common senseassumptions) seems impossible.

The recent discussion of intuitions in epistemology has barely made animpact on philosophical reflections about thought experiments. As faras philosophical thought experiments are concerned, this is as itshould be, according to Williamson. In this respect George Bealer canbe cited in support of Williamson, because for Bealer the talk aboutphilosophical thought experiments reveals a conceptual confusion.Philosophy, he claims, is about “rational intuitions” andthought experiments can be only about “physicalintuitions” (see Bealer 1998, pp. 207–208, and 2002, p.74). To many, this is an implausible claim based on a deeplyproblematic “phenomenology of intuitions” resulting in astrict separation of “rational intuitions” from“physical intuitions”, on such grounds as an allegedimmutability of “rational intuitions”. There are goodreasons to believe that thought experiments appeal to intuitions inorder to give us new insights about different realms of investigation,including philosophy. This kind of positive connection is whatWilliamson has in mind when addressing the role of intuitions inphilosophical thought experiments like the famous Gettier cases, whichovernight found acceptance by the philosophical community in their aimto refute the view that knowledge is justified true belief. WhileWilliamson expects “armchair methods to play legitimately a moredominant role in future philosophy” (Williamson 2009, p. 126),he thinks that “we should stop talking about intuition”(Williamson 2004b, p. 152). This does not impress proponents of whatwe call an intuition-based account of thought experiments, andprobably for good reasons, given the problems in Williamson’sapproach (see, e.g., Dohrn 2016; Ichikawa and Jarvis 2009; Schaffer2017), and the strong empirical evidence in favour of the positiverole that intuitions does play in human cognition (see Myers2004).

4.2 The Intuition–Based Account

What we term the “intuition–based account” ofthought experiments comes in a naturalistic version (see Brendel 2004;Gendler 2007), and in a Platonic version (see Brown 1991a [2011]). Webegin with a discussion of the latter. Brown holds that in a fewspecial cases we do go well beyond the old empirical data to acquire apriori knowledge of nature (see also Koyré 1968). Galileoshowed that all bodies fall at the same speed with a brilliant thoughtexperiment that started by destroying the then reigning Aristotelianaccount. The latter holds that heavy bodies fall faster than lightones (H >L). But consider Figure 6, in which aheavy cannon ball (H) and light musket ball (L) areattached together to form a compound object (H+L);the latter must fall faster than the cannon ball alone. Yet thecompound object must also fall slower, since the light part will actas a drag on the heavy part. Now we have a contradiction:H+L >H andH >H+L. That’s the end of Aristotle’stheory. But there is a bonus, since the right account is now obvious:they all fall at the same speed (H =L =H+L).

Figure 6. Galileo: “I don’teven have to look”

Brown claims this is a priori (though still fallible) knowledge ofnature, since there are no new data involved, nor is the conclusionderived from old data. Moreover, is it some sort of logical truth (fora technical challenge of this claim see Urbaniak 2012). This accountof thought experiments can be further developed by linking the apriori epistemology to accounts of laws of nature that hold that lawsare relations among objectively existing abstract entities. It is thusa form of Platonism, not unlike Platonic accounts of mathematics suchas that urged by Kurt Gödel.

The two most often repeated arguments against this sort of Platonismare: it does not identify criteria to distinguish good from badthought experiments, and it violates the principle of ontologicalparsimony. These seem weak objections. Perhaps they find widespreadacceptance because Platonism seems to be unfashionable these days (seeGrundmann 2018), given the general popularity of various forms ofnaturalism. If intuitions really do the job in a thought experiment,the first objection is weak because neither rationalists norempiricists have a theory about the reliability of intuitions. So theobjection should be that intuitions probably just do not matter inhuman cognition. However, there are good reasons to question the truthof this claim (see Myers 2004). This is not to marginalize theproblems that arise when admitting intuitions as a source of knowledgeand justification, especially in philosophy (see Hitchcock 2012).

As for the second objection, the appeal to Occam’s razor is ingeneral problematic when it is employed to rule out a theory. Whateverwe eliminate by employing the principle of parsimony, we can easilyreintroduce it by an inference to the best explanation (see Meixner2000). And this is exactly what a Platonist contends his or herPlatonism about thought experimenting to be, while conceding that thePlatonic intuition appears miraculous. But are they really moremiraculous than sense perception, which seems similar in many respectsto Platonic intuition? One might want to say yes, because supposedlywe have no clue at all how Platonic intuition works but we do havesome idea about the nature of sense perception. We know that if anobject is far away it appears smaller in vision, and under certainlight conditions the same object can look quite different. However, isit really impossible to state similar rules to capture the nature ofPlatonic intuition? If you are drunk or lack attention you mostprobably will not be very successful in intuiting anything ofphilosophical value.

A review of the relevant psychological literature will reveal furthercriteria that could be employed to identify good and bad conditionsfor Platonic intuition while thought experimenting. Yet, proponents ofthe naturalistic version of the intuition–based account wonderhow necessary Platonism is once this move is entertained in defence ofthe reliability of intuitions (see Miščević 2004).Elke Brendel defines intuitions as mental propositional attitudesaccompanied with a strong feeling of certainty. In her view, we cantell two stories to make sense of their cognitive power andplasticity. One story relates to our biological constitution andevolutionary past. The other is about membership in specializedcommunities. Brendel’s account raises many questions, but it isdifficult to resist its appeal. A universal set is appealing to anyonenot trained in logic because most things we are familiar with can comein sets, such as books, tables, and philosophers. A set of all setsseems intuitively plausible. The intuition disappears once you workedyourself through the problems arising from the idea of a set of allsets. Brendel is quick to insist that such relativity of ourintuitions doesn’t imply that they are cognitively useless.Without intuitions, we probably wouldn’t have knowledge, andthought experiments are sometimes the only way to access theintuitions that guide us in our cognitive lives (see Brendel2004).

4.3 The Argument View

John D. Norton is the most influential advocate of what we call“the argument view” of thought experiments (see Norton1991, 1993, 1996, 2004a,b, 2008). Even though the argument view seemsto be a natural option for empiricists, it seems that most empiricistsfind Norton’s argument view too strong. For this reason, manyparticipants in the debate about thought experiments place themselvesbetween the extreme views of Norton and Brown, which function asuseful foils for apparently more moderate outlooks. Perhaps (withtongue in cheek) they could agree with Bernard Shaw on the virtues ofmoderation, when Shaw said of the typical member of the middle classthat he is moderately honest, moderately intelligent, and moderatelyfaithful to his spouse. Norton claims that any thought experiment isreally a (possibly disguised) argument; it starts with premisesgrounded in experience and follows deductive or inductive rules ofinference in arriving at its conclusion. The picturesque features ofany thought experiment which give it an experimental flavour might bepsychologically helpful, but are strictly redundant. Thus, saysNorton, we never go beyond the empirical premises in a way to whichany empiricist would object.

There are three objections that might be offered against Norton.First, his notion of argument is too vague. However, this might not bethe best objection: arguments can be deductive (which are perfectlyclear) or inductive. If the latter are unclear, the fault is withinduction, not with Norton’s argument view. Second, it is arguedthat Norton simply begs the question: every real world experiment canbe rephrased as an argument, but nobody would say that real worldexperiences are dispensable. The account does not address thequestion: where do the premises come from? A thought experiment mightbe an essential step in making the Norton-style reconstruction. Third,a thought experiment that is presented in argument form loses itstypical force. The soft-point in Brown’s Platonism is linked tothe strength of Norton’s account because Norton claims that anyother view implies a commitment to “asking the oracle.”“Imagine an oracle that claims mysterious powers but neverdelivers predictions that could not be learned by simple inferencesfrom ordinary experience. We would not believe that the oracle had anymysterious powers. I propose the same verdict for thought experimentsin science” (Norton 1996, pp. 1142–1143). Defenders ofempiricist alternatives deny this dispensability thesis. Brendel(2018) offers a most comprehensive review of merits and perils of theargument view.

4.4 Conceptual Constructivism

“Conceptual constructivism”, as we could call it, is amongthe empiricist alternatives to the argument view. The position hasbeen taken up by Van Dyck (2003) to account especially forHeisenberg’s ɣ-ray microscope; but also by Gendler (1998)to makes sense of Galileo’s falling body thought experiment.Gendler’s proposal was advanced in more general terms byCamilleri (2014) in order to establish a firm middle ground betweenthe views of Norton and Brown. Conceptual constructivism was firstproposed by Thomas Kuhn (1964). He employs many of the concepts (butnot the terminology) of his well-knownStructure of ScientificRevolutions. On his view a well-conceived thought experiment canbring on a crisis or at least create an anomaly in the reigning theoryand so contribute to paradigm change. Thought experiments can teach ussomething new about the world, even though we have no new empiricaldata, by helping us to re-conceptualize the world in a new way.Accordingly, some have entertained the option of conceptualconstructivism in the form of a Neo-Kantian reading ofEinstein’s famous clock in the box thought experiment. Such anapproach is inspired by Michael Friedman’s proposal to conceiveof scientific revolutions as times when a Kantian kind of naturalphilosophy plays a major role in guiding scientists from one paradigmto another. The work of Kuhn left us with a puzzle: if scientificrationality is absolutely dependent on a paradigm, and if duringscientific revolutions one paradigm replaces another, not in degreesbut absolutely, comparable to a “Gestalt” switch, thenthis transition from one paradigm to the next cannot be a matter ofscientific rationality. Are scientific revolutions irrational periodsin the history of science? Not necessarily; some kind of naturalphilosophy may guide the process. Friedman has a Kantian naturalphilosophy in mind; his proposal did not earn wide acceptance, but theproblem remains (see Fisch 2017). Be that as it may, it is true thatthought experiments are a valuable currency in times of scientificrevolution. For example, Lennox (1991) has argued that the revolutionbrought about by Charles Darwin in 1859 was made possible by thoughtexperiments (among other things, of course).

4.5 Experimentalism

What we might term “experimentalism” encompasses a widerange of different approaches which all advance the view that thoughtexperiments are a “limiting case” of ordinary experiments.Experimentalism was proposed first by Ernst Mach (1897 and 1905). Hedefines experimenting in terms of its basic method of variation andits capacity to destroy prejudices about nature. According to Mach,experimenting is innate to higher animals, including humans. Thethought experiment just happens on a higher intellectual level but isbasically still an experiment. At the centre of thought experimentingis a “Gedankenerfahrung”, an experience in thought. Suchan experience is possible because thought experiments draw from“unwillkürliche Abbildungen von Tatsachen”(non-arbitrary images of facts) acquired in past experiences of theworld. Some thought experiments are so convincing in their resultsthat an execution seems unnecessary; others could be conducted in areal-world experiment, which is the most natural trajectory of ascientific thought experiment. In any case thought experiments canresult in a revision of belief, thereby demonstrating theirsignificance for scientific progress. Mach also appreciates thedidactic value of thought experiments: they help us to realize whatcan be accomplished in thinking and what cannot.

In the spirit of Mach, Sorensen (1992) has offered an aspiring versionof experimentalism that accounts for thought experiments in scienceand philosophy, and tackles many of the central issues of the topic.Sorensen claims that thought experiments are “a subset ofunexecuted experiments” (1992, p. 213). By their logical naturethey are paradoxes that aim to test modal consequences ofpropositions. The origin of our capacity of thought experimentation isexplained in terms of Darwinian evolution (as in Genz 1999, pp.25–29), though the explanation has been criticized to be onlylittle more than a ‘just so story’ that fails, on aposteriori grounds, to epistemically underwrite that capacity (seeMaffie 1997). Others are more optimistic (see Shepard 2008).

Experimentalism does not have to take a naturalistic turn as it doesin Sorensen’s case. In a number of contributions Marco Buzzonihas defended a Neo-Kantian version of experimentalism (see Buzzoni2004, 2007, 2008, 2011, 2011b, 2013, 2013b). Buzzoni (2008) argues forthe dialectical unity of thought experiments and real-worldexperiments. Thought experiments and real-world experiments areclaimed to be identical on the “technological-operational”level, and at least in science, one is impossible without the other:without thought experiments there wouldn’t be real-worldexperiments because we would not know how to put questions to nature;without real-world experiments there wouldn’t be answers tothese questions or experience from which they could draw. Given themany scientific thought experiments that cannot be realized in thereal-world, Buzzoni might be conflating thought experiments withimagined experiments to be carried out in the real-world (see Fehige2012, 2013b; and Buzzoni 2013b).

Idealizations are common in both real experiments and thoughtexperiments. So-called Aristotelian idealization might ignore, say,the colour of a falling object. Galilean idealizations ignore somephysical aspects, such as air friction, to get at the underlyingphysics (McMullin 1985). So-called Platonic idelization goes beyondthis and ignores what would be actually seen even in a Galileanidealization. For instance, a rapidly moving object in specialrelativity would not look contracted, but rather would look rotated(surprisingly, this phenomenon is not well known). This rotation isignored as an irrelevant optical phenomenon to yield the correctthought experiment visualization, which is the well-known Lorentzcontraction (Brown 2013).

4.6 The Mental–Model Account

The last of the many accounts that emerged in the discussion aboutthought experiments is what could be called the“mental–model account.” It attracts the mostfollowers (see Andreas 2011; Bishop 1998; Cooper 2005; Gendler 2004;Palmieri 2003; Nersessian 1992, 1993, 2007; McMullin 1985;Miščević 1992, 2007, 2021). When we conduct a thoughtexperiment, according to champions of this view, we manipulate amental model instead of the physical realm: “The general claimis that in certain problem-solving tasks people reason by constructinga mental model of the situations, events, and processes that indynamic cases can be manipulated through simulation” (Nersessian2018, p. 319). Like physical models, mental models arenon-propositional in nature. This means first of all “that thecarefully crafted thought–experimental narrative focuses on theconstruction of a model of akind of situation andmanipulating that model through simulation affords epistemic access tocertain features of current representations in a way that manipulatingpropositional representations using logical rules cannot”(Nersessian 2018, pp. 319–320). A narrative functions as a kindof user-manual for building the model, but it isn’t identical toa thought experiment. The biggest problem for the mental–modelaccount is to explain how something non-propositional like a mentalmodel can make an impact on the propositional realm, which happenswhen a thought experiment causes a revision in beliefs.

5. Going Forward

The mental–model approach is one of the most promising of allthe accounts the literature on thought experiments has to offer, andthis for several reasons. First, it does not seem to be much of astretch to draw connections to the intuition–based account. Infact, intuitions maybe the missing link to connect the essentiallynon-propositional activities surrounding mental models, on the onehand, and the propositional aspects of thought experiments, on theother. After all, thought experiments involve propositional reasoning,and somehow the non-propositional and propositional aspects of thoughtexperiments must be linked in any account of thought experiments. Thisis urgent insofar as thought experiments are credited with ameaningful role in theory discovery and theory choice. Second, themental–model approach also allows for inclusion of importantelements of experimentalism and the argument–view. Thoughtexperiments are realized in the mind on mental models, and the methodof variation is employed such that the results of the experiment maybe subject to a careful reconstruction of propositional lines ofreasoning to submit it for careful assessment and critique. Third, themental–model approach enables us to bring an aspect into focusthat has been widely neglected in the discussion so far: the bodilycomponent of (thought) experiments. The exception is the work of thelate David Gooding (1992, 1993, 1994, and 1999). Fourth, in criticalengagement with such naturalistic proposals, those theories of thebody may be put to work that the philosophical school of phenomenologyhas produced (see Fehige and Wiltsche 2013). To be welcomed,therefore, is the entry of phenomenology into the discussion onthought experiments (see Hopp 2014; Wiltsche 2018). Fifth, themental–model account also relates naturally to the mostintriguing discussions about the role of literary fiction in thoughtexperiments.

Some have placed “literary fiction on the level of thoughtexperiments” (Swirski 2007, p. 6). There are two readings ofsuch a claim. According to the first, some literary fiction may be ofcognitive power due to the fact that they are thought experiments. Inother words, we shouldn’t outright reject the idea thatliterature can be of cognitive value. Dystopian novels such asOrwell’s1984 and Huxley’sBrave NewWorld are obvious examples. According to the second reading, thepower of thought experiments is partially a function of the narrativethat conveys it. The work of Novalis remains relevant for theexploration of this link between narrative development and thoughtexperiment: experimental writing and experiments on imagined scenariosgo hand in hand; words and thoughts coincide; mind and matter areentangled (see Daiber 2001). According to the mental–modelapproach, both readings have a valid point. Literary fiction andnarratives of thought experiments can be powerful in establishingmental models in such a way that we can even learn new things aboutthe world at times from the fictional elements of them. The commondenominator is the work on mental models each may facilitate. It is inthis context that an appreciation can grow for Catherine Elgin’stheory of exemplification to argue against the “valorization oftruth in epistemology” (2004, p. 113). This is also the place toconsider Andras Kertesz’s (2015) work on conceptual metaphorresearch in its relevance to the epistemological puzzle that thoughtexperiments pose.

Finally, sixth, mention could be made of visual reasoning inmathematics, which often seems closely related to thought experiments.The standard view of mathematics is that the one and only source ofevidence is a proof, and a proof is a derivation from axioms or firstprinciples. Let’s overlook the problem of where the firstprinciples come from. A simple example such as the following castsdoubt on the standard view:

Theorem:1 + 2 + 3 + … + n = n²/2 + n/2

Proof: See Figure 7.

Figure 7. Picture proof.

The proof works like this: Start at the top and work down, letting thelittle squares represent numbers,1 + 2 + 3 + 4 + 5. Thetotal number of squares in the picture is equal to this sum. Noticealso that the numbers of squares is equal to a large square with sidesof length 5 that is cut in half along the diagonal, i.e.,5²/2, plus the shaded bits that were cut off bythe diagonal cut, i.e.,5/2. It is plausible to claim thatthe diagram is a perfectly good proof of the theorem. One can“see” complete generality in the picture. Even though itonly illustrates the theorem forn = 5, somehow we can seethat it works for every number, all infinitely many of them. Thediagram does not implicitly suggest a “rigorous” verbal orsymbolic proof. The regular proof of this theorem is by mathematicalinduction, but the diagram does not correspond to an inductive proofat all, since the key element in an inductive proof is the passagefromn ton + 1. The simple moral we could draw fromthe example is just this: We can in special cases correctly infertheorems from pictures, that is, from visualizable situations. Thereis an intuition and from this intuition we can grasp the truth of thetheorem (see Brown 1999 [2008]).

Our assessment of the prospects of the mental–model account isvery rough and speculative, though certainly not implausible. Ofcourse, there are challenges to such a vision of a greater synthesisof the many different takes on thought experiments under the umbrellaof the mental-model account. For example, some see additional supportarising for the argument–view from computer simulations (seeBeisbart 2012). Others find that “computational modeling islargely replacing thought experimenting, and the latter will play onlya limited role in future practice of science, especially in thesciences of complex nonlinear, dynamical phenomena” (seeChandrasekharanet al. 2012, p. 239). But, there are alsoproposals such as that by Marcus Schulzke (2014) to think of videogames philosophically as executable thought experiments. Whatever themerits of this particular proposal, future explorations of therelationship between computer simulations and thought experiments canbuild on outcomes of closer inquiries into it (see Behmel 2001, pp.98–108; Di Paoloet al. 2000; El Skaf and Imbert 2013;Lenhard 2011; Stäudner 1998; Lenhard 2018). The work on thenature of the importance of scientific understanding (see, e.g.,Stuart 2018) will inform that exploration as much as the fruits ofcontinuing efforts to clarify the role of the imagination in thoughtexperiments (see, e.g., Meynell 2014; Stuart 2017 and 2021).

We conclude with an interesting, but still relatively unexplored issuethat concerns the relative importance of thought experiments indifferent disciplines. Physics and philosophy use them extensively.Chemistry, by contrast, seems to attract less attention in thisrespect. Why is this the case? Perhaps it is merely an historicalaccident that chemists never developed a culture of doing thoughtexperiments. Perhaps it is tied to some deep feature of the disciplineitself (see Snooks 2006). Economics and history use thoughtexperiments, but apparently not anthropology. A good explanation wouldlikely tell us a lot about the structure of these disciplines.

Related to this is the question of the difference, if any, betweenthought experiments in the sciences and those in philosophy. We haveassumed throughout this entry that they are the same kind of thing.Not everyone sees them this way, so perhaps it should be considered anopen question. On the one hand, philosophy and science seem to many tobe different kinds of activities. That might suggest that thoughtexperiments would differ in the two areas. On the other hand, there isa huge difference between thought experiments within a single field,e.g., Newton’s bucket attempts to establish absolute space whileSchrödinger’s cat aims to show QM as then understood to beabsurd. Is the difference between them less than the differencebetween either of them and Searle’s Chinese Room orThomson’s violinist? The case one way or the other is notobvious. Of course, there are differences between constructive anddestructive thought experiments, but this is true within anydiscipline. Perhaps for now the default attitude ought to be thatthere is no categorical difference between scientific andphilosophical thought experiments. This should not be treated as adogmatic principle, but rather a stimulus to look deeper for importantsubtle contrasts.

Bibliography

The number of papers, anthologies, and monographs has been growingimmensely since the beginning of the 1990s. It might be useful tohighlight that in existing literature, Kühne (2006) remains themost substantial historical study on the philosophical exploration ofthought experiments. And Sorensen (1992) remains the mostcomprehensive philosophical study of thought experiments. More thanother monographs both of these studies well exceed the author’sown systematic contribution to what is widely considered the primaryepistemological challenge presented by thought experiments. Also, thisbibliography does not include the many (we count about eight) popularbooks on thought experiments (likeWittgenstein’s Beetle andOther Classical Thought Experiments by Martin Cohen); nor do welist fiction that is related to the subject (likeThe End of Mr.Y by Scarlett Thomas, orGod’s Debris by ScottAdams). Further, for undergraduate teaching purposes one might want toconsiderDoing Philosophy: An Introduction Through ThoughtExperiments (edited by Theodore Schick, Jr. and Lewis Vaughn,fifth edition, 2012, Boston: McGraw Hill Higher Education), andchapter 5 of Timothy Williamson’s short introduction tophilosophical method (Oxford University Press, 2020). Moreover, anumber of philosophical journals have dedicated part or all of anissue to the topic of thought experiments, including theCroatianJournal of Philosophy (19/VII, 2007),Deutsche Zeitschriftfür Philosophie (1/59, 2011),Informal Logic (3/17,1995),Philosophica (1/72, 2003),Perspectives onScience (2/22, 2014),Berichte zurWissenschaftsgeschichte (1/38, 2015)), as well asTOPOI(4/38, 2019),HOPOS (1/11, 2021), andEpistemologia(12/2022). Furthermore, a companion to thought experiments exists now:The Routledge Companion to Thought Experiments was publishedin 2017. Each includes substantial state of the art reports. Thebibliography that follows aims to list only publications that addressthought experiments as such. Not included are the many specializedpapers that discuss a particular thought experiment in its systematiccontribution to the discussion of a particular issue (such asPutnam’s twin earth scenario to support semantic externalism).An exception is made, of course, when such work is cited. Unlike inprevious versions of this entry, we no longer aim forcomprehensiveness in the bibliography that follows.

• Aligica, Paul D., and Anthony J. Evans, 2009, “ThoughtExperiments, Counterfactuals and Comparative Analysis”,TheReview of Austrian Economics, 22: 225–239.
• Andreas, Holger, 2011, “Zur Wissenschaftslogik vonGedankenexperimenten”,Deutsche Zeitschrift fürPhilosophie, 59: 75–91.
• Arthur, Richard, 1999, “On Thought Experiments as A PrioriScience”,International Studies in the Philosophy ofScience, 13: 215–229.
• Bailey, Cyril, 1950,Lucretius on the Nature of Things(translation ofDe Rerum Naturae), ninth reprint, Oxford:Clarendon Press.
• Batens, Diderik, 2008, “On Possibilities and ThoughtExperiments”, in R. Almeder (ed.),Rescher Studies,Heusenstamm: Ontos Verlag, 29–57.
• Bealer, George, 1998, “Intuition and the Autonomy ofPhilosophy”, in M. DePaul and W. Ramsey (eds.),RethinkingIntuition: The Psychology of Intuition & Its Role in PhilosophicalInquiry, Lanham: Rowman & Littlefield, 201–239.
• –––, 2002, “Modal Epistemology and theRationalist Renaissance”, in T. S. Gendler and J. Hawthorne(eds.),Conceivability and Possibility, Oxford: ClarendonPress, 71–125.
• Beck, Simon, 1992, “Should we Tolerate People whoSplit?”,Southern Journal of Philosophy, 30:1–17.
• –––, 2006, “These Bizarre Fictions:Thought Experiments, Our Psychology and Our Selves”,Philosophical Papers, 35: 29–54.
• Becker, Alexander, 2018, “Thought Experiments inPlato”, in M. T. Stuart et al. (eds.),The RoutledgeCompanion to Thought Experiments, London and New York: Routledge,44–56.
• Behmel, Albrecht, 2001,Was sind Gedankenexperimente?Kontrafaktische Annahmen in der Philosophie des Geistes – derTuringtest und das Chinesische Zimmer, Stuttgart: Ibidem.
• Beisbart, Claus, 2012, “How can Computer Simulations ProduceNew Knowledge?”,European Journal for Philosophy ofScience, 2: 395–434.
• Belkin, Aron, and Tetlock, Philip E. (eds.), 1996,WorldPolitics: Logical, Methodological, and PsychologicalPerspectives, Princeton: Princeton University Press.
• Bishop, Michael, 1998, “An Epistemological Role for ThoughtExperiments”, in N. Shanks (ed.),Idealization IX:Idealization in Contemporary Physics, Amsterdam/Atlanta, GA:Rodopoi, 19–33.
• –––, 1999, “Why Thought Experiments areNot Arguments”,Philosophy of Science, 66:534–541.
• Bokulich, Alisa, 2001, “Rethinking ThoughtExperiments”,Perspectives on Science, 9:285–307.
• Boniolo, Giovanni, 1997, “On a Unified Theory of Models andThought Experiments in Natural sciences”,InternationalStudies in Philosophy of Science, 11: 121–142.
• Borsboom, Denny, et al., 2002, “Functional ThoughtExperiments”,Synthese, 130: 379–387.
• Borstner, Bojan, and Gartner Smiljana (eds.), 2017,ThoughtExperiments between Nature and Society, Newcastle upon Tyne, UK:Cambridge Scholars Publishing.
• Brendel, Elke, 1999, “Gedankenexperimente als Motor derWissenschaftsdynamik”, in J. Mittelstraß (ed.),DieZukunft des Wissens. XVIII. Deutscher Kongress fürPhilosophie, Konstanz: UVK, 785–792.
• –––, 2004, “Intuition Pumps and the ProperUse of Thought Experiments”,Dialectica, 58:88–108.
• –––, 2018, “The Argument View: Are ThoughtExperiments Mere Picturesque Arguments?”, in M. T. Stuart et al.(eds.),The Routledge Companion to Thought Experiments,London and New York: Routledge, 281–293.
• Brooks, David H. M., 1994, “The Method of ThoughtExperiment”,Metaphilosophy, 25: 71–83.
• Brown, James Robert, 1986, “Thought Experiments since theScientific Revolution”,International Studies in thePhilosophy of Science, 1: 1–15.
• –––, 1991 [2011],Laboratory of the Mind:Thought Experiments in the Natural Sciences, London: Routledge,Second Edition.
• –––, 1999 [2008],Philosophy of Mathematics:A Contemporary Introduction to the World of Proofs and Pictures,London and New York: Routledge, second edition.
• –––, 2004, “Why Thought Experiments DoTranscend Empiricism”, in C. Hitchcock (ed.),ContemporaryDebates in the Philosophy of Science, Malden, MA: Blackwell,23–43.
• –––, 2004b, “Peeking into Plato’sHeaven”,Philosophy of Science, 71:1126–1138.
• –––, 2007, “Counter ThoughtExperiments”,Royal Institute of Philosophy Supplement61, 82: 155–177.
• –––, 2011, “Über das Leben im Labordes Geistes”,Deutsche Zeitschrift fürPhilosophie, 59: 65–73.
• –––, 2013, “What Do We See In a ThoughtExperiment?”in M. Frappier et al. (ed.),Thought Experimentsin Science, Philosophy, and the Arts, New York and London:Routledge, 52–68.
• Bunzl, Martin, 1996, “The Logic of ThoughtExperiments”,Synthese, 106: 227–240.
• Buschlinger, Wolfgang, 1993,Denkkapriolen.Gedankenexperimente in Naturwissenschaften, Ethik und Philosophy ofMind, Würzburg: Königshausen & Neumann.
• Butkovic, Ana, 2007, “What is the Function of ThoughtExperiments: Kuhn vs. Brown”,Croatian Journal ofPhilosophy, VII: 63–67.
• Buzzoni, Marco, 2004,Esperimento Ed Esperimento Mentale,Milano: FrankoAngeli.
• –––, 2007, “Zum Verhältnis zwischenExperiment und Gedankenexperiment in den Naturwissenschaften”,Journal for General Philosophy of Science, 38:219–237.
• –––, 2008,Thought Experiment in the NaturalSciences, Würzburg: Königshausen & Neumann.
• –––, 2011, “Kant und dasGedankenexperimentieren”,Deutsche Zeitschrift fürPhilosophie, 59: 93–107.
• –––, 2011b, “On Mathematical ThoughtExperiments”,Epistemologia: Italian Journal for Philosophyof Science, XXXIV: 61–88.
• –––, 2013, “Thought Experiments from aKantian Point of View”, in M. Frappier et al. (ed.),ThoughtExperiments in Science, Philosophy, and the Arts, New York andLondon: Routledge, 90–106.
• –––, 2013b, “On Thought Experiments andthe Kantian A Priori in the Natural Sciences: A Reply to Yiftach J. H.Fehige”,Epistemologia: Italian Journal for Philosophy ofScience, XXXVI: 277–293.
• –––, 2015, “Causality, Teleology, andThought Experiments in Biology”,Journal for GeneralPhilosophy of Science, 46: 279–299.
• –––, 2018, “Pierre Duhem and Ernst Mach onThought Experiments”,HOPOS: The Journal of theInternational Society for the History of Philosophy of Science,8: 1–27.
• Camilleri, Kristian, 2012, “Toward a ConstructivistEpistemology of Thought Experiments in Science”,Synthese, 191: 1697–1716.
• Cargile, James, 1987, “Definitions andCounterexamples”,Philosophy, 62: 179–193.
• Chandrasekharan, Sanjay et al, 2013, “ComputationalModeling: Is This the End of Thought Experiments in Science?”,in M. Frappier et al. (eds.),Thought Experiments in Philosophy,Science, and the Arts, London: Routledge, 239–260.
• Clark, Stephen R. L., 1992, “Where have all the AngelsGone?”,Religious Studies, 28: 221–234.
• Clatterbuck, Hayley, 2013, “The Epistemology of ThoughtExperiments: A Non-eliminativist, Non-platonic Account”,European Journal for Philosophy of Science, 3:309–329.
• Cohnitz, Daniel, 2006,Gedankenexperimente in derPhilosophie, Paderborn: mentis.
• Cole, David, 1984, “Thought and Thought Experiment”,Philosophical Studies, 45: 431–444.
• Cooper, Rachel, 2005, “Thought Experiments”,Metaphilosophy, 36: 328–347.
• Corcilius, Klaus, 2018, “Aristotle and ThoughtExperiments”, in M. T. Stuart et al. (eds.),The RoutledgeCompanion to Thought Experiments, London and New York: Routledge,57–76.
• Cowley, Robert (ed.), 1999,What-If? The World’s MostForemost Military Historians Imagine What Might Have Been, NewYork: Putnam.
• ––– (ed.), 2001,More What-If?: EminentHistorians Imagine What Might Have Been, New York: Putnam.
• Crick, Nathan, 2004, “Conquering our Imagination: ThoughtExperiments and Enthymemes in Scientific Argument”,Philosophy and Rhetoric, 37: 21–41.
• Daiber, Jürgen, 2001,Experimentalphysik des Geistes:Novalis und das romantische Experiment, Göttingen:Vadenhoeck & Ruprecht.
• Dancy, Jonathan, 1985, “The Role of Imaginary Cases inEthics”,Pacific Philosophical Quarterly, 66:141–153.
• Davenport, Edward A., 1983, “Literature as ThoughtExperiment (On Aiding and Abetting the Muse)”,Philosophy ofthe Social Sciences, 13: 279–306.
• Davies, David, 2007, “Thought Experiments and FictionalNarratives”,Croatian Journal of Philosophy, VII:29–45.
• De Baere, Benoit, 2003, “Thought Experiments, Rhetoric, andPossible Worlds”,Philosophica, 72: 105–130.
• De Mey, Tim, 2003, “The Dual Nature View of ThoughtExperiments”,Philosophica, 72: 61–78.
• –––, 2005a, “Remodeling the Past”,Foundations of Science, 10: 47–66.
• –––, 2005b, “Tales of the Unexpected:Incongruity-Resolution in Humour Comprehension, Scientific Discoveryand Thought Experimentation”,Logic and LogicalPhilosophy, 14: 69–88.
• –––, 2006, “Imagination’s Grip onScience”,Metaphilosophy, 37: 222–239.
• Dennett, Daniel C., 1984,Elbow Room: The Varieties of FreeWill Worth Wanting, Cambridge: MIT Press.
• –––, 1991,Consciousness Explained, NewYork: Little Brown.
• –––, 1995, “Intuition Pumps”, in J.Brockman (ed.),The Third Culture, New York et al.: Simon& Schuster, 181–197.
• –––, 2005,Sweet Dreams, Cambridge, MA:MIT Press.
• DePaul, Michael, and William Ramsey (eds.), 2002,RethinkingIntuition: The Psychology of Intuition & Its Role in PhilosophicalInquiry, Lanham: Rowman & Littlefield.
• Diamond, Cora, 2002, “What if X Isn’t the Number ofSheep? Wittgenstein and Thought-Experiments in Ethics”,Philosophical Papers, 31: 227–250.
• Di Paolo, E. A. et al., 2000, “Simulation Models as OpaqueThought Experiments”, in M. A. McCaskill et al. (eds.),Proceedings of the Seventh International Conference on ArtificialLife, Cambridge: MIT Press, 497–506.
• Ducheyne, Steffen, 2006, “The Argument(s) for UniversalGravitation”,Foundations of Science, 11:419–447.
• Duhem, Pierre, 1913,La théorie physique son objet etsa structure, 2^nd edition, Paris: Chevalier &Rivière, used here in reprint of 1981, Paris: LibrairiePhilosophique J. Vrin (translated by P. Weiner, 1960,Aim andStructure of Physical Theory, Princeton: Princeton UniversityPress).
• El Skaf, Rawad, and Cyrille Imbert, 2013, “Unfolding in theEmpirical Sciences: Experiments, Thought Experiments and ComputerSimulations”,Synthese, 190: 3451–3474.
• Elgin, Catherine Z., 1993, “Understanding: Art andScience”,Synthese, 95: 13–28.
• –––, 1993b, “True Enough”, in ErnestSosa and Enrique Villanueva (eds.),Epistemology: PhilosophicalIssues: Volume 14, Boston, MA: Blackwell Publishing,113–131.
• –––, 2007, “The Laboratory of theMind”, in J. Gibson (ed.),A Sense of the World: Essays onFiction, Narrative and Knowledge, Oxon: Routledge-Taylor Francis,43–54.
• –––, 2014, “Fiction as ThoughtExperiment”,Perspectives on Science, 22:221–241.
• Fehige, Yiftach J. H., 2012, “Experiments of Pure Reason.Kantianism and Thought Experiments in Science”,Epistemologia: Italian Journal for Philosophy of Science, 35:141–160.
• –––, 2013, “The Relativized A Priori andthe Laboratory of the Mind: Towards a Neo-Kantian Account of ThoughtExperiments in Science”,Epistemologia: Italian Journal forPhilosophy of Science, 36: 55–63.
• –––, 2018, “Thought Experiments andTheology”, in M. T. Stuart et al. (eds.),The RoutledgeCompanion to Thought Experiments, London and New York: Routledge,183–194.
• –––, 2024,Thought Experiments, Science, andReligion, Leiden: Brill.
• –––, and Harald Wiltsche, 2013, “The Body,Thought Experiments, and Phenomenology”, in M. Frappier et al.(eds.),Thought Experiments in Philosophy, Science, and theArts, London: Routledge, 69–89.
• –––, and Michael T. Stuart, 2014, “Originsof the Philosophy of Thought Experiments: The Forerun”,Perspectives on Science, 22: 179–220.
• Fisch, Menachem, 2017,Creatively Undecided: Toward a Historyand Philosophy of Scientific Agency, Chicago: Chicago UniversityPress.
• –––, 2019, “Gulliver and the Rabbis:Counterfactual Truth in Science and the Talmud”,Religions:An Open Access Journal of Theology,https://doi.org/10.3390/rel10030228.
• Focquaert, Farah, 2003, “Personal Identity and itsBoundaries: Philosophical Thought Experiments”,Philosophica, 72: 131–152.
• Fodor, Jerry, 1964, “On Knowing What We Would Say”,Philosophical Review, 73: 198–212.
• Frappier, Melanie, and Letitia Meynell, and James Robert Brown(eds.), 2013,Thought Experiments in Philosophy, Science, and theArts, London: Routledge.
• Fuhrer, Therese, 2007, “Der Geist im VollkommenenKörper: Ein Gedankenexperiment in Augustins De civitate22”, in D. Frede and B. Reis (eds.),Body and Soul inAncient Philosophy, Berlin: de Gruyter, 465–491.
• Gähde, Ulrich, 2000, “Gedankenexperimente inErkenntnistheorie und Physik: Strukturelle Parallelen”, in J.Nida-Rümlin,Rationalität, Realismus, Revision,Berlin: de Gruyter, 457–464.
• Gendler, Tamar S., 1998, “Galileo and the Indispensabilityof Scientific Thought Experiment”,The British Journal forthe Philosophy of Science, 49:397–424.
• –––, 2000,Thought Experiment: On the Powersand Limits of Imaginary Cases, New York: Garland Press (nowRoutledge).
• –––, 2002a, “Personal Identity andThought-Experiments”,Philosophical Quarterly, 52:34–54.
• –––, 2002b, “Thought Experiment”,Encyclopedia of Cognitive Science, New York; London:Nature/Routledge.
• –––, 2004, “Thought ExperimentsRethought— and Reperceived”,Philosophy ofScience, 71: 1152–1164.
• –––, 2005, “Thought Experiments inScience”,Encyclopedia of Philosophy, New York:MacMillan.
• –––, 2007, “Philosophical ThoughtExperiments, Intuitions, and Cognitive Equilibrium”,MidwestStudies in Philosophy of Science, 31: 68–89.
• Gendler, Tamar S., and John Hawthorne (eds.), 2002,Conceivability and Possibility. New York/Oxford:Clarendon/Oxford University Press.
• Genz, Henning, 1999,Gedankenexperimente, Weinheim:Wiley-VCH.
• Georgiou, Andreas, 2007, “An Embodied Cognition View ofImagery-Based Reasoning in Science: Lessons from ThoughtExperiments”,Croatian Journal of Philosophy, VII:29–45.
• Glas, Eduard, 1999, “Thought Experimentation andMathematical Innovation”,Studies in History and Philosophyof Science, 30: 1–19.
• Gomila, Antoni, 1991, “What is a Thought Experiment?”,Metaphilosophy, 22: 84–92.
• Gooding, David C., 1992, “The Cognitive Turn, or, Why DoThought Experiments Work?”, in R. N. Giere (ed.),CognitiveModels of Science, Minneapolis: University of Minnesota Press,45–76.
• –––, 1993, “What is Experimental AboutThought Experiments?”,Proceedings of the Philosophy ofScience Association, 2: 280–290.
• –––, 1994, “Imaginary Science”,British Journal for the Philosophy of Science, 45:1029–1045.
• –––, 1999, “Thought Experiments”, inE. Craig (ed.),Encyclopedia of Philosophy: Volume 9, London:Routledge, 392–397.
• Grant, Edward, 2007, “Thought Experiments and the Role ofthe Imagination”, inA History of Natural Philosophy: Fromthe Ancient World to Nineteenth Century, Cambridge: CambridgeUniversity Press, 200–211.
• Gregersen, Niels H., 2014, “The Role of Thought Experimentsin Science and Religion”, in Michael Welker (ed.)Scienceand Religion: Past and Future, Frankfurt/Main: Peter Lang,129–140.
• Grosdanoff, Boris, 2007, “Reconstruction, Justification andIncompatibility in Norton’s Account of ThoughtExperiments”,Croatian Journal of Philosophy, VII:69–79.
• Grundmann, Thomas, 2018, “Plato and the A Priori in ThoughtExperiments”, in M. T. Stuart et al. (eds.),The RoutledgeCompanion to Thought Experiments, London and New York: Routledge,293–308.
• Grudmann, Thomas and Joachim Horvath, 2014, “ThoughtExperiments and the Problem of Deviant Realisations”,Philosophical Studies, 170(3): 525–533.
• Hacking, Ian, 1993, “Do Thought Experiments have a Life ofTheir Own?”,Proceedings of the Philosophy of ScienceAssociation, 2: 302–308.
• Häggqvist, Sören, 1996,Thought Experiments inPhilosophy, Stockholm: Almqvist & Wiksell International.
• –––, 2007, “The A Priori Thesis: Acritical Assessment”,Croatian Journal of Philosophy,VII: 47–61.
• –––, 2009, “A Model for ThoughtExperiments”,Canadian Journal of Philosophy, 39:56–76.
• Hauerwas, Stanley, 1996,“Virtue, Description, andFriendship: A Thought Experiment in Catholic Moral Theology”,Irish Theological Quarterly, 62: 170–184.
• Helm, Hugh, and John Gilbert, 1985, “Thought Experiments andPhysics Education – Part 1”,Physics Education,20: 124–131.
• Helm, Hugh, and John Gilbert, and Michael D. Watts, 1985,“Thought Experiments and Physics Education – Part2”,Physics Education, 20: 211–217.
• Herfeld, Catherine, 2019, “Imagination Rather ThanObservation in Econometrics: Ragnar Frisch’s HypotheticalExperiments as Thought Experiments”,HOPOS: The Journal ofthe International Society for the History of Philosophy ofScience, 9: 35–74.
• Hershenov, David B., 2008, “A Hylomorphic Account of ThoughtExperiments Concerning Personal Identity”,American CatholicPhilosophical Quarterly, 82: 481–502.
• Hintikka, Jaakko, 1999, “The Emperor’s NewIntuitions”,The Journal of Philosophy, 96:127–147.
• Hitchcock, Christopher, 2012, “Thought Experiments, RealExperiments, and the Expertise Objection”,European Journalfor Philosophy of Science, 2: 205–218.
• Hopp, Walter, 2014, “Experiments in Thought”,Perspectives on Science, 22: 242–263.
• Horowitz, Tamara, 1998, “Philosophical Intuitions andPsychological Theory”,Ethics, 108: 367–385.
• Horowitz, Tamara, and Gerald Massey (eds.), 1991,ThoughtExperiments in Science and Philosophy, Lanham: Rowman &Littlefield.
• Humphreys, Paul, 1993, “Seven Theses on ThoughtExperiments”, in J. Earmanet al. (eds.),Philosophical Problems of the Internal and External World,Pittsburgh: University of Pittsburgh Press, 205–227.
• Ichikawa, Jonathan, and Benjamin Jarvis, 2009,“Thought-experiment Intuition and truth in Fiction”,Philosophical Studies, 142: 221–246.
• Ierodiakonou, Katerina, 2005, “Ancient Thought Experiments:A First Approach”,Ancient Philosophy, 25:125–140.
• –––, 2018, “The Triple Life of AncientThought Experiments”, in M. T. Stuart et al. (eds.),TheRoutledge Companion to Thought Experiments, London and New York:Routledge, 31–43.
• Ierodiakonou, Katerina, and Sophia Roux (eds.), 2011,ThoughtExperiments in Methodological and Historical Contexts, Leiden:Brill.
• Inwagen, Peter van, 1998, “Modal Epistemology”,Philosophical Studies, 92: 67–84.
• Irvine, Andrew D., 1991, “Thought Experiments in ScientificReasoning”, in T. Horowitz and G. Massey (eds.),ThoughtExperiments in Science and Philosophy, Lanham: Rowman &Littlefield, 149–166.
• Jackson, Frank, 1982, “Epiphenomenal Qualia”,Philosophical Quarterly, 32: 27–36.
• –––, 2009, “Thought Experiments andPossibilities”,Analysis, 69: 100–109.
• Jackson, M. W., 1992, “TheGedankenexperimentMethod of Ethics”,The Journal of Value Inquiry, 26:525–535.
• Janis, Allen I., 1991, “Can Thought ExperimentsFail?”, in T. Horowitz and G. Massey (eds.),ThoughtExperiments in Science and Philosophy, Lanham: Rowman &Littlefield, 113–118.
• Kalin, Martin G., 1972, “Kant’s TranscendentalArguments as Gedankenexperimente”,Kant Studien, 63:315–328.
• Kassung, Christian, 2004, “Tichy · 2 = ?:Relativitätstheorie als literarisches Gedankenexperiment”,in T. Macho and A. Wunschel (eds.),Science & Fiction,Frankfurt am Main: Fischer, pp. 17–32.
• Kertesz, Andras, 2015, “The Puzzle of Thought Experiments inConceptual Metaphor Research”,Foundations of Science,20: 147–174.
• King, Peter, 1991, “Mediaeval Thought-Experiments: TheMetamethodology of Mediaeval Science”, in T. Horowitz and G.Massey (eds.),Thought Experiments in Science and Philosophy,Lanham: Rowman & Littlefield, 43–64.
• Klassen, Stephen, 2006, “The Science Thought Experiment: HowMight it be Used Profitably in the Classroom?”,Interchange, 37: 77–96.
• Klee, Robert, 2008, “Physical Scale Effects andPhilosophical Thought Experiments”,Metaphilosophy, 39:89–104.
• Knox, Philip, and Jonathan Morton, and Daniel Reeve (eds.), 2018,Medieval Thought Experiments: Poetry, Hypothesis, and Experiencein the European Middle Ages, Turnhout, Belgium : Brepols.
• Kolak, Daniel, 1993, “Metaphysics and Metapsychology ofPersonal Identity: Why Thought Experiments Matter in Deciding Who WeAre”,American Philosophical Quarterly, 30:39–50.
• Koyré, Alexandre, 1968,Metaphysics andMeasurement, London: Chapman and Hall.
• Krauthausen, Karin, 2015, “Ermittlung der Empirie: Zu ErnstMachs Methode des Gedankenexperiments”,Berichte zurWissenschaftsgeschichte, 38: 15–40.
• Krimsky, Sheldon, 1973, “The Use and Misuse of CriticalGedankenexperimente”,Zeitschrift für allgemeineWissenschaftstheorie, 4: 323–334.
• Kuhn, Thomas S., 1964, “A Function for ThoughtExperiments”, reprinted in T. Kuhn,The EssentialTension, Chicago: University of Chicago Press, 1977,240–265.
• Kühne, Ulrich, 2005,Die Methode desGedankenexperiments, Frankfurt: Suhrkamp.
• Kujundzic, Nebojsa, 1995, “Thought Experiments: Architectureand Economy of Thought”,The Journal of the British Societyfor Phenomenology, 26: 86–93.
• –––, 1998, “The Role of Variation inThought Experiments”,International Studies in thePhilosophy of Science, 12: 239–243.
• –––, 2013, “Thought Experiments, Puzzles,and Paradoxes”,Philosophy Study, 3:750–758.
• Kukkonen, Taneli, 2014, “Ibn Sina and the Early History ofThought Experiments”,Journal of the History ofPhilosophy, 53: 433–453.
• Laymon, Ronald, 1991, “Thought Experiments of Stevin, Machand Gouy: Thought Experiments as Ideal Limits and as SemanticDomains”, in T. Horowitz and G. Massey (eds.),ThoughtExperiments in Science and Philosophy, Lanham: Rowman &Littlefield, 167–192.
• Lenhard, Johannes, 2011, “Epistemologie der Iteration:Gedankenexperimente und Simulationsexperimente”,DeutscheZeitschrift für Philosophie, 59: 131–154.
• Lennox, James G., 1991, “Darwinian Thought Experiments: AFunction for Just-So Stories”, in T. Horowitz and G. Massey(eds.),Thought Experiments in Science and Philosophy,Lanham: Rowman & Littlefield, 223–245.
• Ludwig, Kirk, 2007, “The Epistemology of ThoughtExperiments: First Person versus Third Person Appraoches”,Midwest Studies in Philosophy, 31: 128–159.
• –––, 2018, “Thought Experiments andExperimental Philosophy”, in M. T. Stuart et al. (eds.),TheRoutledge Companion to Thought Experiments, London and New York:Routledge, 385–405.
• Maar, Alexander, 2014, “Possible Uses of CounterfactualThought Experiments in History”,Principia, 18:87–113.
• Mach, Ernst, 1883,Die Mechanik in ihrer Entwicklung:Historisch-Kritisch Dargestellt, seventh edition, used here inreprint of 1988, Berlin; Akademie Verlag (trans. by J. McCormack,The Science of Mechanics, LaSalle Illinois: Open Court).
• –––, 1897, “ÜberGedankenexperimente”, in:Zeitschrift für denphysikalischen und chemischen Unterricht, 10: 1–5.
• –––, 1905, “ÜberGedankenexperimente”, inErkenntnis und Irrtum,Leipzig: Verlag von Johann Ambrosius Barth, 181–197, (translatedby J. McCormack, inKnowledge and Error, Dordrecht: Reidel,134–147).
• Macho, Thomas, and Anette Wunschel (eds.), 2004,Science andFiction:Über Gedankenexperimente in Wissenschaft, Philosophie undLiteratur, Frankfurt am Main: Fischer Verlag.
• Machery, Edouard, 2011,“Thought Experiments andPhilosophical Knowledge”,Metaphilosophy, 42:191–214.
• Maffie, James, 1997, “Just-so Stories about ‘InnerCognitive Africa’: Some Doubts about Sorensen’sEvolutionary Epistemology of Thought Experiments”,Biologyand Philosophy, 12: 207–224.
• Malmgren, Anna-Sara, 2011, “Rationalism and the Content ofIntuitive Judgements”,Mind, 120: 263–327.
• Massey, Gerald, 1991, “Backdoor Analyticity”, in T.Horowitz and G. Massey (eds.),Thought Experiments in Science andPhilosophy, Lanham: Rowman & Littlefield, 285–296.
• Mayer, Verena, 1999, “Was zeigenGedankenexperimente?”,Philosophisches Jahrbuch, 106:357–378.
• McAllister, James, 1996, “The Evidential Significance ofThought Experiments in Science”,Studies in History andPhilosophy of Science, 27: 233–250.
• –––, 2004, “Thought Experiments and theBelief in Phenomena”,Proceedings of the 2002 BiennialMeeting of the Philosophy of Science Association, Philosophy ofScience, 71: 1164–1175.
• –––, 2005, “The Virtual Laboratory:Thought Experiments in Seventeenth-Century Mechanics”, in H.Schramm et al. (eds.),Collection, Laboratory, Theater: Scenes ofKnowledge in the 17th Century, New York: Walter de Gruyter,35–56.
• McComb, Geordie, 2013, “Thought Experiment, Definition, andLiterary Fiction”, in M. Frappier et al. (eds.),ThoughtExperiments in Philosophy, Science, and the Arts, London:Routledge, 207–222.
• McGinnis, Jon, 2018, “Experimental Thoughts on ThoughtExperiments in Medieval Islam”, in Michael T. Stuart et al.(eds.),The Routledge Companion to Thought Experiments,London and New York: Routledge, 77–91.
• McMullin, Ernan, 1985, “Galilean Idealization”,Studies in History and Philosophy of Science, 16:247–273.
• Meinong, Alexius, 1907, “Das Gedankenexperiment”, inÜber die Stellung der Gegenstandstheorie im System derWissenschaften (reprinted in the fifth volume of the collectedworks of Meinong, edited by Rudolf Haller and Rudolf Kindinger,Graz-Austria: Akademische Druck und Verlagsanstalt, 1973,273–283 [67–77]).
• Meixner, Uwe, 2000, “On Some Realisms Most RealistsDon’t Like”,Metaphysica, 1: 73–94.
• Malament, David B., 2008, “Norton’s SlipperySlope”,Philosophy of Science, 75: 799–818.
• Meynell, Letitia, 2014, “Imagination and Insight: A newAcount [sic!] of the Content of Thought Experiments”,Synthese, 191: 4149–4168.
• Milligan, Tony, 2019, “Thought Experiments andNovels”,Studia Humana, 8: 84–92.
• Miščević, Nenad, 2021,ThoughtExperiments, Cham: Springer.
• –––, 1992, “Mental Models and ThoughtExperiments”,International Studies in the Philosophy ofScience, 6: 215–226.
• –––, 1997, “Categorial and EssentialistIntuitions: A Naturalist Perspective”,Acta Analytica,12: 21–39.
• –––, 2004, “The Explainability ofIntuitions”,dialectica, 58: 43–70.
• –––, 2007, “Modelling Intuitions andThought Experiments”,Croatian Journal of Philosophy,VII: 181–214.
• Morrison, Margaret, 2009, “Models, Measurements, andComputer Simulations: the Changing Face of Experimentation”,Philosophical Studies, 143: 33–57.
• Moue, Aspasia S. et al., 2006, “Tracing the Development ofThought Experiments in the Philosophy of the Natural Sciences”,Journal for General Philosophy of Science, 37:61–75.
• Myers, Mason C., 1968, “Thought Experiments and SecretStores of Information”,International PhilosophicalQuarterly, 8: 180–192.
• –––, 1986, “Analytical ThoughtExperiments”,Metaphilosophy, 17: 109–118.
• Myers, David G., 2004),Intuition. Its powers and perils,New Haven; London: Yale University Press.
• Nersessian, Nancy, 1992, “How Do Scientists Think? Capturingthe Dynamics of Conceptual Change in Science”, in R. Giere(ed.),Cognitive Models of Science, Minneapolis: Universityof Minnesota Press, 3–44.
• –––, 1993, “In the Theoretician’sLaboratory: Thought Experimenting as Mental Modeling”,Proceedings of the Philosophy of Science Association, 2:291–301.
• –––, 2007, “Thought Experiments as MentalModelling: Empiricism without Logic”,Croatian Journal ofPhilosophy, VII: 125–161.
• –––, 2018, “Cognitive Science, MentalModels, and Thought Experiments”, in Michael T. Stuart et al.(eds.),The Routledge Companion to Thought Experiments,London and New York: Routledge, 309–326.
• Norton, John D., 1991, “Thought Experiments inEinstein’s Work”, in T. Horowitz and G. Massey (eds.),Thought Experiments in Science and Philosophy, Lanham: Rowman& Littlefield, 129–148.
• –––, 1993, “Einstein and Nordström:Some lesser known Thought Experiments in Gravitation”, in J.Earman et al. (eds.),The Attraction of Gravitation: New Studiesin History of General Relativity, Boston: Birkhäuser,3–29.
• –––, 1993b, “Seeing the Laws ofNature”,Metascience, 3:33–38.
• –––, 1996, “Are Thought Experiments JustWhat You Thought?”,Canadian Journal of Philosophy, 26:333–366.
• –––, 2004a, “On Thought Experiments: IsThere More to the Argument?”Proceedings of the 2002Biennial Meeting of the Philosophy of Science Association, Philosophyof Science, 71: 1139–1151. [Preprint available online].
• –––, 2004b, “Why Thought Experiments DoNot Transcend Empiricism”, in C. Hitchcock (ed.),Contemporary Debates in the Philosophy of Science, Oxford:Blackwell, 44–66. [Preprint available online].
• –––, 2008, “The Dome: An UnexpectedlySimple Failure of Determinism”,Philosophy of Science,75: 786–798.
• Ornelas, J., A. Cintora, and P. Hernandez, 2018,Trabajando enel Laboratorio de la Mente: Naturaleza y Alcance de los ExperimentoMentales, San Luis Potosí: Universidad Autónoma deSan Luis Potosí.
• Palmer, Anthony, 1984, “The Limits of AI: ThoughtExperiments and Conceptual investigations”, in S. Torrance(ed.),The Mind and the Machine: Philosophical Aspects ofArtificial Intelligence, Chichester: Ellis Horwood, pp.43–50.
• Palmieri, Paolo, 2003, “Mental Models in Galileo’sEarly Mathematization of Nature”,Studies in History andPhilosophy of Science, 34: 229–264.
• –––, 2005, “‘Spuntur lo scogliopiù duro’: Did Galileo Ever Think the Most BeautifulThought experiment in the History of Science?”,Studies inHistory and Philosophy of Science, 36: 305–322.
• Parfit, Derek, 1987,Reasons and Persons, Oxford:Clarendon Press.
• Peijnenburg, Jeanne, and David Atkinson, 2003, “When areThought experiments Poor Ones?”,Journal for GeneralPhilosophy of Science, 34: 305–322.
• –––, 2007, “On poor and not so PoorThought experiments: A reply to Daniel Cohnitz”,Journal forGeneral Philosophy of Science, 38: 159–161.
• Perler, Dominik, 2008, “Thought Experiments: TheMethodological Function of Angels in Late MedievalEpistemology”, in Isabel Iribarren and Markus Lenz (eds.),Angels in Medieval Philosophical Inquiry, Aldershot: Ashgate,143–153.
• Picha, Marek, 2011,Kdyby chyby. Epistemologiemyšlenkových experimentů, Olomouc:Nakladatelství Olomouc.
• Popper, Karl, 1959, “On the Use and Misuse of ImaginaryExperiments, especially in Quantum Theory”, inThe Logic ofScientific Discovery, London: Hutchinson, 442–456.
• Poser, Hans, 1984, “Wovon handelt einGedankenexperiment?”, in H. P. Poser and H.-W. Schütt(eds.),Ontologie und Wissenschaft. Philosophische undwissenschaftliche Untersuchungen zur Frage derObjektkonstruktion, Berlin: TUB-Dokumentationen, Issue 19:181–198.
• Praem, Sara Kier, and Asbjorn Steglich–Petersen, 2015,“Philosophical Thought Experiments as Heuristics for TheoryDiscovery”,Synthese, 192: 2827–2842.
• Prudovsky, Gad, 1989, “The Confirmation of the SuperpositionPrinciple: The Role of a Constructive Thought Experiment inGalileo’sDiscorsi”,Studies in History andPhilosophy of Science, 20: 453–468.
• Pulte, Helmut, 2007, “Gedankenexperiment und PhysikalischesWeltbild: Überlegungen zur Bildung von Naturwissenschaft und zurnaturwissenschaftlichen Bildung im Anschluss an A. Einstein”, inDie Kultur moderner Wissenschaft am Beispiel Albert Einstein,München: Akademie Verlag, 39–68.
• Puskaric, Ksenija, 2007, “Brown and Berkeley”,Croatian Journal of Philosophy, VII: 177–180.
• Rehder, Wulf, 1980a, “Thought Experiment and ModalLogic”,Logique et Analyse, 23: 407–417.
• –––, 1980b, “Versuche zu einer Theorie vonGedankenexperimenten”,Grazer Philosophische Studien,11: 105–123.
• Reiss, Julian, 2009, “Counterfactuals, Thought Experimentsand Singular Causal Inference in History”,Philosophy ofScience, 76: 712–723.
• –––, 2013, “Genealogical ThoughtExperiments in Economics”, in M. Frappier et al. (eds.),Thought Experiments in Philosophy, Science, and the Arts,London: Routledge, 177–190.
• Rescher, Nicholas, 1991, “Thought Experiments in PresocraticPhilosophy”, in T. Horowitz and G. Massey (eds.),ThoughtExperiments in Science and Philosophy, Lanham: Rowman &Littlefield, 31–42.
• –––, 2003,Imagining Irreality,Chicago; La Salle: Open Court.
• –––, 2005,What If?: Thought Experimentationin Philosophy, New Brunswick, NJ: Transaction Publishers.
• Roberts, Francis, 1993, “Thought Experiments and SocialTransformation”,Journal for the Theory of SocialBehaviour, 23: 399–421.
• Robinson, Howard, 2004, “Thought Experiments, Ontology andConcept-Dependent Truthmakers”,Monist, 87:537–553.
• Saam, Nicole J., 2017, “What is a Computer Simulation? AReview of a Passionate Debate”,Journal for GeneralPhilosophy of Science, 48: 293–309.
• Saint-Germier, Pierre, forthcoming, “Getting GettierStraight: Thought Experiments, Deviant Realizations and DefaultInterpretations”,Synthese.
• Schabas, Margaret, 2008, “Hume’s Monetary ThoughtExperiments”,Studies in History and Philosophy ofScience, 39: 161–168.
• Schildknecht, Christiane, 1990,Philosophische Masken:Literarische Formen der Philosophie bei Platon, Descartes, Wolff undLichtenberg, Stuttgart: Metzler.
• Schlesinger, George N., 1995, “The Power of ThoughtExperiments”,Foundations of Physics, 26:467–482.
• Schöne, A., 1982,Aufklärung aus dem Geist derExperimentalphysik: Lichtenbergsche Konjunktive, München: C.H. Beck.
• Schrödinger, E., 1935, “The Present Situation inQuantum Mechanics”, trans in J. Wheeler and W. Zurek (eds.),Quantum Theory and Measurement, Princeton: PrincetonUniversity Press.
• Schulzke, Marcus, 2014, “Simulating Philosophy: InterpretingVideo Games as Executable Thought Experiments”,Philosophyand Technology, 27: 251–265.
• Shafer, Michael J., 2017, “’Filling In’, ThoughtExperiments and Intuitions”,Episteme, 14:255–262.
• Shepard, Roger N., 2008, “The Step to Rationality: TheEfficacy of Thought Experiments in Science, Ethics, and FreeWill”,Cognitive Science, 32: 3–335.
• Snooks, R. J., 2006, “Another Scientific Practice SeparatingChemistry from Physics”,Foundations of Chemistry, 8:255–270.
• Sorensen, Roy A., 1992,Thought Experiments, Oxford:Oxford University Press.
• Souder, Lawrence, 2003, “What are we to think about ThoughtExperiments?”,Argumentation: An International Journal onReasoning, 17: 203–217.
• Sriraman, Bharath, 2006, “An Ode to Imre Lakatos:Quasi-Thought Experiments to Bridge the Ideal and Actual MathematicsClassrooms”,Interchange, 37: 151–178.
• Starikova, Irina, 2007, “Proofs and Pictures”,Croatian Journal of Philosophy, VII: 81–92.
• Stern, J. P., 1963,Lichtenberg: A Doctrine of ScatteredOccassions, London: Thames and Hudson.
• Stich, Stephen, and Kevin Toba, 2018, “Intuition and ItsCritics”, in M. T. Stuart et al. (eds.),The RoutledgeCompanion to Thought Experiments, London and New York: Routledge,369–384.
• Stinner, Arthur, 1990, “Philosophy, Thought Experiments, andLarge Context Problems in the Secondary Physics Course”,International Journal of Science Education, 12:244–157.
• Stöltzner, Michael, 2003, “The Dynamics of ThoughtExperiments-Comment to Atkinson”, in M. Galavotti (ed.),Observation and Experiment in the Natural and SocialSciences, Kluwer: Dordrecht, 2003, 243–258.
• Stuart, Michael T., 2014, “Cognitive Science and ThoughtExperiments: A Refutation of Paul Thagard’s Skepticism”,Perspectives on Science, 22: 264–287.
• –––, 2016, “Taming theory with ThoughtExperiments: Understanding and Scientific Progress”,Studiesin History and Philosophy of Science, 58: 24–33.
• –––, 2016b, “Norton and the Logic ofThought Experiments”,Axiomathes, 26:451–466.
• –––, 2017, “Imagination: A Sine Qua Non ofScience”,Croatian Journal of Philosophy, XVII:9–32.
• –––, 2018, “How Thought ExperimentsIncrease Understanding”, in M. T. Stuart et al. (eds.),TheRoutledge Companion to Thought Experiments, London and New York:Routledge, 526–544.
• –––, 2021, “Towards a Dual ProcessEpistemology of Imagination”,Synthese, 198:1329–1350.
• Stuart, Michael T., Yiftach Fehige, and James R. Brown (eds.),2018,The Routledge Companion to Thought Experiments, Londonand New York: Routledge.
• Swirski, Peter, 2007,Of Literature and Knowledge:Explorations in Narrative Thought Experiments, Evolution and GameTheory, London & New York: Routledge.
• Taliaferro, Charles, 2001, “Sensibility and Possibilia: ADefense of Thought Experiments”,Philosophia Christi,3: 403–420.
• –––, and Elliot Knuths, 2017, “ThoughtExperiments in Philosophy of Religion: The Virtues of PhenomenologicalRealism and Values”,Open Theology, 3:167–173.
• Tetlock, Philip E. et al. (eds.), 2009,Unmaking the West:What-If? Scenarios That Rewrite World History, Ann Arbor:University of Michigan Press.
• Thagard, Paul, 2014, “Thought Experiments ConsideredHarmful”,Perspectives on Science, 22:288–305.
• Thaler, Mathias, 2016, “Unhinged Frames: Assessing ThoughtExperiments in Normative Political Theory”,British Journalof Political Science, 48: 1119–1141.
• Thoma, Johanna, 2016, “On the Hidden Thought Experiments ofEconomic Theory”,Philosophy of the Social Sciences,46: 129–146.
• Thomson, Judith Jarvis, 1971, “A Defense of Abortion”,Philosophy and Public Affairs, 1: 47–66.
• Urbaniak, Rafal, 2012, “”Platonic“ ThoughtExperiments: How on Earth?”,Synthese, 187:731–752.
• Urbaniec, Jacek, 1988, “In Search of a PhilosophicalExperiment”,Metaphilosophy, 19: 294–306.
• Van Bendegem, Jean Paul, 2003, “Thought Experiments inMathematics: Anything But Proof”,Philosophica, 72:9–33.
• Van Dyck, Maarten, 2003, “The Roles of One ThoughtExperiment in Interpreting Quantum Mechanics: Werner Heisenberg MeetsThomas Kuhn”,Philosophica, 72: 1–21.
• Walsh, Adrian, 2011, “A Moderate Defence of the Use ofThought Experiments in Applied Ethics”,Ethical Theory andMoral Practice, 14: 467–481.
• Weber, Erik, and Tim DeMey, 2003, “Explanation and ThoughtExperiments in History”,History and Theory, 42:28–38.
• Wilkes, Kathleen V., 1988,Real People: Personal Identitywithout Thought Experiments, Oxford: Oxford UniversityPress.
• Williamson, Timothy, 2004a, “Armchair Philosophy,Metaphysical Modality and Counterfactual Thinking”,Proceedings of the Aristotelian Society, CV: 1–23.
• –––, 2004b, “Philosophical‘Intuitions’ and Scepticism about Judgement”,dialectica, 58: 109–153.
• –––, 2008,The Philosophy ofPhilosophy, Malden, MA, et al: Blackwell.
• –––, 2009, “Replies to Kornblinth, Jacksonand Moore 1”,Analysis, 69: 125–135.
• Wilson, James, 2016, “VII–Internal Validity andExternal Validity in Thought Experiments”,Proceedings ofthe Aristotelian Society, CXVI: 127–152.
• Wiltsche, Harald, 2018, “Phenomenology and ThoughtExperiments: Thought Experiments as Anticipation Pumps”, in M.T. Stuart et al. (eds.),The Routledge Companion to ThoughtExperiments, London and New York: Routledge, 342–365.
• Winchester, Ian, 1990, “Thought Experiments and ConceptualRevision”,Studies in Philosophy and Education, 10:73–80.
• Witt-Hansen, Johannes, 1976, “H. C. Orsted, Kant and TheThought Experiment”,Danish Yearbook of Philosophy, 13:48–56.
• Ylikoski, Petri, 2003, “Thought Experiments in ScienceStudies”,Philosophica, 72: 35–59.
• Young, Garry, 2013,Philosophical Psychopathology: PhilosophyWithout Thought Experiments, Hampshire, UK: PalgraveMacmillan.
• Yourgrau, Wolfgang, 1964, “On the Logical Status ofso-called Thought Experiments”,Proceedings of the XthInternational Congress of the History of Science, Paris: Herman,359–362.
• –––, 1967, “On models and thoughtexperiments in quantum theory”,Monatsberichte der DeutschenAkademie der Wissenschaften zu Berlin, 9: 886–874.

Academic Tools

	How to cite this entry.
	Preview the PDF version of this entry at theFriends of the SEP Society.
	Look up topics and thinkers related to this entry at the Internet Philosophy Ontology Project (InPhO).
	Enhanced bibliography for this entryatPhilPapers, with links to its database.

Other Internet Resources

• Thought Experiments, entry in theOxford Bibliographies, by James R. Brown andMichael T. Stuart.
• Thought Experiments and Religion, entry in theOxford Encyclopedia of Religion, by YiftachFehige.
• Goodies, a collection of intriguing questions in the philosophy of science,some about thought experiments, by John Norton.
• An Interactive Version of Thomson’s violin thought experiment.
• Six famous thought Experiments Explained Quickly, a video tutorial.
• Ethical Thought Experiments Like the Trolley Dilemma, a video tutorial.

Related Entries

Descartes, René |Galileo Galilei |intuition |Leibniz, Gottfried Wilhelm |Mach, Ernst |moral psychology: empirical approaches |Platonism: in metaphysics |rationalism vs. empiricism