The Ethics of Paranormal Investigations

Exorcising the Ghosts from Ghost Hunters

Samuel Zinaich, Jr., Ph.D.

Purdue University Calumet

In this essay, I will analyze the approach used by the members of The Atlantic Paranormal Society (hereafter TAPS) dramatically portrayed in the popular television show Ghost Hunters. I will proceed by giving a general description of their methods. Next, because the members of TAPS believe their methods are scientific, I will then try to show what methods they use that may be considered scientific. After that, I will set out my arguments against their approach by showing that, although the members of TAPS have attempted to scientize their investigations, their approach falls short of a crucial element needed to be judged truly scientific. Finally, I will consider two objections to my critical remarks about their methods.

Before I begin my analysis, I would like to make some amiable remarks. Although, my family and I don’t watch too much television—most of what’s on is so absurd and ridiculous, one of the shows we like to watch is Ghost Hunters. It is enjoyable for many reasons. The first reason is that, like most people, we enjoy shows that spotlight hauntings and scary—occult—situations. In fact, we’re big fans of the science fiction and especially science fiction films. Second, we also enjoy watching the show because of the personalities of the members of TAPS. There is a refreshing sense of professional integrity among the members of TAPS. A corollary to the last point is the fact that TAPS does not charge for any of the investigations they perform. Finally, unlike many of the other television shows that draw attention to (alleged) hauntings, the members of TAPS (especially the founding members, Jay Hawes and Grant Wilson) typically maintain a sense of skepticism about the claims they investigate. That said, there are times when the main members of TAPS declare that a structure is haunted.[1] With this in mind, I will now move on to my exorcism.

From what I have gathered from the television show and the TAPS website, there has emerged a theory (or system of beliefs) about the phenomena of hauntings.[2] There are several points to underscore about the theory. First, there are two kinds of hauntings: active hauntings and residual hauntings. Because the show rarely explores residual hauntings, I will not discuss it at all except to briefly illustrate it. According to the theory, a residual haunting is the aftereffect of the presence of a spirit that remains behind, and the aftereffect repeats itself analogously to a video set on continuous playback.

Active hauntings point to the existence of a spirit, and according to the theory, spirits can manifest themselves in a variety of ways. First, they can manifest themselves visually in terms of vaporous partial or full body apparitions, orbs of lights, and as small (opaque) beings called shadow people. Next, they can manifest themselves in terms of voices and other kinds of noises. Thirdly, they can manifest themselves in terms of smells. The odors are often described as obnoxious and even sometimes pleasant. After that, spirits can, on certain occasions, manifest in physical ways. For example, they can move objects, touch people, create cold spots, generate thermal images, and create electro-magnetic fields.

The theory of hauntings also maintains a difference between the kinds of spirits in active hauntings. There are intelligent hauntings, poltergeist phenomena, and demonic activity. Intelligent hauntings point to spirits who are aware of their surroundings. Although they are limited in what they can do, they are not confined to one spot—they can move around freely. Additionally, these spirits typically manifest themselves between the hours of 11 PM and 4 AM. Poltergeists are regarded as noisy spirits. The manifestation of poltergeists typically begins with knocks and bangs, and, as the theory maintains, the manifestations may turn intense and on rare occasions the poltergeist will cause furniture to slide across the room and shake beds. According to the TAPS website, the poltergeist is typically a female spirit in her teen years. On rare occasions, demons are responsible for hauntings. Demons are considered to be inhuman spirits. Demons manifest themselves through abnoxious smells, e.g., rotting flesh or sulfuric acid. They are also known to let loose growls, and they make contact by pushing, shoving, hitting, and even scratching.

The question of what means are available to document hauntings is an interesting aspect of the theory of hauntings and the TAPS investigations. Unlike many groups that investigate hauntings, the TAPS utilize various devises to perform an investigation. This is important to note because TAPS rarely employs the help of (suspicious) individuals who consider themselves mediums or psychics. Rather, TAPS attitude seems committed to the value of documentation. For example, there appears to be five devises used for investigation. First, in order to document visual manifestations and to document some of the physical ways spirits may manipulate the environment, TAPS uses live video feeds and cameras. In order to document noices and electronic voice phenomena, TAPS members make use of tape recorders. Thirdly, to measure cold spots that spirits create, TAPS uses handheld thermometers to measure the ambient temperature of the rooms in structures. Fourth, a thermal imagining devise is used to detect thermal images. And finally, TAPS attempts to document electro-magnetic fields created by spirits by using electro-magnetic detecting devises.[3]

The question now to discuss is whether the methods TAPS employs constitutes a scientific approach to the investigation of a haunting. Although what makes an approach like TAPS scientific or not is too complicated to fully explore in this paper, I will focus on the crucial role that observation plays in the role of a theory that is considered scientific. Here I will follow Gilbert Harman’s discussion of the role of observation in science.[4]

According to Harman, observation plays a crucial role in science. There are a number of features to underscore. I will (briefly) summarize these points and then apply it to the discussion of TAPS.

Before I unfold Harman’s point of view, I will make some preliminary remarks in order to clarify my own understanding of Harman’s position. Let’s assume the existence of some sort of event,[5] and that this event is describable in terms of several features which make up the event.[6] I am not using the term ‘event’ in any deep or technical sense. I mean to use it in the most familiar sense of the term. Thus, at least for the purposes of this paper, when I use the term ‘event’ I am merely referring to the common experience of having a dinner with your friends or family, walking a dog, watching a movie, and so forth. Again, just for the purposes of this paper, I want to make a qualifying remark, viz., that during the event—whatever affair or occurrence we are talking about—that at least one of the features that make up the event is observable and identifiable by an individual who is part of the event or who is watching the event take place.[7] One final assumption is necessary. A theory is describable as a system of beliefs. The beliefs or concepts constitute the theory itself, and the concepts of the theory are used to explain the occurrence of our observations when they are activated by an event. [8]

The first detail is this: According to Harman, there are no pure observations.[9] Thus, all observations are always theory laden. This means that we observe the feature of an event because of some theory we hold.[10] In particular, Harman adds that forming an observation (or a belief) is possible because we are actually employing the concepts that make up the theory.[11] Perhaps this will illustrate what he means. My son around the age of two observed a Pomeranian dog one day which was loose in our yard. Up to this point, his only experience with dogs was primarily with our two black Labradors. In my conversations with him about the dogs, he could correctly identify both of our Labradors as dogs. When he saw a stray Pomeranian dog, he turned and said to me, “Daddy, what is that thing in our yard?” I responded to him by stating simply that it was a dog. His response was a look of puzzlement—one not entirely unjustified. After our conversation, I thought about his question, and I realized that because his theory of what dogs are was too narrow, i.e., there were not enough concepts in his theory, he was unable to identify that creature according to its species. Harman explains his point in this way:

To recognize a child as a child is to employ, consciously or unconsciously, a concept that is defined by its place in a framework of the stages of human life. Similarly burning is an empty concept apart from its theoretical connections to the concepts of heat, destruction, smoke, and fire.[12]

So, according to Harman, if an individual understands some of the relevant concepts that make up a theory, then by means of at least one of the concepts of a the theory, our individual is able to identify at least one of the features of an event. This leads us to another point to underscore. An observation that is made about one of the features of an event may be used to confirm or disconfirm a theory that is being used.[13] Although Harman does not discuss what he means by the notion of ‘confirmation’, I do not think he implies that the observation confirms the truth of the theory. First, there is nothing in the text to support such a view—Harman never uses it in this way. Second, the publically available meaning of confirmation points to something different as well. Confirmation typically means to strengthen or to the removal of doubts. Essentially what this means is that if an individual observes a feature of some event, and the feature is consistent with the theory, then the observation of the event strengthens the theory or it removes doubts about the theory in question.[14]

That said, Harman quickly adds a qualifying point. The opposite is also true, viz., if an individual observes a feature of an event and it is inconsistent with the theory, then the observation disconfirms the theory.[15] Harman correctly remarks that before we conclude that an observation of an event disconfirms a theory, we must first decide whether the observation must instead be rejected. I would also add that we should not take Harman to mean that one inconsistent observation will completely undermine a theory. Certaintly, an inconsistent observation (under our understanding of confirmation) raises doubts about the theory. But I imagine that it would take more than one inconsistent observation to completely disconfirm a theory. Thus, it stands to reason that the strength of a theory includes the ability to confirm and disconfirm what features of an event count as confirmation and what features may or may not confirm a theory.[16]

But now the question is this: how does this apply to the methods TAPS employs? In the simplest terms, TAPS methods and approach stem from a theory that minimally explains the observations that take place in the alleged hauntings they investigate. For example, let me describe a typical investigation of the members of TAPS. After receiving permission to investigate a house or some sort of structure, an assessment of the kinds of activity in the house is gathered, then according to the concepts of the theory of haunting, the appropriate kind of equipment is placed in the rooms of the house. For example, if a room is allegedly a hot spot for physical movement, e.g., chairs moving, etc., cameras and live video feeds are used. If a room is allegedly known for unexplainable noises, then recording devises are placed in the room, and so forth. When the equipment is set up, then the occupants of the house are normally required to leave and no one but a few select members of TAPS is allowed to be in the house.

Again, as the show sometimes demonstrates, let us suppose that a member observes a sound that sounds like a voice. On the occasion that a voice phenomena is picked up by the recording devise and the voice is clear enough to understand,[17] then the voice confirms their theory, and in fact this is correct.[18] As the show occasionally indicates, voices, which are not detectable by the normal hearing range of a human being, counts as confirmation of the theory of hauntings. To their credit, there is usually a (minimal) attempt to substantiate the confirming instance by eliminating a non-supernatural explanation. Additionally, on the occasion that other noises are detected by the recording devices, e.g., a loud bang, the investigation team attempts to find a reasonable natural explanation for the bang. When a natural explanation cannot be found, then, on their view, the noise counts as confirmation of the theory. I think this is also correct. Moreover, when an object is moved and detected by a live video feed, barring the obvious staging of the event by unknown individual, the event confirms the theory of haunting. When an electromagnetic field is detected, and when there is no obvious natural source for the field, e.g., when there are no electric junction boxes, this also counts as confirmation of the theory. Finally, cold spots and thermal images are also treated in the same way.

The next point Harman discusses is more difficult to articulate, but that said, it points to the problematic aspect of the methods and approach of TAPS. Observation also plays another role in science, and he adds that it is a role that is a distinguishing aspect of a scientific theory.[19] In short his point is this: in science it is necessary to make assumptions about the facts you are trying to explain in order to make clear the occurrence of the observations that support a scientific theory.[20] Why is this important? Harman discusses two reasons. First, Harman adds that this is evidence for the theory to the extent that the theory can explain the existence of the features of the event better than competing theories can.[21] This is important because the observations of an individual are highly influenced by the psychological states of the individual, we have to rule out psychological states that are not relevant. Here’s what he says:

But, if his having made that observation could have been equally well explained by his psychological set alone, without the need for any assumptions about the features, then the observation would not have been evidence for the existence of [the fact to be explained] and therefore would not have been evidence for the theory.[22]

Harman underscores this point again by drawing out the critical role psychological state plays in science: “His making the observation supports the theory only because, in order to explain his making the observation, it is reasonable to assume something about the world over and above the assumptions made about the observer’s psychology.”[23]

Harman’s second point is very subtle, and related to the first reason: “[T]he observation of an event can provide observation evidence for or against a scientific theory in the sense that the truth of that observation can be relevant to a reasonable explanation of why that observation was made.”[24] Here’s what I think he means. Although a theory tells us what is and is not relevant about the features of an event, it is only until we can support the truth of our observations (again by ruling out other competing theories and especially ruling out our tendency to rely too heavily on the reliability of our own psychological states) that we can confirm the truth of the theory we employ.[25]

What is the implication for this second point for the theory of hauntings? As Harman indicates above, in order to confirm the truth of our observations, it is necessary to make assumptions about the facts we are trying to explain in order to make clear why our observations are not explainable by another theory or by our own psychological states. So, in this case, what are the assumptions that we need to make about spirits to explain objectively our observations? Although this is never discussed by the members of TAPS, I have come up with a list of four assumptions for active haunting that I have gathered from the show and website: 1. Spirits are a concentration of power. 2. They are conscious and possess intentions. 3. Spirits are disembodied, and, 4. Spirits can draw upon the physical sources of energy in the environment in order to manifest themselves. For example, in order for a spirit to manifest itself, i.e., to reflect light in order to be seen, to create sound waves in order to be heard, to create detectable smells, or to move physical objects, it must draw energy from known physical elements of our universe. As the show sometimes indicates, when a spirit intends to make itself known, it may drain the power of batteries (often the batteries of the cameras being used by investigation team), it may use the surrounding molecules of the environment, which the theory indicates is responsible for cold spots and is answerable for unexplainable thermal images and unexplainable electromagnetic fields.

Although more discussion is needed, I will now (briefly) draw out the problematic aspects of the theory of hauntings used by TAPS. The first, and perhaps the most knotty aspect of the theory of hauntings, is the first assumption. The power of spirits, i.e., what they are essentially, is never discussed or if it is, the topic is glossed over very fast without much intellectual discussion. But why is this problematic? The answer is that before we can explain why we make the observations we do during an (alleged) haunting, we need to know what spirits are. Unfortunately, as far as I know, no one knows the fundamental make up of spirits! And this leads me to my second criticism. Without knowing the elementary composition of spirits, the fourth assumption is radically undermined, i.e., we cannot know how it is possible for a spirit to draw upon the physical sources of energy contained within our environment.[26]

Before I end this discussion, three objections to my view should be considered. First, someone might object that my discussion does not rule out the existence of spirits. I agree because my point is slightly different. I just want to show that the approach and the methods used by the members of TAPS, contrary to their own assessment and the assessment of others, is not a truly scientific approach. That said, without knowing what spirits are, I see no reason to believe in them. For example, when I hear or someone tells me that they saw a spirit or had some sort of supernatural experience, I take my queue from David Hume and reason in this way: Either they are attempting to deceive me, they have been deceived by someone else or there really are spirits.[27] Since spirits cannot be scientifically demonstrated to exist—and since I have never seen such activity—they must be attempting to deceive me or they have been deceived by someone else.

The next objection is this: Although the causal chain from the assumptions of what spirits are to the observations of spirits is broken, the probability of the supernatural explanation is extremely high, and therefore reasonable to believe, because baring the staging of the events showcased on Ghost Hunters, a chair cannot move itself without some sort of supernatural intervention. Likewise, a voice detected by an electronic recording voice devise (i.e., a voice not measurable by the normal hearing range available to humans) or detectable audibly (and recorded) by an investigator with no obvious natural source, what other explanation is reasonable except a supernatural explanation? The point is this: if a supernatural explanation clarifies the abnormal occurrence in a haunting, then the abnormal occurrence would not occur without the supernatural event occurring first.

I take this to be an interesting objection to my view because, like Harman’s own view, it turns on a version of what is called “Inference to the Best Explanation.” Although this idea is initially credited to C. S. Peirce, Harman is credited mainly as the scholar who has attempted to explain what it means and to explain its uses.[28] In rough terms, the rule of Inference to the Best Explanation states that if we have evidence E, and, say two competing hypotheses about E, viz., H and H’, we should infer the truth of H if it is a better explanation of the E than H’.[29] Here’s what I take to be the structure of Harman’s rule:

1. If H explains E, then the assumptions about the feature of E explain the occurrence of the observation that one makes about the features of E. (The Observation Test)

2. H explains E better than any alternative H that satisfies the Observation Test. (The Competing Hypothesis Test)

Therefore,

3. H is the best explanation of E.

I mention this issue because the second objection, as I mentioned earlier, employs a rule very similar to Harman’s rule. That said, there is an important difference between the two. Whereas Harman employs The Observation Test, the objection above employs what I call the Counterfactual Test. Here’s an outline of the rule:

1. If H explains E, then E would not occur without H occurring first. (The Counterfactual Test)[30]

2. H explains E better than any alternative H that satisfies the Counterfactual Test. (The Competing Hypothesis Test)

Therefore,

3. H is the best explanation of E.

In the simplest terms, the objection argues—as do the members of TAPS on different occasions—that it is perfectly reasonable to believe that on certain occasions, the existence of a spirit is the best explanation of, say a chair moving, because the existence of a spirit explains the chair moving better than any alternative explanation that satisfies the Counterfactual Test. That is, since there isn’t a better explanation, the chair would not have moved without the presence of a spirit moving it.

I have three responses to this objection. First, the Counterfactual Test is not satisfied because, as I argued earlier, without knowing what spirits are and how they can manipulate the environment, we cannot know whether the chair would not move without the presence of a spirit to move it. Second, even if we assume for the sake of argument that the Counterfactual Test is satisfied, there are many reasons to believe that the Competing Hypothesis Test is not satisfied. First, if we assume for the moment that the members of TAPS are honest people, their methods are often less than thorough. Two problems are worth noting. I have noticed that, on a number of separate occasions, the members of TAPS fail to thoroughly investigate an event that they declare to be a haunting or they fail to take seriously the importance of trying to reproduce the event they eventually declare to be a legitimate haunting.

One example of the first problem is illustrated in the episode where a light house in St. Augustine, FL, was investigated. Live camera feeds, focused on the internal circular stairwell and ceiling of the lighthouse, appeared to capture the presence of some sort of apparition on the stairs far above the members of TAPS. Several times, the apparition looked as if it had it hands on the railings as it looked down at the TAPS crew. The team went up to the very top of the lighthouse, but didn’t find anything. But here’s a simple point they failed to do: the members of TAPS did not dust for finger prints where they saw the apparition looking over the railing. In fact, there was no attempt to collect any kind of physical evidence to rule out the possible of an individual staging the event. There are other examples like this, but they are too numerous to mention.

The second problem is just as serious as the first problem. That said, on several different occasions, the members of TAPS attempt to reproduce an event they are investigating in order to make sure the event is better explained in just natural terms. For example, when the occupant of a house complains about loud knockings or other kinds of noises, there is usually an attempt to see if there are loose pipes underneath the floor. On several episodes, the members of TAPS ruled out suspicious noises by locating loose pipes that shake when fluid passes through them. While such a skeptical attitude is commendable, the investigations are hardly consistent. For example, on one occasion, when a short image was detected by the thermal imaging device—roughly an image thought to be four feet tall—there was no attempt to reproduce the event. Although there was an attempt to rule out the presence of some sort of animal, the members of TAPS did not think that there was any sort of natural explanation available.

My last response to the objection points to whether the Counterfactual Test is an acceptable rule to use in alleged haunting investigations. Although it may be useful in certain contexts—even contexts like the one under consideration—it does not show that supernatural claims are empirically testable in the relevant sense under question. Let me briefly summarize my point. On the one hand, the Counterfactual Text can be used to point to features of a haunting that can explain things that can be observed. On the other hand, the Counterfactual Text cannot do the following: it cannot explain why the observations made during a haunting are the result of the causal interaction between the properties of a spirit and our sense perceptions. It is the second issue that has to hold if supernatural claims are to be tested in all the ways scientific claims can be tested.

The final objection to my view is this. Harman argues, and I think appropriately, that in science it is necessary to develop a set of assumptions about the features of an event in order to explain the occurrence of the observations an individual makes about one or more of the features that make up the event. As he insists, this is the distinguishing aspect of a scientific theory. Unfortunately, as the objection goes, if this is true, then most of what we take to be (legitimate) science will not count as science per se. The reason is that, as it was related to me, it is not untypical for a scientist to begin an empirical research project without any assumptions about the features of an event that is the focus of the project. Again, as it was communicated to me, it is often the case that a scientist will begin an empirical research project without knowing what it is that they want to discover.[31] I have responses to both points.

First, the latter point is just plainly false. I should think it very unlikely that scientists begin an empirical research project without knowing what they want to discover. On the contrary, an empirical research project always begins with a hypothesis that states what the scientist will attempt to prove. Second, the former point will not succeed either. The reason is that a legitimate scientific research project will not begin until a set of assumptions is developed about the features of an event that is the focus of the research project.

In the end, I want to draw out four conclusions. First, without any reasonable discussion of the first and fourth assumptions discussed earlier, the observations TAPS team members make cannot be taken as evidence for the truth of the theory of hauntings. Thus, TAPS’ methods fail the test of a legitimate scientific approach to hauntings. Second, even though I have tried to be charitable to the members of TAPS by listing some of the assumptions associated with spirits—a discussion I believe is entirely missing from their literature—it would seem that we need only make assumptions about the psychology of the people who are making observations about the existence of spirits and hauntings. Again, if I am right about this second point, this also points to the mistake of calling their methods scientific. The explanatory chain from what spirits are to the observations that are made is broken. Fourthly, the Counterfactual Method typically employed by paranormal investigators cannot give them the empirical evidence they do desperately desire. Finally, this paper points to what is needed for a scientific approach to hauntings, a discussion largely missing in the popular account of hauntings, and, I dare say, a discussion that will never emerge.

[1] While I appreciate professional integrity the member of TAPS and their (albeit limited) skepticism, my intuitions tell me that the members are not prepared to reject their own theory in the light of defeating evidence.

[2] The show has investigated open areas and forests as well. For the purpose of my discussion, I will focus simply on the investigations that show case structures.

[3] As far as I know, the members of TAPS do not have any devises to document obnoxious or pleasant smells except for their own antidotal evidence.

[4] Gilbert Harman, “Ethics and Observations,” in Geoffrey Sayre-McCord (ed.), Essays on Moral Realism (Ithaca and London: Cornell University Press, 1988), 119-124. This essay was originally published as chapter 1 of Gilbert Harman, The Nature of Morality (New York: Oxford University Press), 3-10.

[5] Event = {e}.

[6] Event {e} = features {f₁, f₂, . . . , f_n}.

[7] S observes feature {f₂} of event {e}.

[8] Theory T = concepts {c₁, c₂, . . . c_n}.

[9] Harman, “Ethics and Observation,” 120.

[10] If S observes a feature {f₂} of an event {e}, then S hold some sort of theory T.

[11] If S has an observation (or belief) about a feature {f₂} of an event {e}, then S understands some of the relevant concepts, say, {c₂}, of theory T.

[12] Harman, “Ethics and Observation,” 121.

[13] Ibid.

[14] If S observes a feature {f₂} of an event {e}, and feature {f₂} is consistent with theory T, then the observation of {f₂} confirms theory T.

[15] If S observes a feature {f₂} of an event {e}, and feature {f₂} is inconsistent with theory T, then the observation of {f₂} disconfirms confirms theory T.

[16] Harman, “Ethics and Observation,” 121.

[17] I mention this point because the (so called) voices detected by the recording devises are not always understandable. They are often garbled. I have noticed that occasionally the members of TAPS are guilty of fudging an understanding of the meaning of the voice phenomena in much of the same way that people are guilty of seeing figures, e.g. rabbits, flowers, etc. in the formation of clouds.

[18] This is just a simple case of modus ponens: 1. If S observes a feature {f₂} of an event {e}, and the feature {f₂} is consistent with theory T, then the observation of {f₂} confirms theory T. 2. During the event of an investigation, S (a member of TAPS) records a (understandable) voice phenomena and the electronic voice phenomena is consistent with the theory of haunting. Therefore, 3. The voice phenomena is confirmation of the theory of haunting. All of the other observations by the members of TAPS follow the same method of reasoning.

[19] Harman, “Ethics and Observation,” 121.

[20] In science it is necessary to develop a set of assumptions, e.g., {a₁, a₂, . . ., a_n} about the features {f₁, f₂, . . ., f_n) of event {e} in order to explain the occurrence of the observations S makes about one or more of the features {f₁, f₂, . . ., f_n} that make up the event {e}.

[21] If the assumptions {a₁, a₂, . . ., an} about the features {f₁, f₂, . . ., f_n} of the event {e} best explains the occurrence of the observation S makes about the features {f₁, f₂, . . ., f_n} of event {e}, then this counts as evidence for the truth of theory T.

[22] Harman, “Ethics and Observation,” 122.

[23] Ibid. If the observations of the features {f₁, f₂, . . ., f_n) of the event {e} can be explained without any of the assumptions {a₁, a₂, . . ., a_n} about the features {f₁, f₂, . . ., f_n} of event {e}, then the observations of the features {f₁, f₂, . . ., f_n} of event {e} do not count as evidence for the truth of theory T.

[24] Ibid., 122.

[25] If S makes an observation about event {e} and the observation is determined to be true, i.e., it is determined that the assumptions {a₁, a₂, . . .,a_n} made about the features {f₁, f₂, . . ., f_n} of event {e} best explain the occurrence of the observations S makes about the {f₁, f₂, . . ., f_n} of event {e}, then the truth of the observation provides a reasonable explanation of why the observations of the event {e} was made.

[26] The argument is this: 1. If the observations of the features {f₁, f₂, . . ., f_n) of the event {e} can be explained without any of the assumptions {a₁, a₂, . . ., a_n} about the features {f₁, f₂, . . ., f_n} of event {e}, then the observations of the features {f₁, f₂, . . ., f_n} of event {e} do not count as evidence for theory T. 2. The observations of the TAPS members made during an investigation can be explained without any of the assumptions about spirits. Therefore, 3. The observations made by the members of TAPS during an investigation do not count as evidence for the truth of theory of hauntings.

[27] David Hume, An Enquiry Concerning Human Understanding, Eric Steinbery (Ed.) 2^nd Edition (Indianapolis and Cambridge: Hackett Publishing Company): 77.

[28] Bas C. van Fraaseen, The Scientific Image (Oxford: Clarendon Press, 1980): 19.

[29] Ibid.

[30] Harman attributes the Counterfactual Test to Nicholas Sturgeon. Harman critiques and discusses the limitations of the Counterfactual Test in Gilbert Harman, “Moral Explanation of Natural Facts—Can Moral Claims be Tested against Moral Reality?” The Southern Journal of Philosophy, vol. xxiv (1986 Supplement), 57-68. My own discussion of the Counterfactual Test relies heavily upon Harman’s assessment.

The Counterfactual Test relies upon a counterfactual condition or subjunctive conditional which is a statement indicating what would be the case if its antecedent were true. This is contrasted with an indicative conditional, which indicates what is (in fact) the case if its antecedent is (in fact) true.

[31] Both points were related to me by my colleague, Associate Professor Eugene Schlossberger.