Is Psychology a Science? (Column 405)
This column is dedicated, with appreciation, to the friends in the Beit Midrash on WhatsApp
who asked me years ago for this discussion (Shai, Avrami, and others).
In the previous two columns I dealt with Popper’s criterion for the scientific status of a theory. In Column 403 I mentioned psychology as an example of a field whose scientific standing is dubious, a point that Popper himself already noted. In the past I was asked to discuss this topic, and I decided this is the place to address it, even if briefly.
Preliminary Survey
According to Popper, the scientific status of a theory is determined by its being empirically falsifiable. We saw that this is not a simple criterion, but it is accepted as a minimum criterion (that does not exhaust everything). Among other things, I mentioned two problems in this simplistic formulation:
- Every falsifiable claim about the world is, according to it, considered scientific. The claim that more than 100 people live in our neighborhood, for example, is indeed falsifiable, but it is hard to treat it as scientific. Admittedly this is not terrible, since I already pointed out that this is only a minimum criterion that does not exhaust the matter. Beyond falsifiability, a scientific theory is also supposed to be general (not about a particular neighborhood) and interesting in some sense.
- Even scientific claims par excellence are not truly falsifiable in the strict logical sense. It is always possible to patch the theory ad hoc and thus explain any deviation from it and save it from experimental refutation. One can, of course, also attribute a deviation to an experimental error or an unforeseen side effect. The process of refutation is not as unequivocal as Popper presented it. Thomas Kuhn already noted that refutation of a theory is a sociological procedure no less than a logical one. Isolated refutations do not topple a well-established scientific theory. The scientific community is supposed to arrive at the conclusion that the existing paradigm is challenged from too many directions and there is no point in rescuing it ad hoc any longer. Only in such a state is the conclusion to replace the paradigm, namely, to seek a new theory.
What about psychology? Seemingly we are dealing with factual claims about the world, and they are also falsifiable. For example, the claim that frustration leads to aggression is a factual claim about human beings and ostensibly can be tested empirically. On the other hand, it is clear that it has many exceptions, and the full picture is never captured within it. Therefore, even if we encounter some behavior that deviates from this theory, we will usually not abandon it. We can always claim that there are additional aspects that were not taken into account, and usually we won’t be able to point to them. In such cases many will be satisfied with the assertion that the human being is a complex creature, or that there were additional influences that were not accounted for.
Consider a very frustrated person who is provoked and does not respond with aggression. Has this refuted the frustration–aggression link? Certainly not. We will say that he overcame his tendency to respond aggressively, and what enabled him to do so were environmental circumstances such as education, values, and the like. I think that in this sense no fundamental psychological law is truly falsifiable. I am not an expert, but it is hard for me to think of a psychological law that has been rejected due to empirical refutations. Various hypotheses have been rejected, but not a fundamental law (such as the connection between frustration and aggression).
It is important to understand that I do not intend here to argue against this mode of thinking. It is quite possible that there is a link between frustration and aggression and, at the same time, there are other parameters that can also influence (education, environment, values). [1] The discussion is not about whether it is true, but about the extent to which one can treat the connection between frustration and aggression as a scientific claim. In the end, even if it is true it does not give me unambiguous predictions about specific cases, but at most a general line of thought, and therefore it is also not falsifiable. Such a claim recalls the discussion of the ambiguous prophecies of the Oracle of Delphi that I conducted in the first column. We saw there that ambiguity and polysemy prevent quantitative predictions that are falsifiable, and therefore cannot be considered scientific claims.
Statistical Laws
One can circumvent this difficulty and say that the psychological law deals with groups of human beings and not with the individual. Indeed, with regard to an individual there may be deviations, but the fundamental statement concerns statistics of behavior measured over groups of people. For example, if we succeed in determining that in a group of a thousand people whose frustration level is X (assuming frustration levels can be measured. We must remember these are determined not only by stimuli but also by personality and character), Y percent of them will respond with an aggression level Z. This is already a quantitative claim that, in principle, can be subjected to an experimental test. We take a thousand people, place them in a frustrating situation, and examine how many respond aggressively (which will also be measured). If we find that Y percent responded aggressively, we have corroborated the theory. If not—then we have refuted it. To test the theory better, it is, of course, advisable to take in parallel another control group of a thousand people who are not in a state of frustration and see whether they do not respond in the same way. Or to vary levels of frustration and check for changes in levels of aggression, and so forth. In this way one can get closer to a general and quantitative law, and a psychological law of this kind will ostensibly be falsifiable.
In this sense, the psychological law differs from a law in physics, chemistry, or biology. In the natural sciences the law states what will happen in every situation, and if there is a deviation we can usually point to its cause. A good law in the natural sciences gives us a good forecast even for what will happen in a particular experiment and not only in an experiment on a large group of cases as in psychology. The psychological law, by contrast, is only a statistical law. I will note that medicine is an exception in this regard, even though it is based on chemistry and biology. Medicine deals with the human body, and this is a very complex entity that is influenced in various ways also by the human psyche. In medicine there are many laws that are statistical, and therefore one can find quite a few deviations from the laws (sometimes called medical miracles). And yet there are enough components of chemistry and biology there to allow us to make not-bad quantitative predictions. If so, medicine is indeed a field whose characteristics are closer to psychology, but it is still not quite there. [2]
And I have not yet mentioned the free will we have, which ostensibly neutralizes in principle the possibility of arriving at deterministic laws concerning human behavior. A person always has the possibility of overcoming and choosing otherwise than what the psychological law predicts. [3] I did not enter into this here because we can conduct our discussion even on the assumption that the human being is a complex deterministic creature. [4] On the practical level, this complexity prevents us from arriving at falsifiable claims in this field even without assuming free will.
Take as an example the field in physics called statistical mechanics. There too we are dealing with statistical laws that concern a system containing many particles. But those who are familiar with this field know that, bottom line, it gives unambiguous quantitative predictions about what will happen in the laboratory. The law of large numbers determines what will happen and what the statistical significance bounds are for the scientific prediction. Usually the deviations are completely negligible, and to wit: thermodynamics (the theory of heat), in its familiar macroscopic form—no one would imagine it has statistical components—turned out to be a consequence of calculations in statistical mechanics. This is a salient example of the difference between saying that a law has a statistical character in physics and making a similar statement in psychology.
And still, despite all the differences, as I wrote above, a precise quantitative formulation of a statistical law can be considered a scientific, falsifiable claim. It is a statistical scientific law, and even if it cannot be refuted by an experiment on a single case, it can be refuted by a statistical experiment on a large group of subjects. So-called “scientific psychology” deals with laws of this type. It deals with average statistical claims about human beings and not with the behavior of an individual.
These are indeed claims about the world, and they are also falsifiable in the statistical sense, but I think very few, if any, laws in psychology can be formulated quantitatively and precisely such that they can be refuted even at the statistical level. In principle this may be possible, but in practice, in my impression, it is almost nonexistent. One can point to processes and influences of phenomenon X (frustration) on phenomenon Y (aggression), but usually this is not a quantitative formulation that can be subjected to a decisive statistical refutation test. There is a general statement that in people in a state of frustration the degree of aggression in responses will increase to some extent. You understand that it is quite hard to refute such a general and amorphous claim. Of course, it depends on the type of people and the environment, and there are always excuses as to why it didn’t work here and what additional influences interfered with the experiment, and the like. Incidentally, the success of a statistical experiment can certainly corroborate such an amorphous psychological theory, but failure cannot refute it. I already pointed out this asymmetry in the first column (in the discussion on the attitude toward biblical prophecies, which can be corroborated but not refuted).
The Importance of a Conceptual System: Abduction
Psychology is a systematic attempt to understand human behavior. If we liken it to physics, it is an attempt to arrive at general laws that describe our behavior in different situations. The psychological theory, similar to the physical theory, builds a theoretical conceptual world that is not always accessible to direct experimentation except through its practical implications. In physics the theory includes theoretical entities we cannot observe, and the theory that deals with them is tested via its practical predictions. One can deduce from the theory the results of an experiment to be conducted on objects or events that are accessible to our measuring instruments, and through that examine its truth. We arrive at this conceptual system via a process now called “abduction,” which describes the move from the cases we observed to the theoretical conceptual system that explains them (it is a kind of induction, but not from one case to a set of cases, rather from the cases to the theory that explains them).
A similar enterprise is undertaken in psychology as well. There too abduction is used to build maps of the psyche and different forces within it, some of which are not accessible to direct measurement (psychoanalysis), and they are supposed to be tested through predictions they make about behaviors subject to observation and measurement. These maps contain concepts and forces that have no direct empirical basis and are also not accessible to direct measurement. And yet, ostensibly, they can be examined via their behavioral implications. Therefore, the accusation by Popper and others that psychoanalysis is unscientific because it creates a whole conceptual world not accessible to observation is not, by itself, correct.
The more significant problem of psychoanalysis is that it is difficult to derive from the theory a concrete prediction that can be tested in an experiment and measurement in a decisive way. That is, a result that cannot be explained without the psychoanalytic theory. Psychoanalysis appears more like an ad hoc construction of a theory that will fit all the experiments we have done (or the experiences we have had) and place them into a single conceptual framework. Almost nothing can truly refute it, since it is very difficult to derive from this theory a prediction that can be subjected to a decisive empirical test. It is well known that there is great criticism of psychoanalysis even among psychologists, and many of them see it as a kind of baseless and valueless magic. But none of this moves almost any psychoanalyst from his stance and faith in the field.
It seems to me that the main value of psychoanalysis, to the extent there is one, lies in providing a conceptual framework for psychological discussion, not in understanding how people actually function. Psychoanalysis contributes to our understanding not necessarily through its claims but mainly through the conceptual system it created. I have already written here that Freud’s main contribution (some would say his genius) is not in understanding human behavior, but in creating ex nihilo a new conceptual system that allowed us to discuss that behavior. Within it one can formulate claims and also disagree with Freud himself. Think about the situation two hundred years ago, when we had nothing, and now we want to start discussing psychology. How can this be done? How will we formulate our claims? Frustration and aggression are concepts that existed, but if we want to explain the connection between them we must use a theoretical conceptual system (as in physics). Before Freud established his conceptual system it was not possible to conduct such an abductive discussion. It seems to me this is Freud’s main contribution. I will note that there are also several of Freud’s basic assumptions that are indeed accepted by practitioners in the field to this day (see here, at the beginning of the article).
Between Theory and Practice
Beyond the attempt to describe human behavior, which can also be done via statistical laws, there is also the ambition to treat, with psychological tools, the problems of specific individuals. This is an attempt to use theoretical knowledge in psychology to apply it in treating people. Ostensibly this parallels the application of scientific knowledge in physics for technological purposes, such as producing or repairing some device, or understanding the trajectory of a particular particle in a particular state. In this matter the differences I have described so far come to the fore. Because psychological theory is statistical, it is very difficult, and perhaps impossible, to apply it and derive from it conclusions regarding a particular case. When we want to understand the behavior and mode of thought of a particular person, the use of general psychological claims seems invalid. We have no way to know whether this person indeed behaves according to the general statistical law or not (for the law does not hold for all human beings in all situations). Again, theoretical knowledge can give us general lines of thought (what a person is statistically expected to do in such a state), but this is not a simple application of general knowledge to a particular case (deduction) as happens in the natural sciences, for example.
We must therefore distinguish here between two different questions:
- Is psychological theory scientific? Is a theoretical claim in psychology falsifiable?
- Is the effectiveness of treatment methods scientifically grounded?
Up to this point in the column I dealt with question A. But question B can be asked and addressed independently. Suppose psychological theory yields falsifiable (statistical) claims. There is still room to ask whether the effectiveness of using it to treat people and improve their condition has scientific validity.
To answer this question it is important to sharpen it further. I am asking here whether there is valid scientific evidence for the efficacy of a particular type of psychological treatment, that is, whether it yields significantly better results, statistically speaking, than other treatments. I am also asking whether it was the practitioner’s disciplinary knowledge that made the difference. It may be that the practitioner uses good intuitions and experience and therefore succeeds, but that does not yet mean that what helped him do so is the psychological knowledge he acquired in his studies and training. It may be that his experience and studies exposed him to ways of thinking and various findings, and all this created in him a healthy psychological intuition that enables him to achieve therapeutic success—but it is still not an application of knowledge in its scientific (deductive) sense. One could argue that a person without psychological education, who has similar intuitions and experience, would achieve similar results.
What is nice is that regardless of our opinion on the scientific status of psychology (question A), these questions (of type B) are scientific questions that in principle can be examined experimentally. To do so, we must run a comparison of a particular treatment method applied to one hundred people in a given condition, versus another control group of one hundred people in a similar condition who did not undergo treatment, and preferably another control group of one hundred people in the same condition who underwent treatment by other methods (or by non-psychological methods, administered by practitioners who are not psychologists: coaches, sorcerers, or Rebbes). Such a scientific experiment would also neutralize the placebo effect (the contribution to healing from the confidence placed in the practitioner and his methods) that is to be expected in such situations. [5]
I suppose this won’t surprise you, but it turns out that at least until recently, such experiments were hardly conducted. The question regarding the efficacy of any treatment is sometimes considered almost a heretical question (as in any field where there is no way to validate something scientifically, the discussion shifts to believers and heretics, the wicked and the righteous, to slinging accusations, and so on). In my judgment, the reason for this is twofold: A. The fear that such an examination will not yield the hoped-for results (after all, not a few people make a living from this modern magic). For a similar reason we do not find (certainly among Jews) scientific experiments checking the impact of prayers or of a person’s righteousness on his health and chances of recovery. B. Practically, it is very difficult to run such an experiment. You need 300 people who are all in a similar psychological state and who all consent to the experiment, and you need precise measures of their mental state before and after treatment (measures of its effectiveness). And of course you need approval to conduct an experiment and to administer treatment by unlicensed and unauthorized actors, and the chance of obtaining that is not realistic. [6]
One can partially overcome the second difficulty by taking sufficiently large groups of subjects. The assumption is that statistically the differences between the groups will be small and will cancel out. Indeed, every person is different from his fellow, but among a thousand people there will likely be a similar distribution of people with certain problems and circumstances, so that overall the average result across all of them should give quite significant indications of treatment efficacy. Thus, for example, if we take one hundred people suffering from paranoid schizophrenia and give them treatment X, and in parallel another hundred who receive treatment Y, and another hundred who receive treatment from Oren Zarif or the Admor of Weisskvas, then it is clear that beyond schizophrenia each of them has his own additional files and special circumstances, but nonetheless, if there is a significant difference in results among the three groups, that will give a meaningful indication of the efficacy of the treatment. Therefore, in principle, such experiments can be conducted, though one should always remember that the attitude toward the results should be “respect but suspect.”
Some time ago I was sent a pair of articles by Eyal Pozniak (here and here. The links also appeared in note 1 of Column 403) that deal precisely with these questions, and from here on I will try to discuss his remarks briefly.
Psychoanalysis on Trial I
In his two articles Pozniak examines the effectiveness of psychodynamic treatment, i.e., a set of therapeutic techniques that developed out of psychoanalysis. He distinguishes well between the question of whether psychoanalysis is scientific and the question of whether the effectiveness of psychodynamic treatments can be examined scientifically, exactly as I have explained here. In the first article he examines psychoanalysis vis-à-vis the empirical approach in psychology, and in the second he moves to a scientific examination of the effectiveness of treatments.
In the first article he highlights the problematic nature of the field of psychoanalysis, which does not allow for the empirical testing of claims and their refutation in the Popperian sense. The conclusion that emerges from his words is that you will not find “scientificity” there in the sense of the natural sciences. There are various corroborations of general claims (such as the influence of the early stage of life on what follows, the importance of bringing the unconscious into consciousness, and the like), but even these foundations do not stand up to a decisive refutation test.
At the end of the first article he describes a shift among practitioners from an approach that places full trust in a particular psychoanalytic theory to a postmodern approach that views different approaches and theories as perspectives and ways of looking that can be helpful in treatment. It is no coincidence that he quickly moves to the scientific status of treatment methods and does not dwell overmuch on the scientific status of the theory itself. He concludes by saying that the effectiveness of the treatment methods, which is certainly measurable, repeatedly corroborates the validity of the theories. Even if you will not find corroboration for the validity of any one particular theory, here there is corroboration for psychoanalysis as a whole.
Psychoanalysis on Trial II
In the second article Pozniak turns to discuss the effectiveness of psychodynamic treatment. He opens by noting that Cognitive Behavioral Therapy (CBT) is considered the most scientific, evidence-based method in psychology. Therefore, his way of examining psychodynamic treatments is to compare their effectiveness with that of CBT. Already here I will note that a full scientific study of the effectiveness of psychodynamic treatment also requires comparison with the results of sham treatment or plain quackery to neutralize the placebo effect (above I referred to Erik’s remark here on this matter).
He then continues with a note about the difficulty of assessing the effectiveness of psychological treatment in general. He points out that the common way to examine treatment effectiveness (as in physical medicine) is comparison to a control group, preferably one that also receives some treatment (I will note that to neutralize placebo you also need double-blind conditions, which do not obtain in psychology). If there is a significant difference between the groups, that is corroboration of the effectiveness of the treatment under examination. But there is still the question of how we will know whether the patient’s condition has improved. It seems there is no escape from asking him. But that itself already raises question marks about the significance of the measurement. To what extent is the improvement he feels real? Can this improvement be quantified, and how? Asking a subject how he quantifies it (on a scale of 1–10) is a very imprecise method even if we believe the subject. It is difficult for a person to quantify feelings and distress. But for the sake of the discussion here we will assume that if a person reports an improvement, then the treatment has achieved its goal, even if the improvement itself is illusory or not properly assessed. Let us put it this way: the scientific claim we are testing is not whether there was an improvement but whether the patient feels an improvement (or reports an improvement).
He describes that over decades in which psychotherapy has been examined empirically, significantly greater improvements have been obtained than in control groups. This is not about psychodynamic treatment but psychotherapy in general (if I understood correctly, mainly CBT). As noted, this may stem from the patient’s confidence in the method or in the practitioner. It may be that drug treatment or the treatments administered to the control groups aroused less confidence in the patients. Moreover, psychotherapeutic and psychodynamic treatment is individualized, whereas group treatments do not include ongoing, personal listening to the patient, and this alone can create very significant differences. As far as I understand, the control groups there were not treated by coaches, Rebbes, or alien communicators. The question is to what extent the confidence the patient places in the practitioner and in his method is responsible for these differences (a patient who chooses a practitioner presumably trusts him), and to what extent attention and individualization influence the results. Therefore I seriously doubt how indicative these comparisons are.
Regarding the effectiveness of psychodynamic treatments, the research is less extensive. But he argues there are already some studies examining their effectiveness, and the results are quite controversial. First, we are dealing with a small number of subjects (by its nature, since this is individualized treatment). This changes the statistical meaning of the findings, but also leads to artificial inflation of the results (see the explanation there). There is also publication bias (we are dealing only with results that were published, and this is a tiny minority of treatments; what gets published likely belongs precisely to the more successful subset). Beyond that, there is the claim that a set of mental disorders was examined rather than focusing on one. A scientific comparison should examine the effectiveness of a particular treatment for a particular disorder. There is also a non-trivial overlap among the studies surveyed, and therefore the effect is even smaller (different studies refer to the same cases). There are also claims that psychodynamic practitioners use CBT techniques as well, and therefore the comparison between the two treatment types is problematic.
He concludes that in many studies no significant difference was found between the results of psychodynamic treatment and those of CBT (relative to no treatment). Admittedly, many more studies were conducted on CBT, and therefore the findings regarding its effectiveness are more valid. He himself writes toward the end:
“The ‘equivalence effect’ is a term referring to the tendency of comparisons among different types of psychological treatment to point to similar effectiveness across the different types. The effect was identified already in the 1930s and was called the ‘Dodo bird verdict,’ on the name of the character in Alice in Wonderland who declares at the end of a race that everyone has won, and therefore everyone deserves prizes. Despite the many years that have passed and the different therapy techniques that have developed and improved since then, meta-analyses in the field continue to indicate that this equivalence finding largely holds today as well. That is, no treatment method manages to demonstrate superiority over others in a significant and consistent way over time.
If so, the similar result regarding the effectiveness of psychodynamic treatment and CBT could be part of the “equivalence effect.” If so, we will get similar results also for any other treatment method. But the conclusion is that in this way one can also corroborate the assumption that every treatment helps to the same degree (placebo), and perhaps similar successes would be obtained also in charlatan and other treatments, unrelated to any psychological discipline. We have returned to the starting point: there is hardly any way to seriously validate the claim regarding the effectiveness of any psychological treatment or psychotherapy at all.
Interestingly, Pozniak nevertheless draws conclusions from the findings he described:
“Unlike the variety of pseudo-scientific treatments presented on this blog, psychotherapy in general has passed, for decades, time after time, the test of results; studies show significant improvement in patients’ conditions, for various diagnoses and across different types of psychotherapy.”
I do not know what his remark regarding other pseudo-scientific treatments is based on, unless he means CBT. I did not see in his words a shred of evidence for the superiority of psychological treatment over sessions with a coach or a shaman. From a member of the “Sharp Thinking” group I would expect him to highlight this point. It is true there are no results regarding the effectiveness of shamanism or meetings with Rebbes, but that is simply because such studies were not conducted. It is hard to see how conclusions can be drawn from this absence. Pozniak assumes that CBT is certainly effective, and his discussion is whether psychodynamics is as good. But for me, the comparison between psychodynamic treatment and CBT is akin to a comparison between Oren Zarif and a Native American shaman. The more interesting question is about the psychological framework as a whole and not only the choice of one psychotherapeutic technique or another.
Another remark regarding the conclusions about the scientific status of psychological theory. There is in Pozniak and others an assumption that the effectiveness of a treatment attests to the validity of the theory on which it is based. But this is a very non-necessary claim. Just note how many practitioners and how many treatment methods exist in the field of psychotherapy. All claim successes; no stall in this market is closed due to lack of success. If treatment effectiveness attests to the validity of a theory, then, from the equivalence principle, it follows that there are several different theories that are all equally valid—and some even contradict each other. I think this further strengthens my placebo hypothesis.
I will end with a brief discussion of two examples.
Example A: Repressed Memories
A few weeks ago I read an article by Justice Yitzhak Amit on the issue of repressed memories. [7] This issue has arisen quite a bit in courtrooms in recent years and has stirred heated debates. The case is of a minor who, after many years, complains of sexual harassment or assault, and when asked why they did not complain until now they answer that they did not remember it. There are psychologists who argue for the existence of a phenomenon of repressed memories, and if this is indeed the case, then this testimony can convict the assailant. On the other hand, there is always the possibility that it is imagination, or even a lie. Naturally, we have no way to examine what the “victim” remembers using independent tools (if we had sufficient other evidence, we would use it and would not resort to the argument of repressed memories). In addition, how will we prevent people from harming someone they dislike by claiming they have repressed memories that he harmed them in childhood? On the legal plane this opens a very problematic door to false accusations.
The thesis of repressed memories is a non-scientific thesis, since it is not falsifiable and not confirmable. The main difficulty in examining this phenomenon is that we have no ability to test empirically whether these memories are real or are a product of imagination or falsehood. The law grapples with situations where we do not have independent evidence of what happened, and in such situations we have only the victim’s memories. The judge must decide whether these are repressed memories or imagination (or perhaps a lie). Naturally, in such cases he turns for aid to “the professional,” the psychologist. But that fellow too is not equipped with better scientific tools to examine the phenomenon. He certainly cannot examine the specific person before him, but it is also hard to think of a way to examine the phenomenon more generally in a scientific manner. To do this, we would have to take people who raise memories that, they claim, were repressed for a long time, and examine whether they are indeed real or not. But how can this be checked? How will we even be able to gather a significant group of such people, for each of whom we can examine independently whether his memories are true or not? Nor can we implant traumatic memories in people for the sake of the experiment. It may be convenient for the judge to toss the hot potato into the psychologist’s lap, but the psychologist has no real advantage over him vis-à-vis these phenomena.
In his article Amit notes a dispute on this matter between practitioners and academic researchers. This expresses the distinction between the planes that I pointed out above. A practitioner may gain the impression from patients who passed through his hands that a phenomenon of repressed memory is possible, but on the theoretical plane there is no falsifiable scientific evidence for it. We must remember that a criminal conviction requires that there be no reasonable doubt regarding the defendant’s guilt. Therefore, even if we accept in principle the possibility of repressed memories, the path to conviction on their basis is very long. On the other hand, this is a phenomenon that cannot be scientifically examined, and therefore there is room to relate to it in a more accepting way. It is not reasonable to convict a person solely on the basis of repressed memories, but they can be used as supportive when there is other external evidence bolstering the conviction. Thus Amit indeed writes in another judgment of his that is quoted there:
“In those unique cases that arise from scientific uncertainty or the limits of science, the decision can be based on the inductive model that expresses life experience and common sense; yet the court must exercise utmost caution in such cases and rely on expert opinions and on the overall evidentiary tapestry. Note well: when science cannot rule out a certain phenomenon, this should not be construed as scientific uncertainty, since the statement ‘it cannot be ruled out’ is almost always true for any phenomenon.”
The lack of scientificity of a claim that is, in principle, inaccessible to science is perhaps a deficiency, but it does not suffice to dismiss these claims entirely.
Example B: Munchausen Syndrome by Proxy
In Column 144 I dealt with Munchausen syndrome by proxy as an example of a statistical error in court. There I analyzed the issue, so here I will only touch on the aspect of falsifiability.
We are dealing with a psychological claim that a person sometimes seeks attention by harming himself (Munchausen syndrome) or others close to him (Munchausen by proxy). Thus, for example, if he harms his child, people around pity the family and thus he receives their attention. This syndrome (especially when it is about harming another) is controversial among psychologists, and the main reason is the inability to examine the claim in a scientific manner. How will you know whether the person did what he did to gain attention or for some other reason? We are dealing with an interpretive proposal for people’s actions (and see in my aforementioned column the statistical fallacy that appeared in the reasoning given for this interpretation), but it is difficult to think of an experiment that would put it to a refutation test.
In a well-known case that occurred in Britain a woman was imprisoned for the killing of her two infants, on the assumption that they did not die of crib death but were killed by her. The scientific support for this determination was the testimony of a British physician who testified in court about a psychological theory of Munchausen syndrome by proxy, a syndrome that has no scientific support and is not falsifiable. The judge accepted the “expert’s” claim, and this could have happened only because a psychological theory is presented in court as a scientific finding and as the testimony of an expert, when its basis in many cases is shaky and subjective. Awareness of the subjectivity and non-scientificity of psychology is very important for a judge when giving weight to the testimony of such an “expert.” The judge must know that in not a few cases the psychologist’s testimony is worth little more than the judge’s own opinion and life experience. [8]
Long after this post was written, the following reference reached me regarding prevalent psychological claims that have been found problematic (the wording as if everything turned out to be nonsense is probably too extreme):
[https://m.facebook.com/story.php?story_fbid=453935399665007&id=100051456511332&m_entstream_source=feed_mobile](https://m.facebook.com/story.php?story_fbid=453935399665007&id=100051456511332&m_entstream_source=feed_mobile)
[1] I discussed this at length in my article on okimtot.
[2] It seems to me that, in general, it is more correct to view medicine as a technology and not as a science (a kind of mechanics dealing with a very complex machine).
[3] In my book The Science of Freedom (see also the article here) I explained that in the libertarian picture (which believes in free will), even if we ever arrive at perfect psychological knowledge that will give us the complete laws of psychology precisely, that would constitute a description of the map of pressures and urges within which a person operates. But because of his free will, a person always has the option to decide (choose) to act contrary to those pressures. In this picture, psychology does not describe a person’s behavior but the motivations and pressures that try to push him toward a particular behavior. I likened it to a topographical layout with mountains, valleys, saddles, and ravines, upon which a person walks. The layout exerts forces on him that try to push him (toward the valleys) or stop him (from climbing the mountains), but a person can always decide nonetheless to climb the mountain or not descend to the valley. Psychology works toward as full a disclosure as possible of a person’s topographical map, but this will never give us a precise prediction of the actual behavior. To sharpen: I am not speaking here about the fact that there are differences between Reuven’s map and Shimon’s. That difference also exists in the deterministic picture of the world and represents complexity and intricacy, not free will. I am speaking of a situation where we have fully exposed Reuven’s specific map, and my claim is that still there is no way to deduce from it his concrete behaviors, only possible directions. This is the problem of free will.
[4] See the previous note (the differences among people’s maps).
[5] See Erik’s comment here and the discussion that followed.
[6] For reasons that are partly justified, but there is also the influence of the desire of the psychologists’ guild to jealously guard its status and monopoly. The less tested and valid the tools at its disposal are, the more its zeal and domineering increase. Exactly like in the religious world. Have you ever wondered why when people say “you should go to a professional” they always mean a psychologist? (Not a plumber, auto mechanic, or physicist.) For some reason I have never heard of a physicist being granted the title “professional.” Just “a professional” means a psychologist. Likewise, I have never heard of a guild of physicists coming out against “charlatans” in the field. I already wrote here what I heard from my sister who studied criminology and told me that at the beginning of every course they devoted some time to a preface dealing with the definition of science. I told her that in physics I have never heard lectures on that subject.
[7] “Repressed Memories in Criminal Law,” Sefer Menachem Finkelstein, p. 483.
[8] As noted, it is convenient for a judge to hang his decision on expert testimony, thus relieving himself of the responsibility to determine the psychological truth. But the fact that it is convenient does not mean that the expert’s testimony has real value. In many cases he testifies on the basis of his personal experience, and like Freud he errs in giving scientific weight to that subjective experience.
Discussion
I focused on psychoanalysis because it contains something essential to the discussion, precisely because it is not empirical. The branches that are careful about empiricism may for that very reason lose their scientific status, as I will now explain.
When I speak about science, I am not speaking about the mere collection of facts. In the Milgram experiment you test a direct factual claim (whether there is obedience to authority figures), but, as I explained in the column, in science one also builds a theory that includes abstract concepts that are not directly accessible to experiment in order to explain those facts (this is abduction). Thus, when I ask why there is a tendency to obey authority figures, I am looking for an explanation of Milgram's results, and the theoretical explanation I propose must itself be scientific and stand the test of refutation. Otherwise, even claims like "Tall people usually wear large shoes" are science. Once you enter the theoretical world and construct a conceptual framework that explains the direct facts, you naturally arrive at psychoanalysis. Behaviorism in its direct sense is not science by this definition, precisely because it focuses on direct observational facts. That is fact-gathering, which is only the first step in the scientific process.
Human beings are complex creatures, but that does not mean that the laws describing them are not deterministic in the scientific sense of physics. Physics is also very complex. I referred to my article on ukimtas, where I explained that complexity does not essentially change the situation. It only means that there is more than one law governing the situation, but each such law by itself is still supposed to be scientific in the ordinary sense.
In psychology the laws are not like that, though I mentioned that this may depend on whether you are a libertarian or a determinist.
If I am indeed right that no significant psychological theory has ever been refuted, that is definitely meaningful. It is a (scientific) indication that this is a field that is not really falsifiable. The alternative is that somehow, by sheer coincidence, in this scientific field we always hit upon the truth. That is unlikely.
By the way, I did not rule out any field in psychology here. I discussed only their scientific status, not their value or their truth. That is a different question, which I did not address in this column. I will add here that I am indeed very skeptical about that as well.
Are René Spitz's studies on emotional deprivation, hospitalism syndrome, marasmus, and infants' need for emotional attachment not scientific?
Here is a link, for example:
https://iw.juscogens.net/marasmus-the-importance-of-affection-in-early-childhood-8d8d68
From a quick skim, I could not understand exactly what they did there, so I cannot say.
I do not work in the field and do not understand it, but as I understand it, after World War II there were many orphanages in Europe with infants. The infants received regular food and medical care, and still some of them suffered developmental regression that even led to death. René Spitz investigated the matter and concluded that the infants lacked a stable emotional bond with their caregivers. He proposed that part of the infants' care routine should include fixed hours with a fixed caregiver who would sing to the baby and care for it so that it would not feel alone and would have a stable figure with whom to form a bond. His conclusions were accepted and brought about the end of the infants' developmental regression.
It seems to me that this is scientific research according to the book (a natural experiment, research hypothesis, experiment, confirmation).
If there was a comparison to another group that did not receive treatment (to rule out spontaneous recovery) and to yet another group that received sham treatment (to rule out placebo), and if the groups were large enough, and if the improvement measures were defined sharply and clearly, then maybe yes.
A few points, with your permission:
1. The many experiments on conformity are explained by the basic psychological drives embedded in human beings, such as avoidance of embarrassment, rejection, and judgment, and the desire to be loved and accepted. In other words, as you argue, Milgram's electric shock experiment, Solomon Asch's experiment, and a wide variety of other experiments examining conformity are, overall, merely direct facts, but they are explained by the basic motivations that exist within us—which is essentially abduction. The fact that some people act differently, as you already argued, only means that there are quite a few additional variables, such as education, personality, and so on. But, in my humble opinion, when there are high percentages of people who act in a certain way (in Solomon Asch's experiment, for example, out of 123 people only 1 continued to give the correct answer), I think the abstract concepts explain this well, and I see no need to look for other answers. On the other hand, that does not mean we should not look for other answers, and certainly more and more experiments should be carried out in order to try to refute this. But such experiments do indeed occur, and from time to time we witness those same results of conformity. In addition, I think the trigger for looking for another answer should come דווקא when we witness opposite numbers—namely, that in ambiguous situations people tend to rely on themselves and not on authority figures. The issue is that this does not happen in sufficiently high numbers to warrant that, so I do not see why one must rush to look for other explanations. I remain with the conclusion that seems reasonable to me: people in situations of ambiguity will tend to be conformist. Everyone, and one hundred percent of the time? Certainly not. But the probability is enough to predict how people will behave in similar situations. What is the problem here?
2. I repeat that in my opinion there is no point in looking for rigid laws when it comes to the human psyche. As already mentioned, psychology is a statistical field, and if we see 80% of a certain group of people behaving conformistically in ambiguous situations in order to avoid embarrassment and judgment, then that is the conclusion, and that is that. Of course there are another 20% who will not behave that way, but that does not mean the theory should be abandoned. Rather, one should look for the variables that explain that percentage and set up an experiment to test it. As stated, the human psyche is complex, and in that sense it is much harder than any other field to isolate variables and study them separately. But we work with the tools available to us. It is not ideal, I agree, but they yield not-bad results when it comes to human behavior.
3. Let us take the example of dissonance. One can say that a person experiences dissonance in quite a few situations, almost all the time. The main motivation is to reduce it, and he does so in various ways. Let me present a small study: in the experiment (Aronson and Mills, 1959), students were offered the chance to join a club in which discussions were held on sexuality (in those years the subject was not discussed so much), but they had to pass tests in order to be admitted. One group was asked to go through a very harsh initiation test requiring effort (exposure to, and reading, very embarrassing texts about sexuality), and a second group had to go through an easy initiation test (to read a light sexual text in order to test their level of embarrassment), and a third group had no initiation test (control). After being admitted, the different groups attended the first lecture (the discussion), whose subject was reproduction among insects (not really what they had expected). The students were then asked to fill out questionnaires regarding their level of satisfaction.
Results: those who were admitted without effort, along with those admitted on the basis of mild effort, gave low satisfaction ratings and expressed regret about participating. By contrast, those who went through a harsh initiation test reported interest and a positive experience in the discussion.
Conclusion: when a person agrees to undergo a difficult experience in order to achieve his goal, he will sanctify and justify the goal in order to relieve himself of the dissonance of having made the effort for nothing. In that state, his perception will be more positive and his degree of enjoyment will increase.
There are quite a few studies on dissonance, and I think that for the most part they all point to facts that explain the abstract theory well—the human drive to reduce the unpleasant feelings of dissonance, that is, a gap between beliefs, behaviors, and attitudes. And dissonance is only one example among many that can be explained through different conceptual systems.
Again, I am putting psychoanalysis as a method of treatment aside, because I too do not see much value in it. But I do not think that the various fields in psychology, as research fields, fall entirely short of the scientific conditions you present. If your claim against psychology is only on the basis of the argument that you have not yet encountered a case in which a theory collapsed, I can understand that. But even so, I can completely imagine a situation in which this could happen (and again, I am not sure it has not happened. I have not checked).
A) To the best of my knowledge, behaviorist theory was in fact refuted by research showing that monkeys can learn new processes without direct learning,
1. Abduction does not provide a theory that itself stands tests of refutation. These are simple explanations that anyone can give and no one can refute. As I wrote, what stands to refutation is the direct facts, not the theory. That is not science. When the theory is reasonable, that does not make it science. And again, it may be true, but it still is not scientific.
2. I have no problem at all with statistical laws, and I too regard them as science. I explained this in the column, and I also discussed the difference between them and statistical laws in the natural sciences.
3. Here too I will answer as in 1. There is no real abduction here. You are only describing the direct facts in a more general way. There are no abstract concepts or meta-principles here.
In conclusion, I do not deny the existence of scientific elements in psychological theory. I only claim that this is usually not the case.
And therefore? I wrote that behaviorism is too observational and therefore not scientific, in the opposite sense from psychoanalysis. That does not mean it cannot be refuted. The claim that there are more than a hundred people in my neighborhood can also be refuted. Does that make it scientific?
Are depression and attention deficit disorder moral problems, psychological problems, or bodily problems—or perhaps do they simply reflect human diversity (who said we all have to be focused or happy all day)?
Are Ritalin and antidepressant medications scientific?
Is immoral behavior a mental illness (a person knows what is right to do and does not do it)?
Is a mafioso who murders business rivals mentally ill, or a rational person who wants to profit in a market not regulated by the state?
Was Genghis Khan mentally ill?
Would psychological treatment have cured Genghis Khan, or was his scale of values simply distorted? Or perhaps they should have given him medication?
Is child-rearing beneficial (for example, so that children will be orderly, diligent, etc.)? If the answer is yes, does that show that psychology is true? If the answer is no, does that show that psychology is not true?
These questions are unrelated to my column.
The definition of illness is normative, not scientific, and therefore these are merely semantic questions. The important questions are whether he is a criminal and whether he deserves punishment.
Questions about a specific person, such as Genghis Khan, are not important, and I do not see how one could answer them.
As for education and the truth of psychology, I did not understand the connection.
Before discussing whether psychology is scientific, it would be worthwhile to discuss whether there is even such a thing as a psyche that can be investigated scientifically. These questions bring to the surface the fact that some deny the claim that the psyche exists and argue that people's problems are either physical or moral. The question of education points to the fact that there indeed is a psyche one can work with. If we are convinced that there is a psyche, then we can also ask how to investigate its logic and propose how to change it.
Most treatments are sheer charlatanism. I have never heard of therapists signing a contract with a patient in which they commit to achieving a certain result within a certain number of sessions.
But on the other hand, smart and experienced therapists can identify the problem and provide tools the patient does not have. But that is less connected to theories and more to intuition, experience, and practical wisdom.
Freud too expressed a certain despair regarding treatment.
Moreover, the theories say that personality is formed at very early ages, and therefore it is difficult, if not impossible, to change.
As for scientific status, one can imagine that in the near future, if the measuring instruments of neuroscience develop, it will be possible to confirm or refute all kinds of ideas and concepts formulated in the past and the interrelations among them. In the meantime, it is mainly intuition, experience, and practical wisdom.
Hope is very nice. As of today, psychiatry is truly blind regarding the bodily causes of "mental illnesses" (?).
Two comments: first, this is not the first time I have heard from you this business about an "professional" supposedly meaning only psychologists. That is simply not true. If I have an electrical or plumbing problem in my apartment, I can absolutely hear people telling me not to deal with it myself but to call a professional. You do not have to believe me—google "professional" and you will see that alongside references to therapists there are also references to other fields (electricity or plumbing, for example; I assume there are others as well). The point is that there are fields in which every person has some degree of contact—you do not need to be a certified plumber in order to replace a trap, but if you try to deal with more complicated things, you are going to break something. In the mental sphere too there are many people who offer treatment, and many people who are sure they are capable of treating others, but they are not. Even "professionals" in the treatment field cannot guarantee results (and in truth, the more professional a person is, the more cautious he will be with his promises, at least in my experience), but the use of this term is meant to distinguish between someone whose training is accredited and substantial and someone who took a three-month coaching course—and again, in my experience, quite rightly. I have happened several times to hear coaches speak about their methods, and on the only occasions when smoke did not come out of my ears, it turned out to be a coach with a much broader therapeutic background (a psychologist or social worker, for example, who decided to specialize דווקא in coaching).
Second, I largely agree regarding the scientific status of psychology—that in many cases it really is not a scientific field, and the ability to examine it with scientific tools is questionable. But I very much dislike the tendency to belittle or dismiss the whole field of psychotherapy because it is "not scientific." Freud brought much more into the world than a vocabulary of concepts—he also brought an entire world of therapeutic practice, of understanding the dynamics between therapist and patient, of understanding mental processes that, when described, sound completely mystical until you experience them yourself and see how real they are (processes such as transference and projection, and the like). In my view, the very fact that even after there is no part of his theories that has not been analyzed from every side and subjected to criticism, attacks, and rejection, he is still considered the father of the field, and even those who disagree with him tend to relate to him with respect—usually—says something about his achievement, which goes beyond constructing an efficient vocabulary of concepts. Psychotherapy is not something one learns from books, although books provide quite good knowledge and a meaningful and useful frame of reference. It is something learned in the experience itself, and only if the therapist has himself experienced both what it is to be a patient and what it is to receive supervision from someone more experienced, and is able constantly to observe his own internal processes. In other words, we are dealing here with a craft (in the sense of a precise skill that must be practiced and learned through doing) and with art (in the sense of rules that are not written down but require the development of intuition and sensitive inner understanding, and of course creativity and daring). In art too there is enormous significance to learning techniques, becoming familiar with history and with different approaches and with the materials you work with, but in the end the creative process itself is not the application of things others did—sometimes it is the exact opposite. And psychotherapy belongs to that world. So yes, I completely agree that it is not scientific. Shakespeare's plays and Mozart's music and Leonardo's paintings are not scientific either. But there too there is an entire world of learning, experience, training, and creation. In my opinion, it is very problematic to come from outside and relate to an entire world of thought and practice, in which people have been engaged systematically for more than a hundred years, discussing it, arguing about it, changing it, innovating in it—not to mention the heavy mental price and emotional burden they pay in order to engage in it—and claim that it is all nonsense or that everrryone is biased and does not want to lose his livelihood. Because if after more than a hundred years of observation and experience therapists still believe in certain basic principles (and by and large—they do, for all the disputes), then perhaps it is worth hearing what they have to say and not dismissing it, certainly when you come entirely from outside and have never experienced the position of therapist (or patient—I do not know about that).
You did not argue about the scientific status of psychiatry, meaning pills for treating depression, attention problems, hallucinations, excessive sexual drive, and the like. What is the categorical difference between drug treatment and talk therapy? If one can be grounded, then so can the other.
I assume there is no difference at the principled level, but there are significant differences in terms of the therapeutic and experimental method and in terms of the areas each method treats. In terms of method—it is much, much easier to examine a psychiatric pill scientifically than talk therapy. It is very easy to find a control group, one can create a placebo easily, and in general it is easier to conduct an experiment, both practically and ethically. Second—psychiatry often treats areas that even the best psychologists will say are outside their field. It is difficult to impossible to conduct therapeutic conversation with a person suffering from active psychosis. There simply is no partner for conversation. And we know clearly that medication can greatly alleviate the positive symptoms of schizophrenia. A psychologist who tells you he can do that through conversation would usually be highly suspect, even in the eyes of his colleagues.
Allow me to ask a question out of curiosity, as someone who studies psychology, is very fond of philosophy, but is completely ignorant in physics.
Why, then, is abduction so relevant to science? Would it not be more correct to establish a conceptual framework that can be refuted rather than one that cannot be refuted? What is the point of laying out an abstract theory if it does not even meet the most basic criterion, as you argued, of Popper? Is it not more sensible to explain the facts by means of theories that are themselves testable, and from there to look for the theory that explains the facts better? On the contrary, when the conceptual system itself is open to empirical testing, does that not make the facts much more valid?
Is it really abduction itself that makes a theory "scientific"? That is, is that truly the decisive principle between a scientific theory and a non-scientific one?
I really do not understand why abstract concepts that cannot be proven are precisely what make a theory scientific, rather than a more general conceptual system that is itself testable. Is this really the best way to do science?
This is a question arising entirely from a desire to understand. If you have an article that explains this at length, I would be happy to look at it.
Thank you.
On the first matter, it is hard to argue. That is my impression. When people simply say "a professional," that is what they mean.
As for the value of psychology, that is not the issue here. Here I discussed only its scientific status. It may very well have value, but that is a different discussion.
You have a basic mistake. When I speak about abstract entities, I mean concepts that cannot be directly observed but do have predictions and can be refuted. Otherwise it is not science. My claim is that direct factual claims are not science in the full sense. Science is an attempt to understand what lies behind the facts (and not merely to collect the facts themselves), but those understandings themselves stand tests of refutation. That is the whole scientific trick.
The thing is, even on the first matter one can argue. That is no problem—I suggested a simple experiment to you. Go to Google and search for the occurrences of "professional" and you will see that your distinction simply does not fit the facts. Google has no interest. In my opinion, the point is not therapists' inferiority feelings, but the fact that the field is very open and that people have the illusion that there is nothing here to learn.
And as for the value of psychology—I agree that it is a different issue, but it is closely connected to the question of its scientific status. Here I actually do agree that emotional therapists have an inferiority complex vis-à-vis "real science" (at least in my impression), and that they sometimes refuse to recognize that their field is an art and not a science. And that can be very harmful, mainly because it means that so very much of the studies in the field are a terrible waste of time. And on this there is fairly broad agreement among many therapists I know, both psychologists and social workers—many of the courses are unnecessary nonsense, and the real and meaningful learning is done in the field. There is a lot to learn—an enormous amount; it is endless, highly complex, and very difficult. But it does not take place in academia.
David,
The situation in psychiatry is far from rosy, as described here. I have been working in the field of psychiatric rehabilitation for quite a few years, and I see how unclear it is to them as well how to treat each case. They know that in general there are certain pills or treatments that help psychotic states, others that counter anxiety and depression, but it is very general. Usually, in order to find the best medication, one needs several good years of trial and error, during which they switch the treatment every few months to try other directions. I asked psychiatrists several times questions about patients, and most of them gave me the same answer: "I have no idea." One answered, "Psychiatry is like seasoning. You put in a little salt, a little paprika, until something good comes out." Another psychiatrist answered me, "Don't ask me why, ask me how much—how many mg to add." A third said, "Schizophrenia is the unknown disease. We have no idea about almost anything because of the complexity of the nervous system."
And in general, assessment of the treatment is not based on objective observations, like blood tests, but on the patient's reports and the psychiatrist's personal impression.
In short: this field is very far from science.
I also work in mental health rehabilitation and I did not say anywhere that psychiatry is an "exact science." It absolutely is not. I answered the question of why there is a difference between psychiatry and psychology regarding the degree of scientific status of their claims. The answer: there is a difference because in psychiatry it is much easier to conduct experiments. We also know that at least in the area of positive symptoms there are medications that definitely help (statistically, yes?), and these are things that, again, have been examined in studies that in my view meet the standards of scientific research, and our own eyes see that this is so. That does not make the field of psychiatric treatment a field like ordinary medicine (which also involves a very large degree of educated guesswork and statistical assumptions), but as a scientific discipline—yes, it is much closer to real science than psychology is.
Whoa!!! A colleague?
A book came out not long ago by one of the editors of the DSM (the psychiatric diagnostic manual), Allen Frances, called "Saving Normal." There he writes with disarming candor how diagnoses in general do not come from a place of "science" or "psychiatric professionalism" at all, but from many economic considerations, such as insurance companies or cultural values. Interesting that in the previous century there were 6 psychiatric diagnoses; today there are more than 200… not because the scientific literature has developed so much, but because the culture encouraged the formulation of all kinds of behavior as "disorders."
As for empirical experimentation—I agree that there is serious testing, but it is really not by the criteria of objective science. Measuring depression according to the patient's report is useful. Scientific? Well, nu, all right.
But I agree that the metrics there are much clearer and more empirically testable than in the analytic literature, which is more metaphors and helpful stories.
Psychiatric diagnoses are a completely different question. It is a very problematic area, and I think most people who work in it admit this openly. Even aside from the cultural and political issues, diagnosis itself is problematic, because you do not know whether what you are diagnosing even exists at all, what its boundaries are, and so on. Still, symptoms (as opposed to diagnoses) are more measurable—certainly when it comes to acute symptoms such as hallucinations. And again, here modern psychiatry shows results, unequivocally. It does not always help, not everyone, but there are results. That is why I specifically mentioned "positive symptoms of schizophrenia." That is something relatively easy to identify, and one need not rely only on the patient's report (or at least, the report here is not vague and is usually not open to interpretation. If he hears voices, he hears voices. It is not "depression").
Behaviorism tries to give a reductive answer to complex behaviors; the basic claim of the behavioral approach is that no matter how complex the behavior is, I can still break it down into its components and explain it according to the simple principles of reinforcement. If we refer to the spectrum of normal and abnormal human behavior that behaviorism tries to explain, it is hard for me to see how this is not science,
P.S. one of the components of a good theory is giving a concise explanation for a range of phenomena
My question is whether the cataloguing of mental disorders turns psychiatry into a science. If the scientific project is characterized by finding general and necessary laws from a cluster of phenomena, it is not at all clear that psychiatry is such a thing. What law does it discover in man? Let us take schizophrenia, for example. We are talking about a broad cluster of disorders (say—delusions and auditory hallucinations) whose common feature is a "split in the mind" (schizo-phrenia = split mind). Where does "science" enter here? It is a description and cataloguing of impaired mental functioning. What theory can be inferred from it?
David,
That still would not settle it. I do not know how to count the occurrences. In my opinion, what matters is the percentage within the use of the precise professional term ("professional" versus plumber or electrician, and "professional" versus psychologist/social worker).
My feeling is that the use of this term expresses a desire to emphasize that this too is a profession in the disciplinary sense (against the disdain for it. It is the same inferiority feeling that you yourself described), and perhaps that is why I am sensitive particularly to those usages, and a kind of 'confirmation bias' is created in me.
Perhaps you are right, but those are general statements. If you have a good example, that would improve our ability to discuss it. (Most of my education on the matter comes from a Skinner book I read in high school.)
I must say that this article caused me to stop a psychological treatment I started not long ago. That is—I had many questions along the way, but I assumed it was all scientific and evidence-based and did not allow myself to doubt the reliability of the method. It should be noted that not only is it unhelpful, in some cases it is also very harmful. Instead of shrinking a small problem (maybe it is not small, but that itself they are not really willing to examine seriously; rather, they come with a very clear preconception in advance about connections to past experiences), they expand it, deepen it, and cause you to feel that it is not under your control but larger than you and connected to experiences from the 'unconscious,' etc.
In your understanding, are there approaches in psychology that do help with mentally coping with intrusive thoughts that lead to anxiety (beyond CBT)?
I hope and assume that it was not only because of what I said, but because you reached your own conclusions that the treatment is not helping you.
I do not understand the subject, so unfortunately I have no advice. I will only repeat what I wrote: even if psychology is not a science, it is still possible that treatment of one kind or another may help. I would not rule that out.
From your description, it sounds as if you went in a psychoanalytic direction, and perhaps it would be worthwhile to try other directions. If you are suffering, I would suggest continuing to look into it and consult.
Best of luck.
As stated, from the outset I had serious questions and doubts. I assumed that the therapist was relying on authorities greater than both him and me, and I preferred to trust him.
In my opinion, a large part of the need for psychologists stems from the thought that there are personal phenomena we find hard to control because they are connected to deep experiences and unchangeable personality patterns. In other words, psychological theory feeds itself and brings more patients into its orbit, without trying simply to have people work on themselves.
This article is extremely important, and perhaps it would be worthwhile to deal with the subject even more.
By the way, the psychological approach holds that every thought originates in one emotion or another, something you have written against more than once. I think that in this respect it is certainly mistaken and harmful.
There are hundreds of studies supporting the claim that the therapeutic relationship itself helps the patient. One may ask whether a relationship with a Hasidic rebbe or a coffee-ground reader would also help, but here there are two things that must be taken into account: first of all, the obvious assumption is that a focused relationship that lasts about an hour, takes place every week, and revolves entirely around analyzing the person's traits, unloading difficulties, exposing strengths and weaknesses, etc.—the "therapeutic" effect is more effective. Especially when an entire year in clinical psychology studies is focused on the treatment itself, on the way to create a measured distance between therapist and patient, such that on the one hand it does not create condescension, and on the other hand does not bring about dependency, etc. I do not think one needs scientific research for this, only common sense.
Second, Talmud study is also not a scientific field, but I assume you too would agree that in order to deliver a full lesson with novel Torah insights, it is preferable first to grind through tractates, delve into the commentators' interpretations, and become familiar with the accepted style of reasoning. Otherwise, we would take that person's words with very limited confidence.
Everything was explained in the columns.
There are few such studies that stand up to serious scrutiny (once you clean out placebo and spontaneous recovery, and provided there are clear and significant measures), if any. I also do not disqualify a field because it is not scientific. I only say that there are good intuitions there, but not a discipline.
Our Rabbi writes:
For a similar reason, we do not find (certainly among Jews) scientific experiments that test the effect of prayer or a person's righteousness on his health and on his chances of recovery.
And as I recall, in one of our Rabbi's books he describes an experiment that was done (perhaps several experiments), and it turned out that there is
an effect to prayers!
The study received a cool reception from physicians, and as I recall, our Rabbi actually attacked the criticism.
(I do not remember that he agreed with the study's determination—but he criticized the criticism.)
At any rate—among us Jews, such a study was conducted.
I referred in the column to Appendix E in my book God Plays Dice. There I dealt with this a bit. There are no studies there of Jews on Jews. As usual, Christians really believe in their dogmas, unlike us, and are therefore willing to put them to empirical tests. With us, of course, that is forbidden. Better safe than sorry.
With your permission, I would like to address your first question (whether a psychological theory is scientific and whether it is falsifiable) and not the second (whether the effectiveness of treatment methods is scientifically grounded), since I agree with everything you wrote about the second question.
In my humble opinion, you are doing psychology an unnecessary injustice. That is, I do agree that there are fields (psychoanalysis, for example) that are not worth very much. But to take one field and present it as representative of the discipline as a whole is, in my view, a mistake, and certainly says nothing about the other fields that are considered more "scientific." Behaviorism and social psychology, for example, are central fields in psychology, and I do not see how conclusions arising from those fields accord with psychoanalytic conclusions. In addition, the question that should be asked is whether the conclusions drawn from those fields explain reality, regardless of the claim that you cannot think of a psychological law that was rejected as a result of empirical refutation. Whether or not you know of such a rejected law says nothing about the truth of the conclusion.
Let us take Milgram's electric shock experiment as a test case. The conclusion that follows from the experiment is that human beings in ambiguous stressful situations tend to obey a nearby authority figure (what is called "informational conformity"). Moreover, when the experiment is repeated and, instead of the expert experimenter, an ordinary non-expert teacher with no authority is put in place, one sees that there is indeed much less obedience. If so, I do not see how the conclusion is unreasonable, and by the way there are quite a few similar experiments that confirm it. True, you did not claim that psychology cannot be tested empirically, but the claim that you do not know of psychological "laws" that were refuted (I do not really understand what psychological laws are. Human beings are complex creatures, and there is no single law that operates on everyone all the time, but rather different probabilities of behavior in different situations) does not in any way move the conclusion, which seems entirely reasonable to me. Psychology need not search for laws that are always valid in every situation. I am certainly not telling you anything new, but even in physics we have no way of knowing whether those laws are correct one hundred percent of the time. All the more so in psychology. Some conclusions are more reasonable and some are less reasonable.
As you argued, psychology is a statistical field, and when we take a certain conclusion, put it to an empirical test, and see similar results, we assume that the conclusion is correct until proven otherwise. In that sense, it is no different from physics, biology, or any other scientific field. As stated, the fact that you do not know of psychological laws that were refuted (and I am not sure there are none) means nothing, except simply that they have not yet been refuted. Gravity has not been refuted either, so what? In addition, I think that throughout the article you took an example that is indeed very amorphous (the connection between frustration and aggression) as a representative case, but in my opinion this example is not representative at all.
My point, overall, was simply that one should distinguish between psychoanalysis (and other non-scientific fields) and the other fields, because psychology is a very rich branch made up of quite a few areas of research. To take a particular field as a representative sample (psychoanalysis, for that matter—which, as I said, I agree with all your claims about, both as a collection of theories and as a treatment method), ask, "Is psychology a science," and think that if you have refuted psychoanalysis then you have also ruled out the other fields in psychology—that is a mistake in my view.