Philipp Berghofer
Department of Philosophy, University of Graz, Austria
I’m very grateful to Mahdi Khalili, Andrea Reichenberger, and Harald Wiltsche for engaging so carefully with my work and for raising questions and concerns that have pushed me to refine and develop my position. I’m in the fortunate position to reply to “critics” whose support and feedback have been instrumental in shaping the very ideas we are discussing here. In particular, Harald and I have been “phenomenological partners in crime” for some time now, continually advancing our project of establishing phenomenological approaches to physics. Recently, my work and ideas have benefited substantially from Mahdi’s expertise in perspectivism and philosophy of science, jointly working on how to fruitfully connect perspectivist approaches to science and physics. Andrea and I have a shared interest in the work of Paulette Destouches-Février, and Andrea has revived my ambition to work on Destouches-Février. In what follows, I begin by responding to the six questions raised by Mahdi, then turn to the two potential risks uncovered by Harald, and finally comment on the worry raised by Andrea.
1. Khalili’s six questions
1.1. Question 1
Khalili’s first question concerns my claim that “every piece of justification can be traced back to epistemically foundational experiences.” As Khalili rightly notes, in my view, this is not only true for everyday contexts but also for science. But what do I even mean when I speak of epistemically foundational experiences in the context of science? Khalili suggests: “I assume that ‘experience’ in science consists of the results of experimental and observational practices, or in short: empirical results.” Here I need to emphasize that when I say that all scientific insights can be epistemically traced back to experiences, these experiences are not to be understood as technical empirical results incomprehensible to the general public. I really have in mind ordinary experiences as we know them from our daily lives. Of course, these experiences, then, are supposed to (dis)confirm highly complicated scientific theorems and in most cases it will be incomprehensible to the general public how these experiences can (dis)confirm the respective theories. To get a better grasp on this, here are two examples.
Time dilation: Special relativity implies that if a clock moves relative to an observer, this clock will be measured by the observer to tick slower than a clock that is at rest relative to the observer. The greater the relative velocity, the greater is the effect. This can be, and has been, tested by comparing two synchronized clocks such that one of the clocks remains stationary at the lab and the other one is sent in a plane eastward around the world. When the clock has returned from its round-the-world trip and we compare the formerly synchronized clocks, we will see that for the traveling clock less time has passed.
When I say that in such an example the line of justification can be traced back to epistemically foundational experiences, these experiences are not to be confused with the fact that less time has passed for the traveling clock or the fact that clock1 says T and clock2 says T’. The experiences in question are precisely the visual experiences of the experimenter when she looks at the displays of the clocks. What is presented to the experimenter visually is what is displayed on the clocks.[1] Thereby, she has non-inferential justification regarding what is displayed on the clocks. Due to her background knowledge regarding how these clocks work, she in further consequence has inferential justification to believe that for the traveling clock less time has passed. Due to her background knowledge in physics, she thereby is inferentially justified to believe that this supports special relativity. My account implies that the experimenter’s line of justification is different from yours when you read about the outcome of the experiment. What is visually presented to you is not the display of the clock but, for example, the text of an article that reports on these findings. Your line of justification is inevitably more indirect than the experimenter’s. I take this to be in agreement with common sense.
Wave-particle duality: The phenomenon that quantum objects can exhibit either wave or particle behavior depending on the experimental setup is known as wave-particle duality. Let’s consider a typical double-slit experiment. The experimenter sends individual electrons through a double-slit device and at the screen there emerges an interference pattern. In this experiment, the electrons showed wave behavior. Now the experimenter changes the setup such that there are detectors at each slit. Now the experimenter knows which slit each particle has passed through, but no interference pattern emerges on the screen. The electrons show particle behavior.
In this experiment, the epistemically foundational experiences are not the fact that electrons can show wave or particle behavior. It is not the pattern that appears on the screen. The experiences in question are precisely the visual experiences of the experimenter when she looks at the screen and is visually presented with a particular pattern. What is originally given to the experimenter are not the photons or something abstract like “wave-particle duality.” It is the pattern that she observes on the screen. From this, she can infer that photons can manifest both particle and wave behavior, depending on the experimental setup.
With these clarifications in place, let us turn to Kahlili’s first question:
“If we accept the theory dependence of empirical results, in what sense can empirical results be first, primary, or foundational? How can theoretical beliefs be constructed on the foundation of empirical results, whose meaning, relevance, and significance are dependent on those theoretical beliefs?”
In this context, Khalili rightfully emphasizes that even in the case of ordinary experiences we need to take into account that our beliefs can influence our experiences. However, Khalili points out that this “problem” magnifies in science since “in the case of unobservable entities such as bosons, genes, and gravitational lenses, realists and antirealists disagree even about the very existence of these entities. This deep disagreement arises because empirical results are dependent more heavily on theoretical concepts than ordinary experiences on prior beliefs.”
In light of the above, the first thing to note is that according to my account we should not make a clear-cut distinction between ordinary experiences and experiences in science. On the contrary, in many cases the scientist’s experiences are ordinary experiences (or experiences structurally similar to ordinary experiences) that present the scientist with, for example, numbers on a clock display or a pattern on a screen. The reason why there is significantly more disagreement about whether “scientific experiences” justify believing in the existence of unobservable scientific entities like electrons and bosons than disagreement about whether ordinary experiences justify believing in ordinary objects like tables and chairs is precisely because scientific experiences are not originally presentive regarding electrons and bosons but experiences of tables and chairs are originally presentive regarding tables and chairs. (Or, if you believe that only first-order properties such as colors and shapes can be originally given, then the inference from table-shaped objects to tables is sufficiently more straightforward.) For instance, the lines observed on the photographic readout of a bubble chamber experiment might be interpreted as tracks of an electron but the electron is not originally given in such an experiment. Similarly, when in the LHC experiment data is gathered that is interpreted as confirming the existence of the Higgs boson, this is because complicated calculations yield that this is the kind of data to be expected if the Higgs boson exists, but you don’t observe the Higgs directly. What you directly observe is the data or perhaps an artificially intelligent system telling you that there is a match in the experimentally gathered data with the predicted data.
The point is that the “scientific experiences” are not structurally different from ordinary experiences in that the former are somehow more problematic or theory-laden than the latter. Rather, scientific experiences are ordinary experiences in the sense that what is originally presented are ordinary objects such as lines and patterns on a screen, the display of a clock, etc. The difference is that scientific experiences are used to justify statements that refer to entities or laws that are beyond direct experience. (Of course, scientific experiences are also different from ordinary experiences in that the former result from the experimenter’s intention to experience something that is scientifically relevant and typically realized through an experiment.)
Importantly, I do not want to say that scientific observations are not theory-laden. Even our ordinary experiences are theory-laden in the sense that they are influenced by what we think we know about the world. Let us now, finally, return to Khalili’s first question. The question was: How can scientific experiences be both: (i) epistemically foundational in the sense that all justified scientific beliefs can be epistemically traced back to them and (ii) theory-laden in the sense of being dependent on such scientific beliefs? Isn’t there a clear inconsistency between (i) and (ii)? In my view, they are not inconsistent because (i) concerns an epistemic dependence and (ii) a genetic dependence. (i) specifies that scientific beliefs are epistemically dependent on scientific experiences. (ii) specifies that scientific experiences depend for their existence or their exact phenomenal character on scientific beliefs. This is analogous to how ordinary experiences justify beliefs, but are also shaped by them. When I look at a table, my table experience justifies me in believing that there is a table. This experience would be drastically different if I had grown up in an environment where there were no tables such that I lack the concept of “table.” What I know about tables shapes the way I perceive them and helps me to act adequately when I’m invited to dinner. However, this “expertise” may also negatively impact my ability to consider what role such objects could could play in different contexts and lead to false beliefs when presupposing that everybody will utilize these objects in the way I do. Similarly, the theory-ladenness of scientific experiences is not necessarily a bad thing as it will help the scientist to quickly and effectively draw correct conclusions. However, in some cases it may lead to premature conclusions. This, however, does not entail that scientific experiences are not a source of justification. It means that we must be careful in our reasoning and be mindful of what is originally given. This brings us to the next question.
1.2. Question 2
Khalili points out that in contemporary physics many experiments involve data processing by artificially intelligent systems. In this context, he raises the following issue:
“[N]ot only are we unable to directly experience objects, but it is also unclear what it means to experience empirical results (is our experience directed to the data displayed on computer screens?). Thus, my second question can be framed in this way: what does it mean that instrumentally mediated empirical results – for instance, large datasets – have their phenomenology? More generally, how does experience-first epistemology take into account the role of instrumentation, computational systems, and automatic processes in contemporary science?”
This question is closely linked to the previous one. Recall that in my account “scientific experiences” refer to the respective scientifically relevant experiences of the scientist. In the case of experiments whose output is large data sets, neither the data nor the data processing of the computational system is the experience in question but what the scientist sees on their computer screen. In a very simple experiment, all the gathered data may be accessible to the scientist and they can rather straightforwardly compare whether it concurs with the predicted data. In many cases, however, a computational system may select only a tiny fraction of all the gathered data and report to which extent there is congruence with the predictions. Furthermore, typically the experiment will not be conducted by a single experimenter but by a group of scientists. Thus, the line of justification of scientist S will not only depend on what S observed and on what S knows about physics but also on what was observed by S’s colleagues, whether S can trust them to be reliable, and on whether S can trust the data gathering and data processing mechanisms to be reliable. This means that the line of reasoning/justification will be highly complicated. However, this should not come as a surprise and I don’t think that this constitutes a problem specifically for my account. Justification by testimony and the epistemic role of artificially intelligent computational systems inevitably must be taken into account when inquiring into the line of justification of scientific insights. Regarding the role of computational systems and data processing devices in contemporary science, my experience-first approach implies that it is not the gathered/processed data that constitutes the scientist’s evidence and it is not the computational system that has justification. It is always the subject that gains justification and this justification, ultimately, is constituted by their experiences such as being visually aware of a set of data on your screen.
Of course, we can say that the gathered data constitutes evidence in the sense that the data is saved and can be accessed by several scientists. But, strictly speaking, for each and every scientist accessing the data, it is not the data as such that, ultimately, constitutes their evidence but it is their experiencing it. This is like when we say that a bloody knife is evidence that a crime has been committed. Strictly speaking, it is the inspector’s seeing the bloody knife or the judge’s hearing the inspector’s report that constitutes their respective evidence.
1.3. Question 3
Recently I have been contemplating and working on the idea that science, at a fundamental level, does not objectively represent an external world but describes what the experiencing subject should expect to experience next. One crucial motivation for this claim is that quantum mechanics can be interpreted in ways that go in this direction. In this context, Mahdi questions whether quantum mechanics can be understood to be fundamental given that (i) quantum mechanics is the non-relativistic limit of quantum field theory, (ii) general relativity constitutes the other pillar of modern physics and cannot be reduced to quantum mechanics, and (iii) it is unclear how quantum theory and general relativity will be merged in a future theory of quantum gravity.
Regarding (i): One crucial feature of QBism and other interpretations of quantum mechanics that I consider to be surprisingly consistent with phenomenological teachings is that they are not modificatory interpretations. That is, in contrast to Bohmian mechanics and objective collapse theories, they do not modify the quantum formalism. This is relevant because quantum field theory can be considered the successful marriage of quantum mechanics and special relativity, and for modificatory interpretations it remains unclear how they could be made consistent with special relativity. When proponents of non-modificatory interpretations say that quantum theory is fundamental, they mean to imply QFT. The conventional wisdom is that nothing conceptually relevant changes when we move from quantum mechanics to QFT, but I agree with Khalili that it would be a worthwhile project to investigate this further.
Regarding (ii) and (iii): Relativity theory implies that there is no distinguished frame of reference, all reference frames are equally valid. This breaks with the objectivism inherent in classical mechanics. The way I see it, quantum mechanics can be understood as a further radicalization toward non-objectivism. It has been argued that contemporary physics is developing toward theories that “do away with the idea of entities” (Grinbaum 2017) and that the success story of modern physics can be constructed as a story of abandoning metaphysical hypotheses (Mittelstaedt 2011). We don’t know whether a future successful theory of quantum gravity will be similarly amenable to phenomenological interpretations as quantum theory. At this point, the more interesting question is whether phenomenological approaches to modern physics can lead to conceptual clarifications which in turn could be helpful for research in quantum gravity. But this may be an excessively bold hope. The more down-to-earth takeaway is this: Quantum theory is our currently most successful scientific theory (in terms of accuracy and range of applications). Instead of trying to make sense of quantum theory within an objectivist framework, more philosophers should explore the philosophical implications of non-objectivist interpretations such as QBism.
1.4. Question 4
Khalili’s fourth question concerns the nature of the wave function and the explanatory power of quantum mechanics. In philosophy of science, the no-miracle argument says that the success of science would be miraculous if our best scientific theories did not represent reality. Khalili points out that an argument of this type can be raised against QBism. According to QBism, the wave function is not a physical object and it also does not represent physical reality. It only represents subjective degrees of belief and quantum mechanics can be understood as providing us with the mathematical tools that help the experiencing agent to be consistent in their decision-making. Khalili objects that the explanatory and predictive power of quantum mechanics seem miraculous if QBism is true. I don’t agree that there is a problem regarding the predictive power. QBists consider quantum mechanics the best tool we have to predict future experiences. Thus, if QBism is true, the predictive power of quantum mechanics is hardly miraculous. But the following question seems to strike a chord: If QBism is true, how to explain quantum phenomena such as interference patterns? Of course, QBists are very much interested in the following question: If QBism is true, what does quantum mechanics tell us about reality? One lesson, according to QBists, is that “reality is more than any third-person perspective can capture” (Fuchs 2017, 113). Unfortunately, however, not much progress has been achieved regarding the explanatory challenge. For instance, why is there an interference pattern when we fire electrons toward a double slit? It seems that Bohmians, for example, are in the virtuous position to be able to provide a straightforward physical explanation centered around the notion of particles. QBists, by contrast, are not in a position to provide a similarly physical explanation.
Where does this leave us? In my view, it is helpful to look at special relativity. Relativistic phenomena such as time dilatation and length contraction, similar to quantum phenomena, are highly counter-intuitive in the sense that what we observe in our daily lives seems to contradict them. Length contraction is the phenomenon that if an object moves relative to an observer, the length of the object as measured by the observer is shorter than the length as measured by an observer that is at rest relative to the object. Einstein “explained” such relativistic phenomena by showing how they follow from two simple principles: the principle of relativity and the light postulate. But note that this “explanation” also fails to constitute a mechanistic explanation. In fact, prominent contemporaries of Einstein, most notably Hendrik Lorentz, did not accept Einstein’s explanation, looking for a more physical alternative. Recently, physicists working on the foundations of quantum theory, have suggested providing quantum mechanics with a “reconstruction” such that the formalism can be derived from simple physical principles similar to how the formalism of special relativity can be derived from Einstein’s two principles (see, e.g., Goyal 2023). The idea is that a better understanding of quantum theory can be achieved by specifying an analyzing the physical principles from which it can be derived. In my view, the most promising way for QBists to respond to the explanatory challenge is to explain quantum phenomena by means of suitable reconstructions. Accordingly, it is no surprise that QBists, in particular Christopher Fuchs, have been involved in the quantum reconstruction program since the very beginning of this program in the early 2000s. However, although there are now several successful reconstructions, QBists are still waiting for a reconstruction that allows making sense of quantum mechanics on the basis of more straightforwardly physical principles (Fuchs & Stacey 2016).
A final remark: Attempts to derive the quantum formalism from simple principles have existed since the 1930s but only since the rise of quantum information theory notable progress has been achieved. All the successful reconstructions we know today are based on information-theoretic principles. The success of information-based reconstructions could be understood as putting pressure on ψ-ontic interpretations. We return to this below.
1.5. Question 5
Khalili’s fifth question also concerns in what sense QBism could be understood as a realist account. QBists strongly oppose being labeled as anti-realists and insist that they do believe that quantum mechanics offers substantial lessons about the nature of reality. We mentioned the QBists’ prime example in the previous section, namely that quantum mechanics, according to QBists, teaches us that “reality is more than any third-person perspective can capture.” Khalili is not satisfied by this, arguing that for a perspectivist this is almost trivial. Khalili then addresses David Glick’s interpretation of QBism according to which QBism amounts to a normative realism: “quantum theory provides correct answers to questions about what we should do, and provides us with reasons to do as it prescribes. Thus, it would seem that QBism can meet the demands of a conception of realism appropriate for a normative theory” (Glick 2021, 18). Khalili is not convinced and argues that such a normative realism “cannot elucidate how those who adhere to quantum mechanics can enjoy predictive and explanatory power, embodied in technologies that function effectively without caring about guiding our beliefs and actions.” I disagree. Consider, for instance, the no-cloning theorem. The theorem follows from the formalism of quantum theory. Thus, QBists are committed to the assumption that we should expect to experience outcomes as predicted by the theorem. In fact, if results were observed that contradict the no-cloning theorem, this would undermine the QBist assumption that quantum theory offers the best tool to predict what to experience next. Of course, it does not matter whether an agent actually believes in quantum mechanics or belief/action guidance by quantum mechanics. Consider normative realism in ethics. If in a certain situation it is objectively morally wrong for subject S to perform action A, S’s performing A is morally wrong whether or not S cares/knows about this. Similarly, expecting outcomes consistent with the no-cloning theorem is what an agent should expect, whether or not they know about quantum mechanics. Thus, I don’t see why QBism understood as a normative realism cannot explain why quantum mechanics can be successfully applied in technologies.
Khalili then moves on to a claim he defends in philosophy of science, namely that persistence in different perspectives is an important indication that some aspect of a theory is objective. This brings him to the following question:
“the most objective aspects of our scientific knowledge concern those that persist in appearing in different perspectives. But what are those aspects of quantum reality, if any, that persist in the perspectives of different observers? This is my fifth question.”
The way I understand Khalili, with respect to special relativity we can say: phenomena such as time dilatation and length contraction have been independently observed in various contexts and measurements, thus we are justified to believe that these phenomena are objective. In my view, the exact same thing can be said about quantum phenomena such as entanglement, tunneling, interference patterns, no-cloning results, etc. The interesting question is whether we can derive these phenomena from a small set of physically meaningful principles similarly as it is the case in special relativity.
1.6. Question 6
Khalili’s sixth question concerns the underdetermination in quantum mechanics. As is well known, quantum mechanics can be and has been interpreted in radically different ways. As David Mermin famously put it: “New interpretations appear every year. None ever disappear” (Mermin 2012, 8). Many view this situation in quantum mechanics as a prime example of epistemic underdetermination.[2] Researchers working on reconstructing the quantum formalism (as discussed in Section 1.4), have argued that this plethora of diverging interpretations is a symptom of the misguided approach of treating the formalism as a given and trying to read off an ontology from the mathematics involved. Instead, the ambition behind the reconstruction program is to gain a better understanding of quantum theory by first clarifying where the mathematics comes from. For instance, why is the mathematics of complex Hilbert spaces so successfully applied in our attempts to describe nature? The hope is that reconstructing the formalism from a small set of operationally meaningful principles will shed light on such questions. However, it turns out that there is not a single, unique reconstruction. Instead, by now there exist several equally successful reconstructions. One of the research questions I’m interested in is whether it would be possible to derive quantum mechanics not from information-theoretic principles but from phenomenological-epistemological principles about the nature and justificatory force of experience. Khalili wonders whether offering such a reconstruction would worsen the underdetermination problem: “will the reconstruction of quantum mechanics from phenomenological-epistemological principles, and its potential to result in a new interpretation of quantum mechanics, not exacerbate the underdetermination problem?”
In this context, Khalili addresses the optimistic thesis expressed in (Barzegar & Oriti 2022) that reconstructions, along with no-go theorems, may help us to rationally favor one (type of) interpretation over the others and thereby contribute toward solving the underdetermination problem. Khalili questions this optimism, wondering whether new reconstructions will lead to new interpretations and thereby increase the degree of underdetermination.
I would like to make two comments on this: First, while I agree that it is unrealistic to assume that reconstructions will help us to single out one specific interpretation, I do believe that reconstructions can make a great contribution regarding the question of how to interpret quantum mechanics by supporting or undermining certain kinds of interpretations. For instance, it has been argued that successful information-based reconstructions put some pressure on ψ-ontic interpretations and cohere better with ψ-epistemic interpretations (see Koberinski & Müller 2018). If this is true, given that contemporary philosophy of quantum mechanics is dominated by ψ-ontic interpretations, reconstructions could have a great impact on the question of how to interpret quantum mechanics. Second, philosophers have been extensively working on interpreting quantum mechanics for several decades. However, the reconstruction program is a rather new project, and while it enjoys some popularity at the foundations of physics, so far not many works have philosophically reflected on successful reconstructions. I hope that in the near future more philosophers will engage with the reconstruction program. At this point, if it turned out that a reconstruction in terms of phenomenological-epistemological principles is possible, it would be a highly welcome luxury “problem” if this would add to the underdetermination problem by leading to further phenomenology-friendly interpretations.
2. Wiltsche’s two potential risks
Wiltsche nicely summarizes my approaches to epistemology and philosophy of physics, elaborates on how I try to connect them, and highlights virtues and potential risks of this endeavor. Wiltsche is particularly interested in the relationship between mathematical models and the life-world, specifically in how to “bridge the potential divide between the lifeworld and the realm of science.” For him, “[t]he cardinal sin of objectivism lies in the reification of mathematical models and thus in the mistake to take for true reality what is a method.” However, Wiltsche contends that simply rejecting objectivism will not do the job of preventing science and life-world from drifting apart. Instead, what is required according to Wiltsche, is to perform a radical Husserlian Crisis-style “constitutional archaeology” of the mathematical tools and concepts we use in science and our reflections on science. In this context, Wiltsche addresses “two potential risks” of my approach. Here is the first one:
“To begin, Berghofer’s argument exhibits a certain tendency to select the interpretation of quantum mechanics that aligns best with his preferred high-level theory in epistemology. While I acknowledge the necessity of starting from phenomenological descriptions to provide the basic tools for our interpretational endeavors, I am cautious about rigidly maintaining a high-level epistemological framework while attempting to understand physical theories. Such an approach may hinder us from learning valuable epistemological insights from physics itself. This would be particularly detrimental in the context of quantum mechanics, as rightly pointed out by figures like London and Bauer, Heisenberg, Bohr, and others. Therefore, I would encourage a more ‘hermeneutic’ approach that allows for a more flexible oscillation between high-level epistemology and physics without rigidly prioritizing one over the other.”
The background here is that Wiltsche and I agree that it should be a characteristic feature of phenomenological approaches to physics to embody a “physics first!” approach (Berghofer & Wiltsche 2023). In the context of quantum mechanics, this means that when we interpret the theory, we should not start with ontological presuppositions and try to make the theory fit these presuppositions. Instead, we should attempt to look at the theory as unbiased as possible and respect the (quantum) phenomena. For instance, modificatory interpretations such as Bohmian mechanics or objective collapse theories seem to violate this “physics first!” principle. On the other hand, I seek to make several connections between my phenomenological experience-first epistemology and phenomenological approaches to quantum mechanics. Hence Wiltsche’s worry that my epistemological presuppositions may influence my understanding of quantum mechanics. I agree that this is a potential risk but I hope it is one worth taking. In my view, the prospect of establishing a unified conceptual framework that governs epistemic and scientific justification and allows us to interpret and perhaps even reconstruct quantum mechanics is very attractive. I believe that at this point both projects, i.e., a phenomenological experience-first epistemology and phenomenological approaches to quantum mechanics, are sufficiently well-developed to begin inquiring how they could be related to and support each other.
Here is Wiltsche’s second worry:
“However, as I have tried to show, the rejection of objectivism is simply not sufficient for preventing science and lifeworld from drifting apart. Even if we discard the overt reification of mathematical models, it remains a challenge to integrate physical theories like general relativity or quantum mechanics into a phenomenological framework due to their reliance on numerous essential idealizations. This holds true when we represent embodied subjects as points in the space-time manifold or as origins of coordinate systems. However, it also applies when we employ complex numbers, the infinitesimal, Hilbert spaces, or if we associate the wavefunction with the perceptual horizon, as some interpreters of QBism do. In all of these cases, our anti-objectivism does not shield us from allowing unexamined idealizations to foster a division between science and lifeworld. As I perceive it, the only way to avert this danger is to adopt Husserl’s approach in the Crisis. This entails engaging in thorough, methodical, and historically informed constitutional analyses of the ideal building blocks on which modern scientific practice rests.”
The first thing I want to emphasize in this context is that I agree that such constitutional analyses would be of great significance. This brings us back to Khalili’s first two questions regarding the relationship between scientific observations and theories. These questions regarded my thesis that all scientific insights can be traced back to epistemically foundational experiences. In my response to Kahlili, I clarified that such epistemically foundational experiences can be understood as ordinary experiences. In the present context, we note that also scientific concepts, however abstract they may be, can be traced back to what is originally given in experience. The second thing to note is that although Wiltsche’s quest regarding a “constitutional archaeology” and Husserl’s insistence that the scientist must “inquire back into the original meaning of all his meaning-structures and methods, i.e., into the historical meaning of their primal establishment,and especially into the meaning of all the inherited meanings taken over unnoticed in this primal establishment” (Husserl 1970, 56) may appear to be overtly abstract philosophical endeavors of little use to the physicist, the basic idea is rather simple and has been advocated by a number of prominent physicists. Einstein put it particularly nicely when he said:
Concepts that have proven useful in ordering things easily achieve such an authority over us that we forget their earthly origins and accept them as unalterable givens. Thus they come to be stamped as ‘necessities of thought,’ ‘a priori givens,’ etc. The path of scientific advance is often made impassable for a long time through such errors. For that reason, it is by no means an idle game if we become practiced in analyzing the long commonplace concepts and exhibiting those circumstances upon which their justification and usefulness depend, how they have grown up, individually, out of the givens of experience [my emphasis]. By this means, their all-too-great authority will be broken. (Einstein 1916, 102; cited in Howard 2014, 358)
This is to say that by conducting intentional analyses in which we inquire back into how scientific concepts emerged from ordinary experience, we may not only contribute to preventing the world of science from drifting too much apart from the life-world, but these analyses may also help science more directly in that we become more flexible in our thinking.
My third comment brings us back to the quantum reconstruction program. As noted above, proponents of the reconstruction program want to inquire why quantum mechanics has the very mathematical structure it has. The prospect of clarifying highly technical scientific notions by virtue of deriving them from operationally meaningful principles has been expressed as follows:
In short, the postulates of quantum theory impose mathematical structures without providing any simple reason for this choice: the mathematics of Hilbert spaces is adopted as a magic blackbox that ‘works well’ at producing experimental predictions. However, in a satisfactory axiomatization of a physical theory the mathematical structures should emerge as a consequence of postulates that have a direct physical interpretation. By this we mean postulates referring, e.g., to primitive notions like physical system, measurement, or process, rather than notions like, e.g., Hilbert space, C∗-algebra, unit vector, or self-adjoint operator. (D’Ariano et al. 2017, 1)
It seems highly reasonable to expect that successful reconstructions from operationally meaningful principles can greatly contribute to preventing quantum mechanics and the lifeworld from drifting apart. In this sense, reconstructions and constitutional archaeology are complementary projects (see Berghofer et al. 2020). I may, again, emphasize that the link between the life-world and science would be established to be particularly close if we could offer a reconstruction in terms of phenomenological-epistemological principles about the nature and epistemic role of experience.
3. Reichenberger’s worry
Reichenberger points out that (i) experience as a philosophical term is a technical concept with a long history and many different meanings and that (ii) a subject’s experiences can be misleading. Both points can be understood as undermining my experience-first approach. More precisely, this is what Reichenberger says:
“For me, highly philosophical terms and topics such as ‘experience’, ‘justification’, ‘consciousness’, ‘reality’, ‘illusion’ et al. are big names with a long history and a wide range of possible meanings. Every physics and maths teacher knows how misleading students’ everyday experience can be when it comes to calculating acceleration when driving a car.”
In contemporary philosophy, there exist various detailed but opposing views on the nature and justificatory force of experience. In this sense it is true that “experience” is a technical concept. On the other hand, our experiences constitute what is most familiar to us. Regarding the laptop I’m currently looking at: I don’t know how it works, don’t know about its interiority or the atoms it consists of. But I know how it is for me to look at it. As discussed above, in my view, “scientific experiences” are structurally similar to ordinary experiences: observing a pattern or looking at numbers on a screen, reading the display of a clock, etc. Of course, these experiences, then, play a role in justifying highly complicated scientific theories, and, correspondingly, the line of justification is highly complex, and there is much to discuss about what exactly these experiences justify us to believe in. (E.g. whether experiencing what a theory predicts justifies to believe in the entities postulated by the theory.) Unavoidably, such discussions get very technical but this does not mean that the experiences as such are in any problematic sense artificial, technical, or remote from our life-world.
Reichenberger’s second point, namely that one’s experience can be misleading, sheds light on a very interesting tension/paradox. On the one hand, scientific justification and concepts can be traced back to ordinary experiences. On the other hand, our everyday experiences are not a reliable guide for modeling the world of science as quantum objects, for instance, behave very differently than ordinary objects. In other words, it would be a problematic prejudice to assume that always and in all regimes the world behaves as we know it from our life-world. The way I see it, this apparently paradoxical situation does not constitute a contradiction. In particular, this does not imply that ordinary experiences cannot be a source of justification or cannot constitute the ultimate evidence of scientific beliefs. However, it does mean that we must be very careful in our inferences and in our ambition to approach scientific theories and phenomena as unbiased as possible. Generally speaking, in my view, the lesson from the theory-ladenness of experience is not that we cannot trust our experiences but that we must be aware of its limitations and perspectival character.
Finally, Reichenberger addresses the topic of agency:
“[A]gency always operates within and through a social structure and cultural practice. Really? What drives whom? Is every kind of agency a human agency? Is every human agency a responsible agency? Last, but not least: What is meant by ‘agency’ in quantum physics? It seems to me that there is still a great need for clarification here.”
Regarding agency in quantum mechanics, it is interesting to see how differently this is approached in different interpretations. In the objectivist interpretations dominating contemporary philosophy of physics, so-called “quantum theories without observers” such as Bohmian mechanics and the many-worlds interpretation, agency should not play a role at all. The formalism is supposed to objectively describe how external reality evolves in time. No reference to an agent should be required. In QBism, by contrast, the concept of the agent is central and irreducible. The formalism has the function of telling the experiencing agent to predict what they will experience next and to improve their decision-making. QBists concede that they have not defined the notion in a rigorous manner. Perhaps the most sophisticated elaborations can be found in (Fuchs 2023). Here Fuchs characterizes the concept as follows: “An agent is an entity that can freely take actions on parts of the world external to itself and for which the consequences of its actions matter to it” (82). This is how consciousness enters the scene in QBism. An agent is not just any physical entity, it is a subject to whom “the consequences of [their] actions matter.” Electrons and stones do not qualify as agents. For a subject to freely take actions and care about the consequences of their actions, (self-)consciousness is required. In this sense, QBism seems to pave the way for a physics that qualifies as a “human physics” as required by Merleau-Ponty (2003, 97), a “physics with a human face” as elaborated by (Wiltsche 2022).
References
Barzegar, Ali & Oriti, Daniele (2022): “Epistemic-Pragmatist Interpretations of Quantum Mechanics: A Comparative Assessment,” https://arxiv.org/abs/2210.13620.
Berghofer, Philipp & Wiltsche, Harald (2023): “Introducing Phenomenology to QBism and Vice Versa: Phenomenological Approaches to Quantum Mechanics,” in P. Berghofer & H. Wiltsche (eds.): Phenomenology and QBism: New Approaches to Quantum Mechanics: Routledge, 1-46.
Berghofer, Philipp; Goyal, Philip; and Wiltsche, Harald (2020): “Husserl, the Mathematization of Nature, and the Informational Reconstruction of Quantum Theory,” Continental Philosophy Review 413-436.
D’Ariano, Giacomo; Chiribella, Giulio; and Perinotti, Paolo (2017): Quantum Theory from First Principles, Cambridge: Cambridge University Press.
Fuchs, Christopher (2017): “On Participatory Realism,” in Ian Durham & Dean Rickles (eds.): Information and Interaction, Cham: Springer, 113-134.
Fuchs, Christopher (2023): “QBism, Where Next?” in P. Berghofer & H. Wiltsche (eds.): Phenomenology and QBism: New Approaches to Quantum Mechanics, Routledge, 78-143.
Fuchs, Christopher & Stacey, Blake (2016): “Some Negative Remarks on Operational Approaches to Quantum Theory,” in: G. Chiribella & R. Spekkens (eds.): Quantum Theory: Informational Foundations and Foils, Dordrecht: Springer, 283-305.
Glick, David (2021): “QBism and the Limits of Scientific Realism,” European Journal for Philosophy of Science 11, 1-19.
Goyal, Philip (2023): “The Role of Reconstruction in the Elucidation of Quantum Theory,” in P. Berghofer & H. Wiltsche (eds.): Phenomenology and QBism: New Approaches to Quantum Mechanics, Routledge, 338-389.
Grinbaum, Alexei (2017): “How Device-Independent Approaches Change the Meaning of Physical Theory,” Studies in History and Philosophy of Modern Physics 58, 22-30.
Howard, Don (2014): “Einstein and the Development of Twentieth-Century Philosophy of Science,” in M. Janssen & C. Lehner (eds.): The Cambridge Companion to Einstein, Cambridge: Cambridge University Press, 354-376.
Husserl, Edmund (1970): The Crisis of European Sciences and Transcendental Phenomenology, transl. by David Carr, Evanston: Northwestern University Press.
Koberinski, Adam & Müller, Markus (2018): “Quantum Theory as a Principle Theory: Insights from an Information-Theoretic Reconstruction,” in M. Cuffaro & S. Fletcher (eds.): Physical Perspectives on Computation, Computational Perspectives on Physics, Cambridge: Cambridge University Press, 257-279.
Merleau-Ponty, Maurice (2003): Nature: Course Notes from the Collège de France, Evanston: Northern University Press.
Mermin, David (2012): “Commentary: Quantum Mechanics: Fixing the Shifty Split,” Physics Today 65, 7, 8-10.
Mittelstaedt, Peter (2011): Rational Reconstructions of Modern Physics, Springer.
Wallace, David (2022): “The Sky is Blue and other Reasons Quantum Mechanics is not Underdetermined by Evidence,” arXiv:2205.00568.
Wiltsche, Harald (2022): “Physics with a Human Face: Husserl and Weyl on Realism, Idealism, and the Nature of the Coordinate System,” in H. Jacobs (ed.): The Husserlian Mind, Routledge, 468-478.
[1] In Husserl’s terminology, we would say that what is originally given to the experimenter are the clock displays.
[2] For a dissenting view, see (Wallace 2022).