Measuring Things That Measure You: Complex Epistemological Practices in Science Applied to the Martial Arts

Abstract

We argue that an epistemology of martial arts is at least as complex as advanced epistemological positions available to the philosophy of science. Part of the complexity is a product of the epistemic relation between the knower and known, or the scientist and the object of inquiry. In science, we measure things without changing them and, sometimes, complex systems can change as we measure them; but, in the epistemology of sport that we are interested in, each measurer is also an object of inquiry. As such, each martial arts practitioner has to use various epistemic tools to measure a responsive system. We proceed in three steps. First, we discuss three epistemological frameworks in the philosophy of science—perspectivism, productivism, and distributed cognition. Second, we develop an epistemology of martial arts that features components from each of those epistemic frameworks. Third, we close the paper with a brief discussion about the unique complexity available to the martial artist, focusing on the responsive measurements that occur between two systems.

1. Introduction

The philosophy of martial arts (and the philosophy of sport, in general) is viewed by some as a fringe sub discipline, or as more of a fun side hobby than a serious philosophical enterprise. We challenge this sentiment. In particular, we argue that martial arts practices contain epistemologies that are at least as nuanced as the most complex epistemological positions in the philosophy of science. To defend this claim, we present an epistemology of martial arts where the practitioner is measuring phenomena that also serve as measurers. In science, we measure things without changing them and, sometimes, complex things can change as we measure them; but, in the epistemology of sport that we are interested in, each measurer is also an object of inquiry. As such, the martial arts practitioner has to use various epistemic tools to measure changing systems, thus testing the validity of a given martial arts practice. We argue that in this complex epistemic activity, a given martial arts system stabilizes robustly across various contexts over time. We build the epistemic approaches that are represented in martial arts by using more traditional epistemological standards in contemporary philosophy of science, focusing on the reciprocal relations that exist between two systems. Suddenly, the philosophy of martial arts does not seem all that fringe. Rather, it participates in highly relevant epistemic themes that feature in more standard domains of philosophy [1,2,3,4,5,6,7,8,9,10,11,12,13]1.

We proceed in three broad steps. First, we chart out three epistemic frameworks from the philosophy of science, i.e., perspectivism, productivism, and distributed cognition. These frameworks allow us to track the versatility and exchange of knowledge across a community. We then apply these frameworks to specific martial arts examples and discuss three ways in which a given martial arts system can be tested upon. In particular, we demonstrate that a given system can be tested intracommunally, intercommunally, and extracommunally. These different testing methods allow systems to stabilize in different and unique ways. We also discuss how these testing methods relate to self-defense scenarios. In the final section of the paper, we argue that what makes an epistemology of martial arts so complex is that the measurements being made depend on reciprocal relations.

2. Epistemic Frameworks in the Philosophy of Science

There are ongoing questions of how best to model scientific knowledge. Is the scientist developing an ‘accurate’ representation of real systems, or phenomena? In what ways is the scientist involved in experimentation and modeling? How should we understand the development of scientific knowledge over time? Or in a community? All of these questions touch on the epistemic relation between the knower and known, or the scientist and the object of inquiry. So, this section introduces three epistemic frameworks common to the philosophy of science. In particular, we discuss perspectivism, productivism, and distributed cognition. We do not claim that this is an exhaustive survey, but that these views offer a sufficient survey for our purposes.

Naïve realism is the general epistemic view that we have unmediated access to the world [14]. However, more nuanced, less-naïve iterations of our relation to the phenomena have cropped up over time. We see this happening, in particular, with perspectivism. According to perspectivism, scientific modeling and measurement provides vantage points on real systems, and such vantage points are partially adequate [15,16]. The relation between the knower and the known becomes versatile. Perspectivism allows for degrees of ‘accuracy’ as well as ‘adequacy’. That is, even if a given model of a real system is false overall, it can still “get things right” in aspects and degrees.

One key issue with perspectivism is that there are deeper relations between knower and known. For instance, there are ways in which scientific methodology is entangled with the real systems that are being investigated. In other words, in order to measure and model a given system, there is active methodological engagement, which influences the structure of measurement setups and measurement outcomes [16], even to such a degree that new phenomena or real systems can be “produced” [16,17,18,19,20]2.

Productivism [16] is complementary to perspectivism in the sense that both are in agreement about the limitations of scientific models and measurements. But in the latter, attention is often paid to the relation between developed models/measurements and phenomena, while in the former, attention is paid to the process of development in modeling and measurement practice. Productivism focuses on the process of how we stabilize reliable measurements and models with full transparency about how scientific practice is causally entangled in the stabilization of measurement setups, outcomes, models, and new phenomena [16,17]. The benefit of productivism as a methodology is that it explicates how complex things can change as we measure them and to what extent we can maintain reliability in such contexts.

Scientific practice is not reducible to the relationship between experimenter and real system. While that relationship is relevant, there are many ways in which science proceeds through the interaction between multiple perspectives, methods, and models [16,21]. The structure and process of scientific activity can be framed as a process of ‘distributed cognition’ [22]. Multiple scientists, apparatuses, and technologies work in a coordinated relation to achieve an activity or output [22]. It is simplistic to say that the scientist uses an apparatus in order to take a measurement. Instead, a system of complex interactions works to achieve a given epistemic goal [22]. The emphasis is on the relationships between multiple independent perspectives and components, working autonomously but also in coordination, to achieve an epistemic goal. The benefit of distributed cognition as an epistemic framework is the acknowledgement of just how many moving relationships require coordination for something as simple as a measurement outcome, or something as large as a theoretical model, to be realized.

When boiled down, perspectivism, productivism, and distributed cognition provide useful epistemic tools for a given system of inquiry. In perspectivism, the relation between the knower and the known becomes versatile—such that accuracy can vary on a continuum. In productivism, that relation between knower and known becomes more complicated because things can change as we measure them, but we can still maintain some sort of epistemic ‘stability’ in such contexts. Finally, in distributed cognition the lens broadens to acknowledge ‘stability’ as a distributed activity between agents and technologies.

3. An Epistemology of Martial Arts

A robust knowledge system should be able to explain the process of testing the aptness or validity of a given belief. Similarly, a robust epistemology of martial arts should be able to explain the process of testing the aptness or validity of martial principles and techniques. In this section, we argue that martial arts do have robust epistemological standards for testing systems, and that these epistemological standards can account for the internal and external validity of a given system. We proceed in two steps. First, we discuss the epistemic differences between drilling, sparring, and competing. We also define ‘chaos’ as a highly relevant term for the epistemology of martial arts. Second, we explicate multiple ways in which these practices test the validity of martial systems, both internally and externally.

3.1. Degrees of Chaos: Drilling, Sparring, and Competition

In our 2021 paper, “Experimentation, Distributed Cognition, and Flow: A Scientific Lens on Mixed Martial Arts”, we argue that combative techniques become distributed throughout a given practitioner’s cognitive and physiological system, such that, when placed in changing environments, the system itself responds automatically, and with minimal cognitive effort from the practitioner. This, in short, is a flow state in martial arts—the practitioner has saturated their cognitive and physiological system so thoroughly with a given technique(s) that when placed in a chaotic environment, the technique, distributed throughout a system, performs itself [23,24,25,26,27,28]3. Notice here that the stabilization of the precision and effectiveness of a given technique within a given context (internal validity) extends to new contexts (external validity) [29]4. There are a variety of ways to saturate a cognitive and physiological system appropriately, including drills, sparring, and regular competition. Let us discuss the epistemic role of each of these items.

Drilling is essential for a technique to become distributed throughout a practitioner’s cognitive and physiological system. As a jazz musician might practice their scales to improve their improvisation, so too does a martial artist regularly drill techniques in a controlled environment so that they might be more effective in changing contexts [30]5. Drilling allows the practitioner to feel their way through a technique, think about possible counters that the opponent might pull, or even just habituate and reinforce a technique they already intimately know. But the controlled environment of drilling comes in degrees. At the highest degrees of control, drilling requires no effort from the partner—indeed, it might not require a partner at all. For instance, a judoka might require no effort from their uke (the one having the technique performed upon) while practicing their favorite throw; a nak muay may not require a partner at all and just drill their teep and round kick on a heavy bag; or a boxer might not need any equipment and just shadowbox. At the lower degrees of control, drilling requires the variation of elements, provided through the effort of the partner. This may look different across different arts, of course. In boxing, it might look like a mitt holder adding changes in cadence. In grappling arts, it might look like a grappler taking a disadvantaged position and having their partner resist their escape at 50% intensity. There are countless drills that require some fluctuating elements, provided by an active training partner. As we have argued elsewhere, this type of activity isolates and stabilizes specific techniques to be used within particular contexts [23].

A training regimen with no sparring (of any kind) is like a gardener planting seeds without ever watering them. Sparring serves the very practical epistemic end of allowing the practitioner to test or practice their techniques in controlled contexts. If things get out of hand, you are generally safe (barring any major mishaps), but you are still gauging an opponent’s distance, speed, and technical prowess. It is a simulation of uncertain contexts—a construct with a specific set of conditions that we place ourselves under to further embed techniques in our cognitive and physiological systems. In uncertain contexts (varying by degree), the practitioner can apply the techniques that they have been drilling, revisit their ‘go-tos’, or help newer practitioners become comfortable under changing conditions.

Competition is not something that every martial arts practitioner engages in, which is absolutely fine. But, it does serve a very specific end for an epistemology of martial arts. It offers a minimally controlled environment under a given rule set, wherein a practitioner can test the validity of their system against another practitioner of that system or some other system. What differs here, between sparring and competition, is often the degree of familiarity between the two fighters and the limited care that one has for the other’s well-being. Often, in sparring, we are fighting our friends; but, in competition (more often than not), we are fighting a stranger. This is a significant epistemic difference. For one, an opponent in competition represents a greater unknown—we simply do not know them as a person or a practitioner. But further, in sparring, in trying to “measure up to” (a phrase which will shortly become relevant) our friends, we are also trying to make our friends better. There is no such friendly spirit in competition (there may very well be friendly exchanges before or after the match). At stake is not just “measuring up to” but also besting the other practitioner—this applies across the board in martial arts, from Taekwondo to MMA. This is evidenced in cases where training partners meet in competition. The comradery is set aside for the allotted time. Notice that this makes the application of isolated, stabilized, and simulated activities multidimensional. The martial artist is attempting precision and effectiveness of a technique, but it is not in a vacuum (e.g., like it is in drilling), and it is also not in a controlled context where certain variables are held constant, while others are changed. The latter is important to emphasize. In sparring environments, variables like power and aggression can be modulated to aspects and degrees for the purpose of cognitive and physical stabilization. But in competition, variables multiply and evolve without external constraints. This is analogous to measuring in a laboratory with controlled conditions and then going out into “nature” and responding to new factors [17]. The former does not always prepare us for the latter.

What we have now is a better understanding of the various degrees of uncertainty and methodical control that are available to practitioners. Control is the counterweight of uncertainty. As we’ve already seen in this section, there are degrees of control, inversely proportionate to degrees of uncertainty. Controlled environments can vary from the fine-tuned practice during drilling to low stakes contexts of “measuring up to” an opponent in sparring. By shifting between contexts, the practitioner can saturate their cognitive and physiological systems sufficiently to help them deploy their techniques more effectively in varying situations. However, it is important to note that situations or contexts are not “more complex” solely as a function of being constituted by more variables (e.g., more techniques, less rules, more variations, etc.). There is a specific type of complexity in martial arts practice that interests us—the responsiveness between two systems. As we describe shortly, the notion of ‘chaos’ is intended to capture a co-sensitivity, -reactivity, -responsiveness, and -evolution between systems.

Throughout this section, we have been utilizing (and thereby slowly introducing) a technical term that we have yet to define, i.e., ‘chaos’6. This term is essential for our project, however, and needs to be better presented at this point. Technical definitions serve to highlight key features of chaotic systems. According to Shore [31], “Chaos is the stochastic behavior of deterministic physical systems, resulting from extreme sensitivity of the dynamical evolution of such a system to small changes in the initial conditions”. This definition highlights an important general component of chaos, relevant for our discussion: the sensitivity of systems. But for our purposes, we detour from the general definition because we are concerned with a specific kind of sensitivity—the responsiveness of a given system due to interactions with other systems, including systems of measurement [16]. That is, when a martial artist ‘measures’ their opponent, the opponent is not merely reacting with sensitivity to the state of affairs at hand. The opponent is also tracking, anticipating, deceiving, and building models of the other measurer. For our purposes, a given context in martial arts is ‘chaotic’ when it constitutes the sensitivity of two systems (i.e., two opponents) both reactive and responsive to the other system. Notice that this reactive and responsive measurement interaction occurs within a context that contains a relevant set of rules. In a sense, both systems have some sort of boundary conditions. So, a given context in judo is chaotic when each judoka is measuring the other, under the further conditions that each judoka has the aim of throwing, pinning, armbarring, or choking the other under the appropriate rule set. A given context in muay thai is chaotic when each nak muay is measuring the other, under the further conditions that each nak muay has the aim of accumulating points or knocking the other out under the appropriate rule set. Chaos in martial arts is a product of attempting to measure the external signs of another agent with the aim of performing a given technique on them when that very agent is engaged in the same activity. We use the term ‘chaos’ to blanket properties like co-sensitivity, -reactivity, -responsiveness, and -evolution because the contexts that we are describing yield highly complex co-measurements by opponents (with more dangerous than usual physical consequences).

Appropriately, by these definitions, self-defense scenarios represent the most chaotic and least controlled circumstances that a practitioner can be placed in—a context where harm is intended that is not mediated by an agreed upon rule set (and additionally, a context that may involve deadly weaponry). We do not include these in the above list, since the context is presumed completely uncontrolled. There is no way to sufficiently replicate it in a gym environment. The stakes are indeed so high that measuring the opponent accurately and having the appropriate techniques to deploy is of the utmost importance. Self-defense scenarios will serve an important role in evaluating the validity of a given martial system in the next section.

It is also important to note that the chaos and control defined here refer to the external relations between two practitioners and not the internal mental state of each of the fighters. That is, we are only describing the highly sensitive, reciprocated activity of measuring something that is measuring you. Ideally, the internal state of each fighter will be calm and composed under changing conditions. But the psychological aspects of practice are beyond the boundaries of our epistemic account.

3.2. Testing the Validity of Martial Systems

There are robust methods for testing whether a given martial system, represented in a particular practitioner or in a group of practitioners, is valid. The purpose of these methods is to generate evidence that a given martial art possesses an epistemology that allows knowledge to stabilize in unique ways across many different contexts.

Imagine a group of proficient karate-ka at the same dojo (this example can apply to any art—we do not mean to be picking on karate, specifically). This dojo drills and spars, but does not compete or collaborate with other dojos or systems. What is the nature and value of their knowledge? We wager that the knowledge possessed, though stable and even rigorous in that particular context, thus constituting internal validity, is highly insular and ultimately not applicable across many contexts7.

Now imagine a group of karate-ka at a different dojo. This dojo drills, spars, and even competes in karate, but it also encourages its practitioners to go out to other gyms and practice applying their techniques to non-karate practitioners within a different rule set. What we have presented here is a knowledge of karate, represented in its practitioners, that is stable, rigorous, and dynamic across other contexts. Sure, karate may not be effective across all contexts, but the practitioners develop a general sense of what does and does not work across multiple contexts.

What is happening in these two scenarios? We argue that the two schools are engaged in different epistemological methods, one leading to bounded knowledge and the other leading to dynamic knowledge. In short, the methods differ in the testing and experimentation of their respective systems. Martial systems can be tested intracommunally, intercommunally, or even extracommunally, and knowledge stabilizes uniquely when one, some, or all of these experiments are conducted within a given context.

When a martial system is tested intracommunally, the system is only tested within one particular gym’s community. The practitioners will drill with one another and perhaps even spar with one another, but they do not enter more chaotic circumstances, like dual meets or tournaments. Knowledge may be stable within the gym, insofar as the system is internally coherent and responsive to itself. This is akin to the first karate-ka dojo scenario, from above. The limited dynamism is not, necessarily, a consequence of the techniques deployed (though it could be), but rather a consequence of limited exposure to chaos. When a practitioner is not used to measuring new opponents that are measuring them (in other words, they are used to the same opponents), the degree of chaos available has a limited maximal cap. Intracommunal practice may stabilize, but it does not lead to dynamic knowledge.

Additionally, a particular gym not only contains agents but also repertoires, which are composed of methodological activities that embed techniques. For instance, taking a coarse-grained view of drilling and techniques: a given boxing gym can maintain particular types of drilling (e.g., mitt and bag work) with specific techniques, cadence, and other neurocognitive factors—such as reaction-time repertoires. Fighters like Georges St-Pierre, Israel Adesanya, Vasiliy Lomachenko, and Oleksandr Usyk have each implemented specific neurocognition-based reaction drilling into their training, but the reaction repertoires have drastically different profiles. Taking a fine-grained view, repertoires also pertain to types of communication and coordination between trainers and trainees. For example, boxing gyms can contain their own intricacies of communication—such as, signifying each punch with a number for shorthand communication during training and competition. As we have argued elsewhere [23], such intricacies are part of the distributed cognition of a given martial arts gym. If the coordinated cognitive effort is effectively planned, intracommunal repertoires become part of a secret recipe, serving as an advantage. Otherwise, intracommunal repertoires can serve as a rigidifying disadvantage.

When a martial system is tested intercommunally, the system is tested against other gyms and practitioners that operate within the same martial system. This would be like a muay thai gym testing themselves against another muay thai gym(s). The practitioners at each gym drill and spar with one another, but put themselves to the test through various forms of intercommunal competition by hosting fight nights, dual meets, or attending other forms of competition. Knowledge stabilizes in more rigorous ways through these forms of experimentation. This is because the practitioners are exposed to new, unknown (or, at least, less known) opponents, allowing the knowledge that has saturated their cognitive and physiological systems through intracommunal practice to be tested in chaotic circumstances. The subtle stylistic differences represented by gyms that practice the same system allow the practitioners to learn to measure a diverse set of opponents. It allows the practitioners of a given school to know if their intracommunal practices are effective, or whether new methods need to be put in place. Interestingly, if an intercommunal experiment fails, the failure in external validity can lead to a reconsideration of the contextual circumstances that prompted intracommunal internal validity. That is, if intercommunal competition fails, it results in a reconsideration of the repertoires contained in the intracommunal practice. This serves as a way of using intercommunal competition to retrospectively measure the effectiveness of intracommunal practices. This then extends the methodological rigor of the martial system in its ability to measure not just in prospect but also in retrospect.

But there is a third way in which a martial system can be tested, and it is represented most clearly in early mixed martial arts competitions (we will discuss why it is less represented in modern mixed martial arts in due time). We call this extracommunal experimentation. When a given martial system is tested extracommunally, the system is tested against gyms and practitioners that operate within different martial systems. This is where the dynamism of the epistemology of martial arts begins to shine. Here, systems stabilize relative to each other uniquely across a wide array of contexts, where systems are represented by individual practitioners. Systems may also be responsive to given experimental conditions, thus, actually improving over time, stabilizing in a new way that responds to previous weaknesses (or strengths), but this is not necessarily the case. As we saw in the second karate-ka case from above, there are hypothetical scenarios where given systems are tested extracommunally to learn the system’s limits and bounds.

We say that early mixed martial arts practices are especially representative of extracommunal experimentation because ‘mixed martial arts’ had not yet stabilized as a relatively generic style (very roughly, it involves training in Brazilian jiu jitsu, wrestling, and various striking arts). Early mixed martial arts pitted different systems, who had little exposure to one another, against each other and tested the validity of those systems in chaos. There was no mixed martial arts style or system, as is perhaps represented in its earliest form in pankration. Could a karate-ka defeat a wrestler? Could a BJJ practitioner defeat a muay thai fighter? Could a Kuk Sool Won fighter defeat a judoka? The scenarios were endless for early mixed martial arts. But as the conditions of the experiment of mixed martial arts became more familiar, practitioners converged on a system of techniques, strategies, and repertoires that have tended to externally translate to the octagon. And so, a new style of martial arts has stabilized. Nowadays, if you place a mixed martial artist in a BJJ competition, you are testing the grappling prowess of a mixed martial art system against a pure grappling art. You are testing whether the conditions under which a mixed martial artist grapples affect their ability to grapple under other conditions. So, now we can isolate components of a mixed martial arts system to test or fine tune their validity—thus being consistent with the fine-tuning aspects of perspectivism and the stabilizing aspects of productivism. (Recently, the spotlight has been on isolating MMA boxing components and testing those components under traditional boxing rules.) Overall, such fine-tuning serves as a stark contrast to the early stages of MMA where, to be externally valid, styles needed a complete renovation.

But as has been hinted already, extracommunal experimentation does not need to take the form of early mixed martial arts. It can take the form that we saw earlier in the second hypothetical karate dojo. Or, it can be represented in the Jeet Kune Do system, which exposes practitioners to the principles of various arts, allowing those arts to stabilize uniquely, as seen most clearly in the famous collaboration between Bruce Lee and Dan Inosanto [12,13]8.

Extracommunal experimentation allows a practitioner of a given martial system to know whether their system will be competent in a chaotic environment under a specific rule set. It is the most thorough and invasive intervention into a martial system in order to see how it holds up. However, it does not account for whether a given system maximizes preparation for self-defense scenarios. How can this be? How can a maximal intervention still not prepare one for a self-defense scenario? Importantly, this calls into question a lot of self-defense systems that “demonstrate” effectiveness in heavily controlled contexts. For instance, just by showing an intracommunal demonstration of a heavily rehearsed repertoire, one might mistakenly infer extracommunal validity. But, as we have presented, context is everything in the stabilization of a technique. Indeed, sometimes just one factor (e.g., stochastic movement) can throw an idealized demonstration off.

In 2020, UFC light heavyweight, Anthony Smith, was woken in the middle of the night by a home invader [32]9. What followed was, in his words, “one of the toughest” fights of his life. Even though his opponent was “a regular Joe”, he “had a hard time dealing with him”. He implies that the toughest part of the fight was the uncertainty. No matter what he threw, the home invader “took every single one of them and kept fighting me”. This is a striking example of the limitations of intervention in martial systems. When we enter into maximally chaotic circumstances, like a self-defense scenario with uncertain weapons and techniques, the measurement patterns that stabilize in our training simply have not given us enough. We cannot replicate in controlled situations the uncertainty of danger, in addition to the uncertainty of resiliency of a given opponent. The latter is striking because time and referees moderate a given match. But in this case, time and stoppage are no longer relevant parameters. We argue that this presents an unusual case of “measuring up to one’s opponent”. In a situation with boundless time, no external judgment/stoppage, and an unrelenting foe, how does one evaluate the effectiveness of a technique? Moreover, how do we compare extracommunal competition to a maximally chaotic self-defense scenario? We unpack this below.

The rigorous models that martial arts develop never allow us to stabilize in a fully chaotic context. It is important to emphasize what components fail to stabilize. While competition training prepares a fighter to meet a range of uncertain techniques, timing, power, etc., there is a certain neurocognitive stabilization of what to expect in competition. When factors like a low blow or headbutt occur, they are out of bounds and require pauses, deductions, and disqualifications. The uncertainty range in competition seems broader than it actually is. Moreover, it is neurocognitively modeled in a very rigid way. Though controversial, Dubois’s low blow against Usyk resulted in a momentary pause in order for everyone to “measure” the event. Fighters understand the uncertainty ranges of competition. But in a no-rules scenario (like the home invasion), the neurocognitive model of techniques, repertoires, and weapons drastically changes. The uncertainty of techniques, power, startle-effect, boundless time, etc. can constrict the range of what is effective, thus complicating the success of external validity. So, in a sense, a given model always approaches but never fully captures reality. This is because uncertainty ranges are always in flux. Contrast this with training and competition, where the models track environments that are mediated by styles and rule sets—that is, contextual features created by the practitioners themselves.

We have seen in this section that knowledge in martial arts stabilizes uniquely depending on the experimental interventions one applies to a given system. These interventions can test the validity or invalidity of a given system relative to the aims of practitioners. However, the interventions we have available never allow a system to stabilize with respect to a fully chaotic context. This section demonstrates that the martial arts can have robust epistemic methods for testing the validity of its systems. We turn to its application in science.

4. An Epistemology of Martial Arts Applied to Science?

Thus far, we have detailed an epistemology of martial arts that responds to experimental measures. This model is loosely based on three existing epistemological views in the philosophy of science—perspectivism, productivism, and distributed cognition. Perspectivism features insofar as practitioners track the selective adequacy of their techniques. Productivism features insofar as measurements are made in a reciprocal relation between practitioners—consistently updating the conditions under which they measure. Distributed cognition features insofar as the knowledge of a given system is distributed throughout a community across space and time, though represented, in part, in particular practitioners. We turn now to discuss the philosophical import of this view.

We maintain that what makes an epistemology of martial arts so exciting is the reciprocal relation between the measurers. It builds upon and extends an epistemology of science by placing the measurement apparatus (a given practitioner) in a situation where it is being measured back (by another practitioner), allowing for the stabilization and implementation of techniques. Typically, when a scientist measures a given system, it is not often presumed that the system is responsive to the measurement apparatus in the same way that an agent would be, and also that the system can update properties (e.g., repertoires) based on how it is being measured. That is to say, though an object placed within a given system may respond to experimental measures, it is not presumed that the object is anticipating, baiting, or substantively measuring the scientist back.

Of course, one may object and say that measurement and experimentation in psychology are relatively akin to the epistemology of martial arts. After all, part of what makes psychological experimentation so tricky is that scientists are measuring an agent, not an object. We maintain that even an epistemology of psychology, where scientists experiment on human agents, does not capture the complexity of the model operating within martial arts, since there is a relevant asymmetry between the scientist and the subject of study. That is, the psychologist controls the conditions of a given study, while the participant of a study is just that—a participant. Such an asymmetry does not exist in martial arts—agents are presumed equal relative to each other, and the experiment tests which system, represented within a particular practitioner, will effectively overcome the other.

5. Conclusions

Our general aim has been to illuminate a particular type of epistemology of sport, where each measurer is also a dynamic object of inquiry. Specifically, we have argued that in an epistemology of martial arts, the practitioner has to use various epistemic tools to measure a changing, responsive system. To do this, we first presented three epistemic frameworks—perspectivism, productivism, and distributed cognition. We utilized these frameworks to develop an epistemology of martial arts. This epistemology allows for the testing and experimentation of martial. We then argued that the epistemology of martial arts extends beyond the most complex epistemologies of science by noting that placing measurers in a reciprocal relation allows for asymmetric scenarios to emerge and stabilize.