Subscribe to DSC Newsletter

Fallacy of Rational Prerequisite & My Fruitless Existence

Before elaborating on my fruitless existence - about my decision to avoid fruit - I want to emphasize how this blog is actually about something that I call the "Fallacy of Rational Prerequisite." There will be some misunderstanding about this term even after my prolonged explanation. I just want to state plainly at the outset that I am not proposing that people become irrational. If they are already so, I am not suggesting that they further the situation. Consider an analogy. Imagine a society predisposed to biblical explanation. I belong to an organized religious faith. This blog is not about religion. But just recall the times and places in human history in which day-to-day problems and challenges invoked some sort of faith-based explanation. I would describe this more generally as an atmosphere of "belief-based" explanation. Whether the belief system is centered around religious faith, neoliberalism, or socialism, I suggest that when we come across hurdles in life, we rather routinely find ourselves reaching into our bag of beliefs and pulling out some kind of complicated belief-based rationalization. So this blog is about that type of rationalism. I will be questioning the need to fabricate rational explanations to justify, sustain, and manage organizational activities - the most pertinent to forum being the collection and analysis of data. The process of rationalization leads us to direct our focus on beliefs rather than production. I am not saying that beliefs are unimportant. I am just questioning the extent to which they are necessary in a goal-oriented organizational setting driven by data.

There have been comparisons between data science and computer science. Some have suggested that computer science is not a real science, and data science should be regarded similarly. I consider this an interesting debate. The divisiveness can easily cause a conversation to descend into tribalism and theocracy. The fight for epistemological territory ostensibly scientific has really been about controlling the recognition of data in the sense of asserting ownership. Consider a comparison to patriotic or religious fanaticism. "This is an evil act." "We are good people." "This is unacceptable in a civilized society." So sayeth nations that direct enormous budgetary resources to its military - waging war constantly. I realize that these seem to be declarations of ideology. I suppose it would be necessary to consider the situation from the perspective of my own duties where I make observations based on established criteria and personal experience. "This is a mistake." "This will cause problems if not corrected." "This has consequences." I believe if we examine how these assertions define the placement of data as objects of relevance, it would be apparent that underlying intent is to impose a belief system or rational structure over the construction of our lived realities. It is an intellectual incursion - viral in nature - sometimes purely functional and at other times premised on the need for authority and centralized power. We find ourselves basing our handling of material concerns over immaterial beliefs. Through capital, we give life to dogma. Yet dogma doesn't actually give us a return on investment. When we remove the tribalism and theocracy from the discussion, it should become apparent that these themes are philosophical. They can also be rather destructive or at least extremely distracting.

I believe that people have a tendency to rationalize situations, which I suggest means providing a plausible or convincing explanation that is intellectual in nature. A bear touching a fire might learn to avoid it in the future; most of us would agree that it is reasonable to do so. But the bear's behaviours are rooted in its immediate physical needs. Humans on the other hand perform and persist with behaviours without the need for physical stimuli. But we often require rational explanation to substantiate our behaviours: for example, whether as a stock-market speculator or furniture mover, many of us perform our duties and responsibilities on the expectation of monetary reward or compensation. Our rationalization might be incorrect, but we often expect and need it in some form to bolster our commitment. We also judge people by their rationale. The absence of rational explanation can sometimes lead to declarations of irrationality. Why would somebody work without some type of compensation? In a business setting, rationalism is tested against an entirely different creature called "performance." Companies that fail to test against performance cease to exist at some point, a situation that might be blamed on Natural Selection. It is perfectly fine for people to have beliefs, but the beliefs per se do not sustain an organism or organization. Even if we import beliefs from successful organizations, this does not mean that we import their success. I think much the same arguments would apply to the importation of a foreign data system, which might be introduced into a workplace essentially as a faith-based initiative. Sadly there are no door prizes. If the system fails to perform, typically we blame the lemming rather than the charismatic.

Many will recognize one particular reoccurring context in which different forms of rationalization have become persistent: through professional association. On one hand there is commitment to the organization; but on the other, there is loyalty to the profession itself. Some might say that belonging to a profession is quite separate from the issue of rationalization. It is not belonging to a professional association per se that has ever concerned me; because indeed there are many good reasons for such groups to exist and for people to be members. It is precisely that a profession tends to bring with it a paradigm of thinking that sometimes troubles me: one has to question why any kind of imported paradigm might be necessary if there is data. A paradigm is practically an abstraction that can exist without data. I believe that rational explanation is often meant to front-run the data - to create a slot or pigeon hole for it so its relevance to us is controlled. In a moment, I will be discussing fruit. I know some people must be thinking, why in the world fruit of all things? Well, 'tis the season for fruitcake. What is the first thing we are told about fruit? "Fruit is good for you." Without any data about me specifically or my experience with any particular fruit, nonetheless there is this normative. I call it a normative because it actually affects how people regard me for instance if I decide not to eat fruit. Without knowing anything about a company - that is to say, without having any data about a specific organization - there are people that consider it appropriate to simply promote a litany of neoliberal ideals, product lines, and easy 5-step solutions. This is "normal" rather than professional negligence. I go along with it. I don't question it. But I do question an organization that willingly makes itself a victim.

Some people belonging to the data science community probably come from disciplines that, far from trying to understand and adapt to organizational needs, seem preoccupied with the imposition of their disciplinary perspectives. I am not suggesting that they necessarily impose on the outcomes . . . although this might indeed be the case. I mean that they assert their disciplines over the process of reaching outcomes. Nothing can be more hazardous and pointless than characterizing the challenges confronting an organization in the context of a particular discipline or profession as a matter of alliance or intellectual colonization. The main reason I feel so is because an organization owes nothing to these interests. Nor do these interests necessarily recognize a debt or obligation to any organization. So for an organization to voluntarily submit itself to a profession or discipline is an act of betrayal and violation. There is data. There are decisions. There are outcomes. The fact that there are accountants, statisticians, and MBA graduates involved in some of the background processes should not in itself matter; thus, their contributions in terms rational explanation are equally irrelevant. We should seek out results rather than rationalization; returns on investment rather than excuses. It's fine if people belong to a church group, a choir, or recognized profession. But these connections are immaterial to production. Who says so? The market does. As evidence, I point out how companies can be entirely replaced by foreign imports. This is how extremely irrelevant rationalizations are - and by extension those whose primary role is to rationalize. I place much more emphasis on performance - and perhaps more importantly, the tangible (real and local) business contexts giving rise to our perceptions of performance. Consequently, I find myself in dichotomous opposition to (fabricated and foreign) paradigms of thinking heavily infused with rationalism.

We too frequently find ourselves confronting this philosophical argument that data must conform to our rational understanding in order to be valuable. I contend, however, in relation to large amounts of data where the focus is on discovery, it is adequate in most cases to be "reasonable and sensible." It isn't necessary to have any kind of supporting rationalization. In fact, attempting to define and impose reality through rational thought, while great for starting arguments, might actually threaten our adaptability. This is an important issue. I often encounter arguments diminishing the value of data-collection due to lack of rational explanation. The theocrats and tribalists might say for example, "It must be scientific." "It has to lead to efficiencies." "It has to concentrate on our existing product lines." These seem like harmless comments. But the data must overcome belief-based criteria in order to become significant. I want organizations to recognize that this paradigm is fine if one's intention is to serve beliefs. The effective impact however is to structure the organization of data the same way religion might cause us to organize facts in a certain way. Submission to the deity occurs at the outset. What organizations have to keep in mind is how this submission represents an incursion of a foreign entity into decision-making. We test not for performance but rather compliance. By successfully conforming to our rational perception of a situation, we do not necessarily further our performance. In fact, I suggest that the opposite is true. Rationalization is what we turn to when we lack data but need justification; when we want to appear correct and proper; and when we expect the need for a scapegoat. It is a product of ideology and social construction.

My Road to Fruitlessness

I have been posting blogs for some time. I find myself emboldened by the lack of negative comments from readers. (I realize there is always a first time for everything.) I want to share the investigation leading up to my decision to stop eating a major food group. I no longer eat fruit. My existence is fruitless. The purpose of this example is not to promote fruitlessness. Rather, I want to demonstrate how an elaborate "reasoning out" process incorporating large amounts of data is possible without resorting to rational explanation - or indeed any kind of explanation.  It just so happens that the lack of an explanation tends to trigger rationalization; its role is to diminish the need for data. A nation can invade another without any accurate data at all but only suspicion, which I present here as a form of rationalization. The rationale represents our justification to take action. This justification need not be based on facts. While the rationale behind our actions should be important, perhaps they often serve to substantiate excuses in case something goes terribly wrong. "There were weapons of mass destruction. Oh, sorry. We were under the impression that there were.  Nonetheless, the rationale was sound." I am suggesting that generally speaking particularly in terms of organizational decision-making, the rationalization is irrelevant. The channeling of data resources to entertain human distraction is a great threat to our social and technological advancement.

During my undergraduate years, I stopped eating fruit mostly due to lack of financial resources. In light of my limited funds, to me it made sense to concentrate on the other food groups. These days, I have adequate resources to make fruit purchases. But I encountered a "good reason" to stop my consumption. I don't have any rational explanation. Here then is a real-life example highlighting the debate. Does a person need a rational explanation to stop eating a major food group? Most aspects of the illustration below have been covered in earlier blogs. I started collecting personal health data about 20 months ago. I currently keep track of approximately 1,300 reoccurring health-related events, which in my case includes what I eat. In relation to these events, I examine their impacts to about 100 different health metrics. Through the use of these metrics, I obtain "systemic impacts." Placing the events in sequence from poorest to best impacts as shown below, I noticed one day that all of the events relating to fruit consumption were suspiciously below neutral - indicating adverse impacts. I marked the fruit events using vertical lines extending to the bottom of the illustration. Fascinated by this negative bias, I started monitoring event "fruit1a0," which is triggered when I eat no fruit. I was surprised to find fruit1a0 rise above neutral - indicating positive impacts. Based on this preliminary information, it seems that fruit and I are incompatible. By the way, the "No Fruit" line has been pushed further to right over the course of writing this blog.

I am definitely not discouraging readers from eating fruit. Please by all means eat fruit. I am writing about a personal experience. I do not represent myself as a healthcare professional; and it is not my intention to advise people on their healthcare needs. Despite the fact that drinking milk is widely regarded as healthy, nobody would question a lactose-intolerant individual over his or her deliberate avoidance. We also accept that peanut allergies and sensitivity to seafood affect some people enough to justify not just avoidance but considerable dietary care and attention. I have come across some individuals that avoid eating beans. Suffice it to say, the idea of keeping away from certain foods is not all that unusual these days. Let's say for the sake of argument that I am not going down this road at all. I'm not here to discuss whether or not fruit is good for people or the extent to which allergies are an important consideration. Fruit is good for people. Allergies are important. This isn't the discussion. At question is the need to reason out everything at an intellectual level in order to take action. "You seem to have severe breathing problems when eating peanuts. It's possible you've been cursed by a gnome."  What the heck . . . "Extraterrestrial spores have seeped into our underground water aquifers and affected the food system."  Get out of here, no way!  "Those multinational corporations are genetically modifying peanuts. No wonder you can't breathe properly." Of course, all of this debate is not without its special place and time. But does a person actually "need" a rationalization? No, absolutely not. In relation to capital resources, given that problems "like this" (on a more abstract level) might surface organizationally, is exhaustive and time-consuming research requiring hypothesis testing all that meaningful in a process or rapid adaptation? Rationalism and superfluity are close cousins. Organizations need sensible reasons to do things.

My main objective in the next illustration is to show that the fruit events are more concentrated to the left (having negative impacts) compared to the right (having positive impacts). Some readers might recall how I sometimes "plough" data if there are irregular temporal distributions. I have no problems ploughing ordered sequences. I should explain that I am predisposed to 1) thematically ordered sequences and 2) creating plough charts or running tabulations. It's not all that statistical, but these techniques are so simple and easy to understand and explain. I found that fruit events were more frequently associated with adverse metric readings the more negative those readings became. The "cumulative total" shown below represents the total of fruit-event instances: about 20 instances at 477 events; 30 instances at 885 events; reaching a plateau at about 36 instances at 1089 events. Based on these two illustrations, I found that my avoidance of fruit was "reasonable." I had absolutely no explanation for the adverse impacts. This is not to say that I am incapable of rationalization. Just to make my point, I will offer a few explanations that seem plausible but which might be incorrect.

I suspect that the fruit as a commercial product has been altered to improve resistance to insects; reduce processing and shipping losses; and minimize spoilage. I believe that ambient bacterial levels in the environment have changed over the years, and this might be affecting the digestibility of certain foods to certain people. I feel that genetic modification and selection has led to fruit species that appear normal but which have substantively changed. I suspect that some individuals are sensitive to trace levels of the pesticides used on the fruit. It is also possible that I have a genetic predisposition that becomes more pronounced as I get older. So these are rationalizations that are not necessarily correct.  Nor do I ever have to prove that they are actually correct in order to take action.

Confirmation of Benefit

After making a radical lifestyle and dietary change, I consider it important to monitor developments carefully. I have emphasized two major abstractions of data in my blogs: "event data" that might be qualitative in nature; and "proxy data" sometimes called performance metrics. Certain behaviours represent the events that give rise to performance metrics. (In other words, the events are symbols of behaviours or circumstances: [orange1] might mean to eat an orange. Event [orange1] might be associated with a metric for consumption: 50 percent terrible; 30 percent bad; 15 percent neutral; and 5 percent good.) To confirm outcomes, I am usually interested in proxy data, which is something that can be evaluated statistically. Others particularly in this forum can statistically evaluate proxy data much better than me since I am not a statistician. I apologize in advance for resorting to linear regression. Perplexing perhaps is applying linear regression to ploughed metrics essentially to get the slope; yet this is precisely what I have done. I have two charts to share. "Pre-Fruit Metrics" in the first chart shows the plough lines for metrics before fruit avoidance. BREATHE represents my ease of breathing in the evening; BOD at beginning of the day; CARDIO for the overnight period; and STEP for periods of physical activity. Evidently the slopes are quite steady as indicated below. Note the equations to the right. Before going cold turkey, I was already reducing on my fruit intake due to concerns over digestion; this might be apparent in the illustration.

The chart for "Post Fruit Metrics" shows the same ploughed metrics after fruit avoidance. For all of these metrics, higher numbers mean more positive impacts. Therefore slopes resulting from higher metrics indicate improvement. All of slopes are indeed higher in the fruitless period compared to the fruity period as confirmed on the illustration below. The biggest improvement occurred in relation to the STEP metric pertaining to periods of physical activity. Notice how the purple line has moved from third to second place. This change is consistent with my sense of thing. When I was younger, I had exceptional distance-walking capabilities; but this started to decline noticeably about 10 years ago. I recall that my walking performance seemed least affected early in the morning before eating. I never rationalized the situation in relation to any particular food or food group. I certainly wouldn't have blamed fruit since I rather enjoyed eating it.

I have been asking if others plough their numbers the way I do since it seems unlikely that I'm the only person. Ploughing often results in smooth slopes - but usually not lines that are completely straight. It has the effect of reducing the impact of fluctuating distributions. I use the term "ploughing" since it describes a build-up of material. Through ploughing, it becomes apparent that most of the slopes are comparable; however, the fact that all of the slopes together show improvement at the same time is noteworthy. Further, STEP has drastically improved. Consequently, it would seem non-productive or non-constructive for me to resume eating fruit. I reached this conclusion analytically using data irrespective of the normative "fruit is good for you." This assertion has no basis in fact. It serves only to place me at a competitive disadvantage. It is more accurate to say, "Fruit is often good for many people to eat." In any case, I am not trying to make any enemies here. I'm just saying that we have all sorts of computing capacity. We have to collect data. We have to consider that data. We should act on that data. What may have been acceptable decades ago needs rethinking today. It's not enough to rely on codified beliefs. Our intentions may have been sound long ago especially in relation to personal health. In terms of organizations today, there aren't any excuses at all to persist relying on petty production beliefs and theocracy. It's an age of data for good reason: business desperately needs more good reasoning.

Next Post Will Be in 2015

I paused at this part of the blog in part because the ending is so abrupt.  Also, I wanted to wish everybody a happy New Year.  I prepared a closing image, but it seems rather risque.  Well, I will go for it and hope for the best.  Featuring Chico, Mr. Plantain, and Bee-Dee, this is to wish everybody a fruitful or fruitless 2015.

Views: 338

Tags: administration, analytics, approach, behaviours, beliefs, construction, control, corporations, decisions, faith, More…influence, management, methodologies, philosophies, philosophy, procedures, processes, professionalism, professions, rationale, rationalism, rationalization, reasoning, religion, role, scientific, social, systems, theocracy, tribalism

Comment

You need to be a member of Data Science Central to add comments!

Join Data Science Central

Videos

  • Add Videos
  • View All

© 2019   Data Science Central ®   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service