« Worthwhile Fundamental Concepts »
When it comes to understanding how things work, there is a specific set of fundamental concepts that are well worth absorbing. These principles help us break complex problems into manageable pieces, analyze their components systematically, and then synthesize new solutions. From establishing our available resources to recognizing emergent properties, each concept builds on the others to create a robust problem-solving toolkit. And because life continues to evolve and problems rarely stay static, these principles also include tools for breaking free of entrenchment when our existing solutions become trapped in old patterns.
Without regard to the overall complexity of a problem, you will start with a process of analysis, namely breaking things down into their fundamental parts or truths. Synthesis will come later, where you will gain insight into the nature of the problem as you put things back together. The earlier principles in the following list will help you to do that well, such as Occam’s razor, which tells you which parts are really necessary at a given time.
Establish the following:
Resources are understood in terms of an economy, that is to say, it is not just a matter of how much you have to spend (i.e., quantity), but the way(s) in which you spend (i.e., quality), which leads us to…
This problem-solving principle is often misunderstood and applied falsely. It gets colloquially represented as, “The simplest explanation is the best one,” then when any sort of issue needs an explanation, someone just says, “Aliens” (or similar).
See how simple that explanation is? Watch this…
Observed event: The fog in the neighborhood seems to be unusually heavy this morning.
Misapplication of Occam’s Razor: It’s a government conspiracy.
Application of Occam’s Razor: 1) This is the season when the atmospheric and ground conditions tend to lead to heavier fog; and 2) we will stop right there in crafting a proposal for testing (for the time being, at least) because the other explanations have an additional number of pieces (that we do not need to get into while we test this one).
Occam’s razor doesn’t even insist that this particular parsimonious explanation will turn out to be the right one once you have tested it. It really says that when you are going to propose a hypothesis, you should start with the one that has the fewest number of necessary pieces, test it, and then work from there.
So Occam’s razor is a risk analysis designed to help people (mostly philosophers of various sorts) avoid wasting resources.
Of course, that risk analysis presumes that the reasoner wants their economy to be parsimonious, and that’s not always true. Maybe your priority in the moment is just to have fun bullshitting with some friends or something, and the costs are not a big deal. But figuring out how we want to approach our resources (to whatever degree of rigor) is still important because of the following dynamic…
Our decisions are affected by our resources (and our economy), so you want to think about setting a boundary on the rationality of your solution, identifying a preferred balance between satisficing (i.e., finding a good enough solution) and optimizing (i.e., finding the best solution, where the metric for “best” can vary among people).
This selection can affect what sort of agreement might be reached with others; that is to say, you can find a solution that is good enough for you, but sometimes you will also want other people to treat it as good enough for them as well.
Keep in mind that your optimization will be context dependent, not just in terms of planning how you will distribute your personal resources during different stages of the analysis, but also your solution’s resource distribution relative to the needs of one person versus another.
These are kinds of Trade-Off Analyses.
When it comes to distributing resources (and balancing outcomes), you often have to consider the satisfaction of many constraints, such as the welfare of multiple people. A Pareto improvement helps in one area without making any other areas worse. A system that is Pareto efficient or optimal cannot change without making things worse in some area.
Each of the parts that you have gathered is mapped onto a conceptual topography. You want to have some notion of where each fits in the grand scheme of things. This is sort of like looking at a jigsaw puzzle piece and noticing what its contour looks like. Is it typical (i.e., a rectangle modified with concave and convex bits), or is it an unusual piece with a flat edge (or two), or maybe it’s shaped like an animal?
This concept is explained in detail in the CDS tutorial, but the basic notion is that we tend to organize information as if it were sets of overlapping wheels (speaking figuratively for the purposes of illustration); for example, some people imagine a rather generic kind of “long-legged slacks” as a schematic, core example representing the concept of [PANTS], which is then surrounded by an array of more specific and/or less perfect examples, such as golf pants, pajama bottoms, parachute pants, jeans, and so on.
The generic slacks act as a hub of a wheel, namely the generic concept of [PANTS}, surrounded by spokes that radiate outward to the more specific entities on the rim. Each of those more specific entities then acts as the relatively generic hub of its own wheel, where the likes of [GOLF PANTS] would be surrounded in turn by such specific versions as checkered golf pants, linen golf pants, worn out golf pants, miniature golf pants (however parsed), and those pants that are displayed prominently in movies about golf.
A familiar example is that while tomatoes are classified as a fruit in a botanical taxonomy because they are seed-bearing, they are peripheral members of the set because they tend to be savory rather than sweet, hence the rarity of (a) tomato ice cream and (b) blueberry spaghetti sauce.
Radial categorization also applies to the set of principles listed here, arranging them into core and peripheral members of the concept [PROBLEM-SOLVING COMPONENTS}, reflecting (for example) how ordinal a part might be in the process of(i.e., first, second, and so on). It is part of a systems analysis in that it describes an interaction between elements (i.e., which is the priority, the hub). It applies to bounded rationality because it describes the geography of the boundaries. It is part of a state space analysis because it can be used to describe the arrangement of those spaces.
So don’t think overlapping circles, think intersecting spheres. And then stop thinking spheres and start thinking hyperspheres.
State Space Analysis
For each of the necessary pieces that you have gathered, consider each of its possible states. ‘Off’ and ‘on’. Emotional states. Flavors. This information will be needed when you shift to a consideration of the synthesized, holistic system, and the interactions between these states. Their actual and practical characteristics are also states…
While the coining of the phrase, “The customer is always right,” is debated, Henry Selfridge tends to be identified as the originator of the crucial clarification, “in matters of taste.”
The canonical purpose of a screwdriver is to drive screws, but if someone is scanning their environment for a tool with which to open a paint can (i.e., where that is the presenting problem to be solved), then do they perceive a screwdriver when they see one, or a lid pryer?
In the story Mermaids are Practical (from back in 2010-ish), I talk at some length about the the benefits of treating the world according to the way in which you perceive it (i.e., the pragmatics of how you interface with it) rather than slavishly obeying “the truth.” When it comes to radially categorizing the parts of the problem, you will often want to do so according to what the stakeholders actually do, instead of what the objective truth seems to be.
The principles up to this point have focused on breaking things down into their pieces and parts and their properties (i.e., analysis), both in terms of your personal resources and the components of the problem.
Now you are going to look at the interactions and influences between all of the pieces and their states, such as “when the flavor is banana, I am happy.” You will identify feedback loops, delays, and emergent behaviors.
Examine complex, higher-order entities for properties that emerge from the interaction of their lower-order components, in the sense of “the whole being more than the simple sum of its parts”; for example, when you come across arguments that try to categorize entities as being driven by either nature or nurture (i.e., a false binary), consider instead that some such entities emerge from the interaction of nature and nurture to establish a whole new categorical entity (which is likely radially organized, as above).
Language is just such an emergent property, as thoroughly discussed in the following collection of essays:
Tomasello, Michael, and Slobin, Dan I (Eds.) (2005) Beyond Nature-Nurture: Essays in Honor of Elizabeth Bates. [ISBN 9780805850277]
Similarly, interpersonal relationships emerge from the interaction(s) of the personalities of multiple people (where the personality of each such person is an emergent property in itself).
But have you found the most efficient answer?
Let’s say that you were afraid of heights, and that you felt most relaxed when you were closest to the ground, and surrounded by the highest barriers (to keep you from falling over an edge.
No imagine a set of hammocks tied one above another on two very tall trees. The hammocks get smaller as they go up.
You are in the highest and smallest hammock. You are not falling, You are lying in it, and gripping the sides. Relative to that hammock (i.e., your local context), you are exerting as little energy as possible (which we also measure in fear). That is your local minimum.
The other hammocks are other potential minima, albeit not local.
Now, suppose that there is an influx of energy, like a high wind, and it flips you out of the smallest hammock, but you are caught by the next one lower down. It is bigger, so you are more secure. You do not have to grip as tightly. Your fear is somewhat lower. Overall, you are expending less energy now. This is your new local minimum. As above, you will remain at rest there in a relatively low energy state (compared to moving around in the hammock and being more afraid)… until another gust of wind or something flips you out.
Eventually, and ideally, you come to rest on the ground. You have no fear of falling at all in that local context, and you don’t have to grip onto anything. The problem is solved in a way that costs you the least amount of energy, so you have those resources available now that you can devote to some other problem… like figuring out who the hell stuck you in this hammock experiment.
The point is that once you use all of the above principles to come to some sort of conclusion, it can be good to inject some energy into the system to see whether it settles into am even better local minimum. Here are tools that help…
Now that you have a fairly stable, internalized model of the system that you are synthesizing, consider the “what if” scenarios that occur when you change out parts or fiddle with the dynamics. Observing the resulting changes will help you to isolate causal relationships and test your assumptions.
Engage in this step periodically, as you might have an opportunity to settle into a new local minimum.
Reasoned Impulse is an app that I prototyped for an interface design course. It brings disparate concepts together to inject the sort of energy that is described here.
it is also the topic of the next essay.