Monday, January 26, 2026

A Grand Unifying Theory: Creativity, Productivity and Specialization

This post is one of six posts in a series on this topic. The full list of posts are linked here for convenience.

Any philosophical analysis of the arc of history, economics and politics and their purposes has to be grounded in an assumption about whether it is possible to name a single goal which can accurately reflect a goal of the entire population. The easy answer is NO, doing so is impossible because people can have multiple goals and prioritize them differently, based on which goals have already been met to some minimal threshold. For analysis here, it will be assumed the universal goal driving most human behavior is providing the proper necessities for basic life and health. Adding one additional assumption to that initial goal -- that "sustaining" human life results in a growing human population -- sets the conditions in place for everything to be analyzed over the paragraphs and posts to come.


Defining Terms: Creativity, Productivity and Specialization

This first phase of analysis focuses on three key factors that may often be considered synonyms of each other but merit a few key distinctions. For purposes here, creativity involves the ability of humans to formulate unique ideas from the world around them and use those ideas to alter the impact of the environment upon them or their impact on the environment. Productivity is a measure of the rate at which humans can complete tasks required for baseline survival or for satisfying more intangible goals higher up on the Maslow pyramid such as esteem and self-actualization. Specialization is the practice of purposely reducing the VARIETY of tasks performed by a single person or group in order to allow that person or group to become far more productive at a smaller set of tasks.

It is crucial to caveat that the relationships BETWEEN creativity, productivity and specialization are not mathematically precise nor will all people view them the same way. Caveman Og may have found a new source of pigment that allows him to double the number of cave paintings completed per week but cavewoman Thog may likely conclude that has done nothing for Og's productivity in slaying wooly mammoth for tonight's dinner. If the family is starving, cave art is worth considerably less than food.


Creativity Leads to Productivity

Since humans do not emerge from the womb exhibiting adult levels of understanding and physical abilities, even sustaining any existing human population with a 1:1 replacement ratio between parent and child unavoidably creates an addition burden on existing adults as children are raised. The longer dependency period for human children REQUIRES increasing levels of productivity at least for SOME period of time (12-18 years?) from adults. If a couple required 2.0 units of food daily for sustenance, once a child arrives, those same parents require SOME incremental volume of food per day for a child. If 2.0 units of food required 2.0 units of work in their local environment, that couple is going to have to work some additional amount, say 0.X units, to exchange for 0.X units of additional food for a child. Have a second kid? Add another 0.X to the load. And don't forget a bed, additional clothes, daycare while Og and Grog work at the rock wheel quarry, etc.

Where are the extra units going to come from? Certainly, the parents can attempt to work more hours every day but that likely becomes counter-productive after about sixteen hours of work per day and some work might require daylight which is further limited. If Value = Hours x Productivity and Hours are already maxed out, the only other way to get more Value is to increase Productivity. There are only so many ways to improve productivity:

  • continued mastery of a physical task as originally devised ("practice")
  • identification of easier / more productive sources of materials to utilize ("better inputs")
  • devising a better process for accomplishing a result that yields the desired outcome ("a new and improved buggy whip")
  • devising a different solution for the underlying need that requires less work / resources ("a car instead of a horse-drawn carriage")

All of these approaches for increasing productivity can occur even with a single human operating alone. Once the system involves multiple people, increased productivity inevitably leads to specialization.


Productivity Leads to Specialization

In a system with multiple people, workers quickly realize that if tasks A and B are required every day and everyone currently is working at tasks A and B individually for their own needs, everyone would increase their productivity by choosing one of the two, getting even MORE productive at that task, then swapping with others for units from the other task. Everyone gets more total units of A and B generated in a fixed amount of time, giving everyone a share of a bigger pie. Or everyone gets the same amount as before with less total time spent working. An improvement either way.

At a higher level of abstraction, specialization is just a narrower form of trade which will be addressed in the next installment in this series. However, one distinction between specialization and trade is worth making that involves the level of trust required and the degree of formality among the parties. If two parties are each creating 1 unit of A and 1 unit of B independently then agree to specialize so that one creates 2.2 units of A, the other creates 2.2 units of B and they both wind up with 1.1 units of A and B, it is easy for both to conclude the specialization is worthwhile.

However, if the B worker only produces 2.1 units, will the A worker still produce 2.2 units of A and accept 1.05 units of B? If the two workers are related or from the same community, these variations can still be worked out without complex rules and specialization will still provide value. As the universe of workers gets larger, more formality is likely to be expected to ensure some workers are not shirking and free-riding on the higher productivity of others without increasing their own as well.

This is a key point early on... Notice how quickly concerns about actual work come into play among people pondering specialization and productivity? Whether people understand the linkage or not, EVERYTHING with people's perceptions of the value of THINGS comes down to their grasp of the amount of WORK involved to create those things. If someone is going to trade you something from thin air for something you worked eight hours for, you can get that other thing out of thin air as well and keep your object and be better off. A trade only makes sense if approximately equal value derived from work is perceived in both objects involved with the trade.


Funding Creativity / Productivity

So far, the analysis of creativity and productivity have essentially assumed that all creativity just emerges from a random "spark" of an idea in a human that is pursued, experimented with and eventually results in increased productivity that creates extra value. But the spark phase of that process is not totally random. There are things that can be done to increase the likelihood of a new "spark" occurring to a worker and resulting in an improvement. The first involves purposely allocating time to workers to rethink existing tools and processes. To the extent that enough time is set aside for this purpose, this work can be assigned to specific workers as another form of specialization. When a group formalizes work to this degree, this approach acts as an "investment" in the collective system.

Modern business obviously implement this exact strategy and attempt to optimize the amount of resources (money, people, time) to this "research" to maximize productivity gains without spending all incoming cash on research at the expense of profits. In reality, the link between a specific level of resource commitment and growth in productivity is virtually impossible to model with mathematical precision. These factors are all

  • hard to quantify,
  • they have exponential, compounding effects over time,
  • and those effects are spread across many other disciplines

The fact that no one can present an accurate numeric model with a straight face does not mean it is not worth modeling the process to think through what-if scenarios to appreciate out results change from inputs, funding and priorities that vary -- randomly or intentionally. As a first attempt, imagine the entire world of knowledge being neatly divided into eight sectors. All of the sectors are related and start off adjacent to each other with a specific amount of existing knowledge represented by the area of a circle.

Over time, each area of knowledge grows in total area and expands its "frontier" ever outwards. Collectively, all of the sectors do the same thing and the overall scope of human knowledge is represented by the total area and the outer "frontier" of all of those expanding circles.

In reality, all sectors do not expand at the same rate over time. Progress may seem to wax and wane over years / decades, just due to the random interactions between all of the disciplines. But what if interests associated with one sector attempt to boost spending on their sector at the expense of other sectors? In the visual below, the growth rate was changed from 4% yearly for all sectors to 12% for Computer Science while Geology and Education were lowered to 0.5%.

After 20 years, the compounding effect on Computer Science is readily apparent but even this modeling approach is flawed. What if over that twenty year period, progress in computer science actually required discovery of new formations of rare earth metals required to manufacture high density semiconductors? In this example, the extra funding for computer science "research" came at the expense of Geology and Education so it is possible less resources in those areas will "starve" computer science of a needed innovation that will impair future advancements in computer science.

Special interests routinely lobby for explicit boosts on spending in their sector or relaxed regulation that artificially stimulates spending in their sector not just because they think it will someone improve society but because they are uniquely suited to personally benefit from that boosted spending. A corollary to this is that special interests will lobby for boosting spending in their sector AT THE EXPENSE of other sectors. Giving them the benefit of the doubt, they may do so because they don't understand the cascading long term effects of "starving" other sectors. But that doesn't mean these relationships can be ignored. The larger public needs the vocabulary and analytical tools to identify these situations to more closely examine stories being told to decide what is best. This is an underlying theme of this entire analysis.


Impairments to Productivity

So why are creativity and productivity not linearly related to "spend" in any particular sector? In any modern society, there is an upper bound to the rate at which new knowledge can be shared and adopted. Until the internet reached critical mass in the 1990s, it was virtually impossible for a single person to instantly communicate an idea to millions of others without coordination and consent from a multitude of parties (publishers, broadcast media, etc.) Even with the Internet, not every inventor or expert is adept at packaging information about a discovery in a format that is suitable for educating thousands / millions of others about their new invention. It takes time and effort to develop materials to train other people who will train even more to eventually communicate the discovery to the masses. This acts as friction which slows down adoption.

Events over the past four years provide a different, more concerning example of why progress cannot be instantly communicated and adopted. The primary AI technologies in use today based upon large language models were created by ingesting petabytes of existing online content, analyzing the text of that content and calculation trillions upon trillions of statistics about the next most likely "token" to follow a prior "context" of tokens previously seen. This flow is illustrated below.

As illustrated, the universe of online content was mostly human generated (green) until about 2014 when the use of "bots" to fake likes and comments on social media became obvious. By 2022, the launch of competing AI platforms led to an explosion in AI generated content (in red) which has already swamped many platforms. This red content is being fed back into AI training flows and becomes indistinguishable when converted to trillions of statistical probabilities of character sequences. This means AI systems are contaminating their own well with increasing amounts of slop which permanently impairs the ability to separate human-curated content from slop forever which slows down progress rather than speed it up.

This is not hypothetical. Generative AI technology has taken an already flawed publish-or-perish model in academia and swamped it with even higher volumes of suspect content. This cascade of machine generated plagiarism and noise has tarnished the reputation of many online publications found to have released "research" with plagiarized content or synthesized data. It has also overwhelmed the ability of qualified peers to conduct reviews of such papers without taking away more of their time from their research. As a result, many of those publications are REDUCING the number of papers being published. They cannot provide a qualified human review to five times the volume being created with one fifth of the prior effort and care being given by those submitting the papers. The net effect? Tools that should assist in IMPROVING the quality of research are instead being used in ways which are significantly REDUCING the quality of new research in nearly every field.


WTH
Posted at