This post looks at the science of historians—a discipline I suspect many in business will not have encountered. When society was “hypnotized” by the success of scientific laws, an unintended side effect encouraged scientists and society to believe that history is not scientific. Fortunately for those of us doing business, that idea is wrong.

Note, though, that if we are to change our ways of knowledge we must do so with curiosity—exploring options which offer insights without demanding law-like discoveries (expectations of older paradigms). History gives us an excellent start for this curiosity as it works always within complexity while its science is related to non-experimental scientific disciplines.

Before turning to history, let me clarify the knowledge most critically missing today.

GOAL #1: UNDERSTANDING WHOLE RESULTS

Western education helps us be well-prepared to observe the parts but makes us ill-prepared to understand the whole.
Edgar Morin, Sociologist

The challenge begins with how the superb tools we have for analyzing parts in isolation leave us ill-prepared to consider whole complex results. To do business, though, we must study whole results if we are to gain insights into how a business works.

In other words, we need ways to discover whole results in the past, which exist today, and toward which we might guide our businesses. Unfortunately, the challenge of understanding whole results is very different from traditional scientific discovery. Where traditional investigation expects to develop clear, precise views, it is not possible to comprehend whole results all at once.

For example, every human being is, on their own, a complex whole being—a complex whole result. Yet even within deep relationships there are no simple, monolithic views of the “whole” of another person. I can, then, gather together a hundred pictures of my wife of 32 years but that is not enough to portray the whole person who is Judith. Of course, photos cannot capture parts of her person which are not visually evident. Yet even with our superb closeness after 32 years, I continue to be surprised by new things about Judith as she is continually surprised by new discoveries about me. No process can fully encapsulate our whole human results. Fortunately, this reality helps keep our marriage fresh.

If we are to comprehend such whole results, then, we must seek out many views as there are no monolithic, single views of a whole person much less a company. Within this post and those which follow, then, readers will be continually reminded that we must take up a variety of “whole views” if we are to perceive the whole result.

Note that a robust comprehension of whole results may not be critical if we are taking only short-term actions within a situation where complex effects are small. On the other hand, continually maintaining an understanding of the whole complexity of our work is critical, for example, if we want to build a successful business.

GOAL #2: COMPREHENDING INTERACTION AND ADAPTATION AMONG PARTS

…human jobs are not simply collections of independent fixed tasks; most jobs require the jobholder to understand how different tasks relate to one another, to adapt to change on the fly, and, more generally, to be flexible based on the open-ended nature of the real world… “The easier a task is to measure via benchmarks, the less likely it is to represent the kind of complex, contextual work that defines professional practice.”
Melanie Mitchell. Also quoting Sayash Kapoor and Arvind Narayanan,

A second critical reality is that it is far easier to study parts than to study actions, interactions, or adaptation among parts. We cannot know, then, what passed among a group of employees except for those parts explicit in memo’s or emails. Yet even what is explicit in documents omits what is assumed within a group yet such assumptions are critical for effective work and business success.

Consider Melanie Mitchell’s observation about the truth of human work—that it’s not simply a set of tasks. Unfortunately, traditional study cannot comprehend such realities of interaction and adaptation. Thus, we not only struggle to see the whole, we struggle to comprehend how it came about. (Note that many job related claims of artificial intelligence assume a job is simply a set of tasks.)

Once, again, when complex effects are small, traditional social sciences work studying parts can be effective as long as we reject claims of (a) universal paths, (b) remember that what we discover will mislead us if we are wrong about complexity’s importance, and (c) will always eventually become misleading as the world changes around us.

LEARNING ABOUT RIGOROUS HISTORICAL STUDY

We cannot think effectively about our complex world within the strictures of managerialism — instead we must “deeply reform all our way of knowing and thinking.”
Edgar Morin, Sociologist

Understanding the general failure of traditional research to comprehend what is complex, I began to think of historical study—not because I expected absolute answers but because historians must come to understand complex whole results. We begin with lessons from rigorous historical study, then, because today’s whole emergent result has arrived as a result of past interactions.

Wanting to find useful starting points I first reached out to Marjorie McIntosh, emeritus history professor at the University of Colorado, Cambridge University Press author, and my undergraduate history prof. Marjorie pointed me toward Marc Bloch’s superb The Historian’s Craft—a work written during WWII but published posthumously in 1949. (As a member of the French resistance, Bloch had been executed in 1944.) I also turned to Isiah Berlin’s superb essay “History and Theory: The Concept of Scientific History” (History and Theory, 1960, Vol. 1, No. 1 (1960), pp. 1-31). Berlin looks in a broad sense at learning as a whole practice and at the differences between the scientific expectations of the day and his practices as a historian. He primarily rejects the idea that history is “a science” because, at the time, study had to lead to predictions of the future if it was to be a science. I have written about Berlin’s essay here.

John Lewis Gaddis of Yale had studied with Berlin and updates discussion of historical methods in his book The Landscape of History: How Historians Map the Past (2002). Gaddis relies on Bloch and others as well as his work on complexity with the Santa Fe Institute. By the time Gaddis wrote, natural scientists had come to understand a great deal of useful science cannot predict (e.g. evolutionary science). Starting from this change, he discovered historical science is quite similar to the practices of scientists in disciplines which are not inherently experimental such as anthropology, geology, and others. Gaddis concludes that the study of history is more scientific than much of the study in today’s social sciences.

What follows starts from Gaddis’ discussion, informed by Bloch and Berlin. To be clear, I do NOT claim all business study must take up these practices NOR claim that Gaddis offers the only, or most valid, view of historical study. Rather, his work offers us a useful starting point for considering complexity in business.

STARTING POINT: WEAKNESSES IN WHAT WE KNOW ABOUT OUR BUSINESS

…it is from the structures that survive into the present—the “certain residues” of which Bloch wrote—that we reconstruct processes inaccessible to us because they took place in the past.
John Lewis Gaddis

Historians start with the idea embedded in the quote above—that evidence of the past is found through structures or residues left behind and which survive into the present. Our understanding of business present or past is no different—we are working with those certain residues which survive into the present and into our possession. Some of these residues in business will be quantitative while many are qualitative.

Critically, all residues (or accumulations of residues) are necessarily incomplete as all residues have weaknesses. Consider some common residues along with their gaps.

• There is always material the company made public including press releases, white papers, and carefully prepared financial reports. All these residues come with cautions as historians know any official narrative—whether from historical governments or modern companies—has been crafted to achieve a purpose. Such crafting leads some ideas to be included and most to be excluded. While this material is useful, it does not carry all that we need to know to study business. After all, what it contains is, at a minimum, heavily influenced by the mythology the company prefers to put forth about itself.
• Companies also issue formal internal company documents and reports. These are important to study. As we study them, we must remember that companies keep only what they think is important — not what matters most to us in looking back. This material also cannot carry all we need to know and will be influenced by a company’s mythology or perception of itself.
• There are also reports which circulated informally inside a company. These sometimes help us discover what other evidence would hide but they are also skewed by bias as serious as the company’s own bias. There is bias in the preparation of the report AND the choice to keep it.
• We may also discover information through stories recorded or told by individual employees. As with other residuals, these are exceptionally helpful and necessarily incomplete.
• Some information will have been assembled or gathered by those close to a company but not employed by them. Similar cautions apply to this information. While important, the information was gathered to satisfy the observer which means it was not kept to help us study the issues with which we are concerned.

It is particularly important for those of us in business to recognize that a choice to keep certain information or data leads to bias in that information. The business, after all, made its own choice that other information or data is unimportant. Such choices are not bias-free. Thus, our considerations are inherently skewed by the information available to us or unavailable to us.

Having considered Information within and quite close to a company, readers should consider the gaps in structures or residues available in society at large, within the media, and within popular perceptions or writing about companies.

Our first learning from historical rigor, then, is that information from surviving structures or residues always has gaps—we work always with an incomplete record of business past and present. As William Starbuck notes, “nearly all managers hold very inaccurate perceptions of both their firms and their market environments.”

STARTING POINT—AFFECTED BY COMPLEXITY: THE PROBLEM OF VERBS vs NOUNS

Standard economic theory uses mathematics as its main means of understanding, and this brings clarity of reasoning and logical power. But there is a drawback: algebraic mathematics restricts economic modeling to what can be expressed only in quantitative nouns, and this forces theory to leave out matters to do with process, formation, adjustment, and creation…
Brian Arthur, Economist

Gaps in residues grow because business practice is primarily a discipline of verbs—actions, interactions, and adaptation. The primary residues of business, though, are nouns—things, people, programs, bureaucracies, numbers, reports, and more. After all, the details of noun parts are easily recorded while little can be effectively recorded—even seen—of what passes along interactions.

The blind spot inherent in the structures and residues, then, becomes more severe because the residues do not pass along understanding of interactions and adaptations. Thus:

1. ALL information looking at a company’s past or current state is flawed.
2. That information is flawed is simply reality. While we should always work to improve its qualities, the information we work with will always be flawed.
3. We must, then, learn ways to work with data which is flawed in the ways of residues (fortunately historians can help us do this).
4. Many might protest, suggesting data must be perfected. They ignore how ALL data is inherently flawed whether stored in massive data lakes or summarized within an AI.
5. Success in business does NOT arrive by demanding perfect data, but from learning ways to succeed despite the flaws of the information we have available.

Onto the challenge of noun residues, then, we add loss of information about the verbs of action, interaction, and adaptation. Once again, to succeed in business we must work within this flawed data.

STRUCTURES AND RESIDUES: FILLING GAPS

Social sciences during the twentieth century embraced a Newtonian vision of linear and therefore predictable phenomena even as the natural sciences were abandoning it.
John Lewis Gaddis

Sociologist John Goldthorpe notes that “a historical fact is an inference from a relic.” Gaddis notes such relics include “bones and excrement, tools, weapons, great ideas and works of art, or documents that get deposited in archives.” Critically, “in each case processes produced them. We can know these only from the structures they leave behind.” (Gaddis, 35-36)

Business is effectively no different. Our relics might include spreadsheets, printed reports, products, demonstration samples, brochures, advertisements, bureaucratic rules, written processes, and much more. Fortunately, those who study business deeply and with humility can gain key insight into the processes which can be inferred from the relics which are left behind. Business is not, after all, a practice of entire chaos as it is a human practice and we understand a great deal about processes among humans. That said, all analysis of business past, present, or possible future depends on inferring a great deal from the relics or residues available at a given point in time.

To do this, we must not trust what companies proclaim about their formal processes of business as descriptions of those formal processes and their use are also relics. No matter which set of dictated processes (e.g. Six Sigma) a company claims to follow, what happens is always different from such claims. For example, Jack Welch proclaimed that his success at General Electric was due to Six Sigma. Except, later investigation shows his General Electric appearance of success depended heavily on GE Capital whose employees resisted Six Sigma training. (Thomas Gryta and Ted Mann, Lights Out: Pride, Delusion, and the Fall of General Electric (Mariner Books, 2020), Kindle, Chapter 15)

The processes of humans doing business must be our focus—flows of cash and investment, interactions among humans, customers, sales, profits, ideas as they interact and adapt, machinery, software, competition, markets, and much more. Those who are thoughtfully experienced in business will understand these interactions. Further, these processes are not unlimited so we can reliably infer a great deal about what might have taken place if we follow clues within the residues and remain open to challenges.

Note, though, this does NOT mean one single right answer can be inferred from the residues of business. We must remain open to a variety of inferences based on understanding processes and sift them to find those best matching the structures we inherited.

Unfortunately, this process of inferring is one where many journalist authored histories are weak. While journalists are superb researchers and assimilators, they are rarely experienced in the processes of doing business. Too often I read work which takes up company claims verbatim without noticing that they violate process reality.

CAUSALITY IN A COMPLEX WORLD: FORCE AND CONTEXT MODELS

…our aptitude for connecting is underdeveloped and our aptitude for separating is overdeveloped.
Edgar Morin

The most intriguing and useful part of Gaddis’ book looks at how historian’s attempt to understand why things happened as they happened—to understand causes. In this thinking he was heavily influenced by Bloch’s writing.

Of course, as within any business, there are no single causes of any events in history. While historians know this those doing business do not. Business headlines shout out that “Company X Failed Because It Didn’t _____” (enter one thing) or that “Company Y Succeeded Because it ______” (enter one thing). Such headlines are wrong—always wrong. If we are to learn to thrive doing business within a complex world, we must understand issues of cause in more sophisticated and subtle ways.

In my upcoming book I suggest an approach for thinking about causes based on what Gaddis wrote which I call Force and Context Models (it seemed a name was needed). The longer I work with them the more value I discover for thinking through situations. Readers can learn more in my book.

They start by thinking of all the issues of context and forces (within and outside a company) which might have affected the results which emerged. List in hand, we begin to sort these contexts and forces. Generally, but not always, effects are larger for forces or context closer in time and place to what occurred. More critically, we must separate forces which initiated what happened—what Gaddis calls exceptional forces—and set all other forces alongside the context within which it happened.

Doing this helps us envision how interconnections might have influenced what happened. While exceptional forces “begin” what took place, the magnitude and direction of what results depends more on context (forces and other context) than the initiating forces. Thus, two companies which take identical actions will always have different results because the actions take place within unique context.

One of the most egregious errors in business study is declaring one exceptional action to be of universal value. When we comprehend what Gaddis observes, we begin to develop far more useful knowledge of our work. Force and Context Models not only reveal how what happens depends as much on context as on the exceptional force, they also make clear that we cannot make a list of forces and context then define percentage numbers telling us how much each element contributed to what happened. When a result depends on connected factors, such percentages are lies as noted in an earlier post in this series:

Executives, for example, demand that those who do advertising decide whether the most important factor in advertising impact is outstanding creative work or outstanding media strategy and execution. Except, the only valid answer is YES. Both are critical and they cannot be separated. Each is interacting and interlinked so that any error in one affects the impact of the other.

A simple historical example ponders the causes of World War I. Of course, an exceptional force was the assassination of Archduke Ferdinand. This only initiated a world war, though, because of the context in which the assassination took place. It took a vast array of contributing context to turn a relatively insignificant event into years of brutality.

Once, again, it is useful to note that no single Force and Context Model will explain everything. We must remain prepared to consider alternatives where each reveals a different side of how results developed.

NARRATIVES: RIGOROUS DESCRIPTION BEYOND NOUNS

It is not possible to explain time-based actions in the world without creating some type of narrative. This word, though, is quite concerning today because it is used for two very different types of narratives:

• Stories invented for companies (or politicians) to achieve goals are the first type. We should all know the dangers of such narratives as a compelling story is the company’s goal making loosely (at best) connected to facts.
• Historian’s narratives, though, need to be rigorously connected to residuals and inferred facts. They are created because narratives are the only way to express the changeable realities of interaction and adaptation. Historians, then, posit narratives which portray a likely sequence of events based on historical study. These narratives are then critiqued by others in the field before they can be accepted.

When Gaddis discusses narratives he considers only this second type as it is a major failing for a historian to choose the first and reviews of their work will call out its errors. Once, again, narrative is required as we attempt to comprehend what is complex. And, just as in history, no single narrative will encompass whole complex realities. We must, instead, seek out a variety of reliable, history-based narratives, each mapping unique views of what happened. Then, by comparing and considering these maps we can develop a sense of the whole of a company and its results. Only narratives can make sense of parts, interactions, adaptation, and whole results.

Added insight into narratives can be found in an article I wrote for JP Castlin’s Strategy in Praxis newsletter discussing the need for biographical magnification and its inherent danger. Useful narrative must magnify some parts, connections, interactions, adaptations, or environmental effects while minimizing others. Yet, that magnification and minimization also is not fully right. For example, one narrative might better capture and investor point of view while another magnifies market events including accidental benefits to the company.

We need a variety of narratives if we are to understand complexity even in our own business.

IN CONCLUSION

This post introduces ideas for exploring complexity in our businesses based on historical discipline—but they are merely a start. Having only skimmed the surface of Gaddis’ work, I recommend reading it more deeply while also exploring two other works:

• John Lewis Gaddis, The Landscape of History: How Historians Map the Past (Oxford University Press, 2004).
• Mac Bloch, The Historian’s Craft (Vintage, 1964).
• Clifford Geertz, The Interpretation of Cultures, (Basic Books, 1973).

The last of these three books is based on anthropology—a discipline with similar challenges attempting to make sense of what is inherently complex.

Next up in our exploration, JP Castlin will look at using safe-to-fail experiments to explore the verb-based, ever changing world within which we do business.

Until then, be well.

©Doug Garnett 2026—All Rights Reserved

Footnote: Brian Arthur, “Economics in Nouns and Verbs,” Journal of Economic Behavior and Organization 205 (2023) 638–647

My upcoming book, The Complexity Paradigm: Using System Science to Drive Business Success, will be published by Columbia Business School Publishing in January 2027. It can be pre-ordered today through Columbia or on Amazon.

You can read more about my unusual background (math, aerospace, supercomputers, consumer goods & national TV ads) at www.Protonik.net. Through Protonik LLC I consult with companies as they design and bring to market new and innovative products. Since 2001 I have taught marketing, consumer behavior, and advertising at Portland State University and I am an active member of the RetailWire.com Braintrust with regular discussions of unusual retail challenges. Together with my podcast partner Shahin Khan, current issues in marketing and business are discussed on The Marketing Podcast — available on Google, Spotify, the OrionX website, and Apple Podcast.