Corporate entities

This post tries to pull together some of the threads of a conversation with Gabor Molnar and Ramiro Montealegre about orgregores, my term for putatively sentient agents that emerge when large numbers of people are very well connected and have a common purpose. Our discussion focused on corporations.

The term orgregore is inspired by egregore, “an occult concept representing a distinct non-physical entity that arises from a collective group of people.” My speculation is that just as large numbers of well-connected neurons form brains that give bodies agency (and in some cases intelligence, sentience, and even consciousness), so there may be agency (or intelligence, sentience, or even consciousness) that emerges “on top of” organizations. 

Corporate culture

Orgregores raise the question of whether corporate culture is distinct from the people in an organization or is merely the shared beliefs of the individuals. [Endnote 0] Gabor pushed me to clarify this difference, asking whether the internet is distinct from the computers connected to the Internet, or is the shared nature of the computers' connection. He sees corporate culture as a vast network that connects the beliefs, behaviors, and attitudes of the people who are "connected" to the company (from the inside, like employees and managers, but also from the outside, like vendors and customers).

According to Investopedia, corporate culture refers to the beliefs and behaviors that determine how a company's employees and management interact and handle outside business transactions. This definition is formulated completely in terms of individuals, not aggregates. However, there can be different phenomena and behaviors at different levels of aggregation where the behavior of the group cannot be predicted (or predicted easily) from the behavior of the individuals. 

To use a physics analogy, one uses different parameters at different levels of resolution, like the velocities of individual molecules vs. the entropy of a gas volume. New concepts arise at higher levels. Individual molecules don’t have an entropy. Bulk properties arise: water is wet, but not individual molecules. While temperature is an average of molecular behavior, entropy is unrelated to attributes of individual molecules (to first approximation, leaving aside stuff like spin and conformation).

So following the physics analogy, is there something about corporate culture that is like entropy, i.e. it’s not a property of any of the employees? Gabor point me to an HBR paper claiming that culture “is a group phenomenon [that] cannot exist solely within a single person, nor is it simply the average of individual characteristics. It resides in shared behaviors, values, and assumptions …” Similarly, a PressBooks article on corporate culture  he gave me states, “Organizational culture refers to a system of shared assumptions, values, and beliefs that show people what is appropriate and inappropriate behavior.” [Endnote 1] Both definitions are framed in terms of attributes of the constituent entities, not the aggregate. However, this doesn’t get me to where I want to go because simply invoking shared behaviors rests the concept on the individuals that hold those beliefs. I didn’t see any evidence in this paper that explains ways in which culture is not just some metaphorical average.

I suspect there are beliefs and behaviors of organizations that none (or few) of their members subscribe to. This would be a property of the aggregate but not visible in any individual. One example is the military’s habit of killing civilians, e.g. the Kabul airstrike (WSJ) that killed an aid worker loading water bottles into a vehicle. Nobody was held accountable, and I assume most service people would regret if not decry the killing. Yet, this is something the military keeps doing; it’s in the structure and incentives of the institution. (Institutions can change, though. There’s a new U.S. Defense Department directive on mitigating harm to civilians.)

Emergence

Since I’m trying to distinguish between beliefs and behaviors of individuals versus aggregates, I think that at a hand-waving level I’m trying to identify emergent behavior. Further, I’m trying to establish to what extent claims about group action are not just metaphorical but have some reality. For example, if we say a company’s culture has taken on a life of its own, to what extent is that meta-organism literally alive (i.e., to what extent does it satisfy attributes of life ascribed to biological organisms)?

Geoffrey West’s Scale: The Universal Laws of Life, Growth, and Death in Organisms, Cities, and Companies (2018) argues that there are commonalities among aggregates of very different kinds regardless of their constituents. He suggests that all complex systems—including, presumably, institutions—have behavior that cannot be predicted from the behavior of its constituents:

A typical complex system is composed of individual constituents or agents that once aggregated take on collective characteristics that are usually not manifested in, nor could easily be predicted from, the properties of the individual components themselves. [...] The economic output, the buzz, the creativity and culture of a city or a company all result from the nonlinear nature of the multiple feedback mechanisms embodied in the interactions between its inhabitants, their infrastructure, and the environment.

…

In general, then, a universal characteristic of a complex system is that the whole is greater than, and often significantly different from, the simple linear sum of its parts. In many instances, the whole seems to take on a life of its own, almost dissociated from the specific characteristics of its individual building blocks. Furthermore, even if we understood how the individual constituents, whether cells, ants, or people, interact with one another, predicting the systemic behavior of the resulting whole is not usually possible. This collective outcome, in which a system manifests significantly different characteristics from those resulting from simply adding up all of the contributions of its individual constituent parts, is called an emergent behavior. It is a readily recognizable characteristic of economies, financial markets, urban communities, companies, and organisms.

Group intelligence

Gabor pointed to swarm intelligence (marketing video), the "collective behavior of a decentralized or self-organized system," and murmurations of starlings (BBC video). Both examples make a good case that intelligence (i.e., the ability to solve problems) can result from group behavior. However, I’m trying to go beyond aliveness and intelligence. Bacteria and plants are certainly alive; they’re also intelligent, defined as the ability acquire knowledge and solve problems. Arguably viruses (or viral genomes) are alive, too; they learn how to evade changing immune defenses. It seems trivial to assert that organizations are alive and intelligent in these terms. Intelligence doesn’t require a mind. For example, Antonio Damasio talks about ‘non-explicit intelligences’ in Feeling & Knowing: Making Minds Conscious (2021). Minds, in his approach, are neural structures that instantiate maps that represent other things, like sounds and body state. Another way to frame my question is whether some organizations are (or have) minds. If the answer is yes, one can then ask whether they’re conscious.

Another angle on shared intelligence is Ed Hutchins’ distributed cognition, where the entities are not all the same. The paradigmatic examples are landing planes and navigating ships, where crew members with different skills and various tools combine to complete a task. This case also takes me to puzzles over what is known; task completion is “knowing how” not “knowing that.”

Creating culture

Ramiro pointed out that culture is rooted in the way an organization sets its goals and how it interfaces with employees, customers, investors and the greater community.  He noted that founders and executives have a primary role in setting the culture. He also pointed to different types of cultures: “A clan culture, or collaborative culture, is mainly focused on teamwork. In a market culture, the bottom line is the main priority.  Adhocracy cultures are primarily focused on innovation and risk-taking.  A hierarchy culture is one that follows the traditional structure and has a clear chain of command.” He recommended the work of Geert Hofstede, e.g., Culture and organizations: Software of the Mind.  Gabor noted that culture depends on geography and its local values. The organizational cultures of companies headquartered in Frances, England, Spain, China, India or the US can be very different. What makes it even more interesting that companies are now global, raising the question of how to mix the local culture with that of the company as a whole. 

“Culture” seems closely related to the notion of social imaginary. Sheila Jasanoff”s “sociotechnical imaginary” probably applies at a lower scale to companies (tech companies, especially). From Jasanoff & Kim (2015):

We define national sociotechnical imaginaries as “collectively imagined forms of social life and social order reflected in the design and fulfillment of nation-specific scientific and/or technological projects.” Imaginaries, in this sense, at once describe attainable futures and prescribe futures that states believe ought to be attained.

Imaginaries defined this way emphasize a future orientation distinct from culture in the sense of “this is how we do things around here,” or “this is the kind of people we are.” Sociotechnical imaginaries are more along the lines of “this is what we want our future to be,” which reminds me of company mission statements, particularly the ones that use future-oriented rhetoric such as the ones whose market value implies the need to generate massive profits in the future.

Social control

One of the ways in which corporate culture relates to egregores is in the claim that “organizational culture is an effective control mechanism dictating employee behavior” (from the chapter “Understanding Organizational Culture” in Principles of Management). 

Mark Stavish describes egregores as social control mechanisms and his book on the subject refers to “the more technical view that egregores are only those psychic collectives that are actually created and fed via ritual and with a specific purpose …” This isn’t that far away from the “culture maintenance” discussed in the PressBooks article, which includes a discussion of rituals.

This is still framed in terms of employees, though. The bigger question remains: can one talk about organizations without talking about their constituent humans? Mary Douglas, following Emile Durkheim, suggests that institutions shape what individuals think (see How Institutions Think, 1986). For example, in chapter 8, “Institutions Do the Classifying,” she writes

Institutions systematically direct individual memory and channel our perceptions into forms compatible with the relations they authorize. They fix processes that are essentially dynamic, they hide their influence, and they rouse our emotions to a standardized pitch on standardized issues.

Corporate intention and agency

The legal notion of corporate personhood expresses the intuition that an organization can take on a life of its own. The U.S. Supreme Court has increasingly accorded personal rights to corporations. According to NPR, the court has ruled that “corporations have the right to spend money in candidate elections, and that some for-profit corporations may, on religious grounds, refuse to comply with a federal mandate to cover birth control in their employee health plans.”

SCOTUS decisions are just a blip compared to the longue durée; the personhood of corporations goes back a long time. Wikipedia describes how the late Roman Republic “granted legal personhood to municipalities, public works companies that managed public services, and voluntary associations (collegia) such as the early Catholic Church.” 

So, can one observe intention and agency in a company as a whole, as distinct from individuals like leaders and managers? Gabor raised Facebook as an example. Mark Zuckerberg controls the company through his shareholding and can even control the outcome of matters submitted to the shareholders. However, he can no longer control the influence Facebook has on our society. I’d add that Zuckerberg can’t control the beliefs and actions of all his employees and thus the ethos that influences Facebook’s products. Zuckerberg (and the culture he created) can and does shape their beliefs and actions through the ‘culture maintenance’ methods mentioned in the PressBooks article, like attraction-selection-attrition, employee onboarding, and reward systems. However, employees have minds somewhat of their own, and the mismatch between what Bay Area employees value and what the company’s DC lobbyists need has been exposed by Republicans in Congressional hearings.

Ramiro agreed that Zuckerberg can’t control Facebook’s influence on society, but he argued that Facebook attained its influence through the choices it made along the way. Zuckerberg has evolved the company’s goals, from “an online directory for students” (2004) to “the amount of money we make [is] an important thing for us” (2009) to “build community and bring the world closer together” (2017). (See also "The Evolution of Facebook’s Mission Statement," 2009.) This has directly impacted the company’s culture, processes, incentives, and the algorithms it created.

Boundaries

One of the striking differences between orgregores and biological organisms is that orgregores don’t have sharp boundaries. People straddle different organizations, for example, when employees work in standards or professional bodies and behave in ways that don’t serve their employer’s immediate interest. Joint ventures combine the employees of different companies to achieve goals neither none pursue on their own. Industries that consist of different companies (and of those companies) can have blurry edges but yet have goals and needs. An example is trade associations that lobby government. Sometimes different company subsidiaries are in different industries and are at odds with each other. For example, sometimes the message of company advocacy at the FCC depends on which part of the business holds the pen.

Gabor noted, and Ramiro agreed, that big data and data science could help to draw approximate boundaries by collecting data on inter-personal interactions. However, these boundaries won’t be sharp and will vary depending on your analysis. Gabor also pointed out that one could map the informal power within an organization just by sending in an experienced management consultant to observe people. 

In principle, data on information flows on a social network (e.g. email message endpoints and content) could indicate where the boundaries are. Marc Smith has pointed me Innovisor, Optimice and Swoop Analytics, but they seem to be tools that are used inside companies. Information doesn’t seem to be published (understandably, given privacy and competitive concerns), so it doesn’t help outsiders to map communication flow.

The harder problem is whether something that doesn’t have hard boundaries can be compared to biological organisms. Damasio explains that consciousness as being rooted in the sense of ownership of neural maps that comes from maps of the internal state of an organism running in concert with maps of the outside world, and inner cognitive processes (my poor paraphrase). His analysis seems to rest on the existence of a hard boundary between inside and outside (pace the microbiome, which counts as being inside). 

In principle I don’t see why a fuzzy part of network can’t have feelings and be conscious in a Damasio sense, particularly if there’s a relatively marked drop-off in connectivity creating a not-too-fuzzy boundary. However, I can’t think of an example. Even octopuses, the paradigm candidate for multiple semi-distinct consciousnesses in the same body, have all their neurons inside one bag.

Endnotes

[0] I’m not sure what to make of this, but “corporate culture” seems to be a recent concept. With all the usual caveats, this from Google Ngrams:

“Organizational culture” and "corporate culture" suddenly appeared in 1980 and grew much more more rapidly than just “culture.” In the last 20 years "organizational culture" has become more popular than “corporate culture.” (Note that the scale on "culture" has been shrunk 100x compared to the other two.)

[1] Another quote from the PressBooks article about organizational culture resonates with neurons not being aware that they’re part of a brain: “Culture is largely invisible to individuals just as the sea is invisible to the fish swimming in it.” However, the claim isn’t very strong since the individuals are probably largely unaware of all their “assumptions, values, and beliefs,” not just the ones related to the organization.