Summary
Contra popular interpretations, Jo Freeman’s essay, “The Tyranny of Structurelessness”, does not argue for hierarchy. People misinterpret Freeman because they falsely equate Hierarchy with Structure. I provide formal and empirical examples of non-hierarchical structures.
Again and again I hear arguments coming from Hackernews for Hierarchy, supposedly based on a very good essay called The Tyranny of Structurelessness, by a feminist named Jo Freeman.
Just today, in an essay called 100 Years of Whatever this will Be, I read:
…The Tyranny of Structurelessness by Jo Freeman. You should read it. The summary is that in any system, if you don't have an explicit hierarchy, then you have an implicit one.
The main problem with this claim is that the opposite of structurelessness is not hierarchy. There are all kinds of non-hierarchical and quasi-hierarchical structures, like rings and lattices. Moreover, we still lack a satisfying system of (meta)classification of these possible structures.
Perhaps a better way to summarize Freeman’s essay would be:
The right structure for the right task.
For some reason, the HN crowd keep trying to use Freeman’s essay in a contrived binary between hierarchy and non-hierarchy, though these terms never occur a single time in the essay itself, nor can we make sense of them formally.
ctrl-f
the Document, and you will find 64 occurrences of the term “structure” and 0 occurrences of the term hierarchy.
Consider this quote from the start of the essay:
Contrary to what we would like to believe, there is no such thing as a structureless group. Any group of people of whatever nature that comes together for any length of time for any purpose will inevitably structure itself in some fashion. The structure may be flexible; it may vary over time; it may evenly or unevenly distribute tasks, power and resources over the members of the group.
Here, Freeman is clearly arguing that governance structures may be either hierarchical or non-hierarchical. And anyone who has taken an A+ course should know this! Many possible structures are not hierarchical.
Thus, in seeking to avoid Hierarchy, a group may descend into Tyranny, because they avoid Structure altogether.
Structure → Hierarchy → Tyranny
This does not mean that we eschew Hierarchy altogether! But we can often avoid coercive hierarchies at the human-contact level, precisely via non-hierarchical and quasi-hierarchical structures, like rings and lattices.
At the end, Freeman offers her “design principles.” Most of these design principles are for dynamic, ring-like structure to prevent coercive hierarchies: Delegation, Responsibility, Distribution, Rotation, Allocation, Diffusion of Information, Equal Access to Resources.
Most of these principles are precisely for non-hierarchical modes of organization.
Examples from Governing the Commons
In Governing the Commons, Political Economist and Nobel Laureate Elinor Ostrom analyzes a number of collective governance structures for managing so-called Common-Pool Resources (CPRs), and comes up with her own, similar but different principles.
Consider Ostrom’s treatment of the Spanish Huerta system, and the governing role played by the farmers themselves:
While waiting, it is relatively easy to watch what those [farmers] ahead of him are doing and watch the ditch-riders, whom he is paying. The ditch-riders patrol the canals regularly and are watched over by the syndic, who can lose respect, and his job, if the allocation of water is not handled fairly and according to the farmers' rules. Challenges to the actions of a syndic, a ditch-rider, or another irrigator can be aired weekly before the Tribunal de las Aguas.
Of Ostrom’s own design principles for CPR governance, the most relevant principle for the question of hierarchies is her 8th and last:
For CPRs that are parts of larger systems:
8. Nested Enterprises: Appropriation, provision, monitoring, enforcement, conflict resolution, and governance activities are organized in multiple layers of nested enterprises.
Now this sounds hierarchical. But the term hierarchy is still too vague. A semilattice may well-represent “nested enterprises”, but can we even call this structure hierarchical per se? A lattice represents a mixture between a tree-like structure and a ring-like structure. Is is semi-ordered, which means that you can get some concept of depth, but if you try to traverse the nodes programmatically, you could walk forever.
Because of its unique properties, lattices are commonly used for distributed computing, e.g., in Conflict Resolution Data Types (CRDTS). They are also useful for the kinds of knowledge-representation systems that run Google Voice and Alexa. Such structures are themselves limited in their own ways.
Level of Application
The right structure for the job. Beyond our attempts to meta-classify structure per se, there’s the hard question of correct application. It is one thing to say that a Museum needs hierarchies for its taxonomies and it’s another thing to say that a Museum needs hierarchies for its management structure. Invertebrates and Bosses aren’t supposed to be the same kind of thing, relative to an employee. Thus, the level of application of a structure determines its relative usefulness to human life. When Ostrom argues for “nested enterprises”, she is arguing for a quasi-hierarchical ordering of the institutions themselves, not of the human-to-human relationship. So I might add an addendum to my prior summary of Jo Freeman:
The right structure for the right task, at the right level of application.
Conclusion
Hierarchies may have their place in governance discourse, but it is a mushy, mathematically-vague, politically-contested space. The term Hierarchy is certainly not clear enough to be propped up as providing a meaningful dichotomy with Tyranny. Jo Freeman does not argue for hierarchy, but for the correct structure for the job. The Hierarchy/non-Hierarchy dichotomy itself needs clarification, or it just needs to go away.