Online Debates

Communist Dictatorships in Our Midst?

29/9/2026

According to Elizabeth Anderson, in her 2017 book Private Government, most workplaces in liberal democracies are communist dictatorships. After all, the employer owns the means of production, which they organize by hierarchical central planning, by minutely dictating their employees' actions and sometimes even speech, without consultation and without appeal, all the while surveilling them to ensure compliance. We are outraged with communist states do this to their citizens, so why should we not be equally concerned when corporate employers do this to their workers? Why should private government get less scrutiny than public government?

INSTRUCTIONS:
• "Leave a Reply" to this entry to start a new discussion thread.
• "Reply" to a comment to contribute to a pre-existing discussion thread.
• Don't forget the "Notify me of new comments" checkbox if that is what you want.
• Limitations: Comments are limited to about 800 words
• Contact the moderator if you have any problems or questions.

5 Comments

Joe

29/9/2026 06:07:02 pm

According to Elizabeth Anderson, "most workers in the United States are governed by communist dictatorships in their work lives" (39), and she thinks that the private governance of firms (by their CEOs) should be replaced with public governance of them (to include their workers).

But why should the workers have a say? That's not the deal. The employer owns the job, from which he aims to pursue his goals, and the worker primarily owns only her own mental and bodily powers, from which she aims to make a living. Now there is presumably no moral obligation for them to interact. Even so, neither finds this desirable, for if the first is to realize his goals then he needs the job filled, and if the second is to make a living then she needs the money. And so, as a means to their separate ends, and hence for mutual benefit, they make a deal: that she exercises her powers performing his job, that she takes his orders as he sees fit, and that in return he pays her a wage. This deal is justified, since the end justifies the means, if the end is permissible and the means is not wrong-in-itself, and since each of them chose it over non-interaction, free from force or fraud by the other.

Clearly, Anderson rejects this argument, so, apart from its conclusion, what does she think is wrong with it in particular?

She might have problems with the presumption that there is no obligation for employers and workers to interact. She thinks it was a doubled-edged sword for nineteenth century employees to be liberated from bondage to their employers, since their employers were at the same time liberated from any obligation to employ them in turn (60-61). Certainly, communitarians will insist that it is not completely up to us whether we interact with others, since we are all members of the same community, and that this applies to employers, who therefore have no right to simply walk away, and, since they own the jobs, must employ workers. The short response to this, however, is that all individuals have a moral right to decide who they will associate with, and the terms under which they will do so. To think otherwise, to think that the individual may not declare independence from the community, is, in my view, to think that they are a slave to the community.

She certainly does object to the wage-contract, in particular the clause stating that workers take orders as their employers see fit. There are two points.

First, the default wage-contract gives dictatorial power to the employer. "In the United States, the default employment contract is employment at will ... [under which] workers, in effect, cede all of their rights to their employers, except those specifically guaranteed to them by law, for the duration of the employment relationship" (53-4). Actually, given our claim above, that individuals have a moral right to decide the terms under which they will associate with others, it follows that employers anywhere have a moral right to put any conditions they like on the jobs they offer. She provides an extensive list of some such conditions (37, 38, 52): sanctions for consensual sexual activity, choice of sexual partner; regulation of off-duty political activity, or social-media output; and sometimes even degrading regulation of bodily functions, such as insisting that factory workers not use toilets, and urinate in their clothes instead (66); etc. Of course, most employers do not impose such draconian conditions (63), but that is irrelevant, since they claim is that they have the right to demand them, if they wish.

Second, such dictatorial power is unjustified, even with free entry and exit from the contract. She gives a forceful argument: "Under [historical] coverture, a woman, upon marrying her husband, lost all rights to own property and make contracts in her own name. Her husband had the right to confine her movements, confiscate any wages she might earn, beat her, and rape her. ... Imagine a [hypothetical] modification of this patriarchal governance regime, allowing either spouse to divorce at will and allowing any clause of the default contract to be altered by a prenuptial agreement" (61). Historical coverture was clearly unjust—a "contract into subjection", Anderson says— and this does not change by adding free entry and exit, nor does it change if (as a matter of fact) husbands rarely exercised such draconian rights over their wives (64). But, as we have seen, this is exactly like the default wage-contract.

The argument above relies on the seemingly sensible claim that any interaction that is mutually chosen over permissible non-interaction is itself permissible, if the choices are free from force or fraud by the other. It looks as if that is not so. How do I respond?

Joe

2/10/2026 09:41:30 am

Response to Anderson (Part 1) ...

So what are the moral limits of contract? Respect for humanity, I say, which, I gloss as the claim that you must never treat a person as a thing, but must always respect their capacity to choose for themselves how to live—you must never treat them as a mere means, nor as a mere obstacle, nor with complete indifference. This implies a duty of negative respect, which is never to purposely frustrate their attempts to live as they choose (eg, lying to get their help, threatening them with harm), and a duty of positive respect, sometimes to purposely facilitate their attempts to live as they choose (eg, never ignoring requests if only to politely decline them, helping them in small ways, etc). But the first is much more stringent than the second. You must never push someone into the river to drown them, but there is a limit to the costs you must incur to save them if they were drowning already. And the limit seems quite low, since, according to Peter Singer (Practical Ethics, 3rd edn, p 201), a mere $1200 will save the statistical life of someone suffering from extreme poverty, but, from the way we behave, few of us believe that we are obliged to repeatedly donate $1200 to save such lives until, as Singer puts it, we begin to lose something of “comparable moral significance” to a life. So, do the contracts Anderson discusses cross the limits of respect for humanity, so understood?

You might think not. Remember, they are not obliged to interact at all—so the employer is not obliged to help the worker in any way. And their deal is free from force or fraud by either party—so the employer respects the worker’s choice to take the job, and any subsequent revocation of that choice. But—comes the obvious objection, couched in terms of coverture—the woman’s choice is to become a chattel slave (near enough) to her husband, which is hardly consistent with respect for her humanity, and even if the man himself does not force her to choose this, her circumstances must be very dire indeed if she prefers chattel slavery, in which case he exploits her whenever he exercises his patriarchal rights.

This is a good objection, but still arguable. First, it is unclear exactly what respect for a person’s humanity actually implies. For example, everyone values non-violence. But the pacifist honors non-violence, and will never do violence even if that allows greater violence overall, while the pragmatist promotes non-violence, and will do violence if there will be less overall. Who respects non-violence more? It is not so clear that the pacifist is wrong, nor similarly that it is wrong to think that we should prioritize the woman’s initial choice. Second, if we should prioritize the woman’s initial choice, then it is unclear whether the man is obliged to make any accommodations in his dealings with her, even if her dire circumstances are due to injustice. Thus, marriage still occurred under coverture, but that was only because women’s options had been so severely, and so unjustly, restricted. Even so, if we have not wrongfully caused their situation, and so long as we do not wrong them now, we are not obliged to ameliorate the situation of those who have suffered injustice, and the same applies to the man. Rather, the perpetrators of injustice have that obligation (thus, society for the restrictions it placed upon women), and, failing them, perhaps society as a whole, which obligation will fall equally on us all.

But, even if it is arguable, this response will convince few. To repeat, we have a choice. On the one hand, (i) we might prioritize each person’s choice right now, like the pacifist, so that respecting their humanity involves respecting their choice now, even if it means they will have (very) fewer choices later. Obviously, this is out, if we think that coverture is a contract into slavery. (But if it is not wrong to sacrifice one’s life for others, then why is it wrong to become a slave for them? At least that way one is still alive.). And so (ii) we shall prioritize each person’s choices over time, like the pragmatist, so that respecting their humanity means we won’t respect their choice now, if it means they will have (very) fewer choices later. We will not let them choose anything makes them look “too much” like a thing, or, as the lingo has it, “objectifies” them. By this understanding of respect for humanity, then, do the contracts discussed cross the limits? . . .

Joe

2/10/2026 09:42:22 am

Response to Anderson (Part 2) ...

. . . It depends on what the contract demands from the employee.

First, some make demands that are not wrong-in-themselves, but are irrelevant to the employer’s productive goal. Anderson accepts the theory that economies of scale and costs of contracting explains why hierarchically-organised capital-intensive firms are the most efficient mode of mass production (52), which presumably justifies their existence, but insists that “it does not explain the sweeping scope of employers’ authority over workers ... for example, why employers continue to have authority over workers’ off-duty lives, given that their choice of sexual partner, political candidate, or Facebook posting has nothing to do with productive efficiency” (52, italics added).

There are a number of things to say about this. First, some of these things may well be relevant to the employer’s goals—thus, Israel Folau’s Instagram post condemning gays to hell was entirely relevant to Rugby Australia’s goals. Second, it is not clear why the goals should be understood narrowly—thus, one might think that RA’s “productive” goal is to organize rugby competitions, but it is permissible for them to add others, such as the promotion of social inclusion, which is why Folau’s post was relevant. (Any social goal? Could RA have adopted as a goal the promotion of fundamentalist Christianity?) How many demands does this leave? Any? Maybe. But why would an employer want the right to tell people what they can and can’t post on Facebook, even if that has nothing to do with their goals, broadly conceived? The only explanation I can think of is that they simply enjoy exercising control over others for its own sake. And that attitude is wrong. To treat people as play things, just because you can, is inconsistent with respecting their humanity.

Second, other contracts do the opposite, they make demands that may well be relevant to the productive goal, but are wrong-in-themselves. An egregious case in point: “Nabisco once threatened its female production line workers with three-day suspensions for using the bathroom, and ordered them to urinate in their clothes instead” (62). More generally, some employers demand the right to treat their employees “too much” like things, in various ways: (i) devaluing their autonomy, by being too controlling—eg Amazon’s notorious workplace practices, perhaps; (ii) devaluing their subjectivity, by dismissing their feelings as not very important—eg, Nabisco; or by (iii) violating them, treating their physical integrity as of little importance—eg, workplace sexual assault, by the employer. (But here is a dilemma. It isn't wrong to hire someone as a stripper (notwithstanding recent public events), nor, obviously, to hire them as an accountant; so it isn't wrong (even if it would be unusual) to hire someone as a stripper so long as they also do accounts for the boss on occasion — but, in that case, how (if at all) could it be wrong (as most people think) to hire someone as an accountant so long as they also strip for the boss on occasion?) To treat people in this way, or even just to demand the right to treat them like this, is inconsistent with respecting their humanity.

How far do these considerations restrict the demands that an employer may make? Obviously, they have an absolute negative duty never (themselves) to violate others. However, the duty not to be “too” controlling or “too” dismissive is only a positive duty (we might add here the duty, if such it be, to ensure protection from sexual assault by other workers), since even Anderson admits that some degree of hierarchy is justified, and so justified to some degree to order people around (thus controlling their actions) whether they like it or not (thus dismissing their feelings). But how far does that positive duty extend? Who knows? However, since, as we have seen, it is permissible for there to be a considerable discrepancy between the weight a person gives to their own interests over those of others (they do not have to spend $1200 to save a life), then the permissible working conditions will likely still allow considerable alienation in workers.

So we can agree with Anderson about a lot, but not everything. She considers four strategies for protecting workers. Three of them—right of exit, rule of law, and basic constitutional rights—all seem consistent with ensuring that people’s humanity is respected at work. But she prefers the fourth—giving the workers a voice in their government—and this is not consistent, if employers do not wish to give them a voice. For, so long as the demands an employer makes are consistent with respecting the humanity of their workers, as understood above, then the employer may indeed be the only one who gives the ord

Ari

2/10/2026 12:14:44 pm

OUTLINE OF ORAL ARGUMENT
PART I: THE ISSUE
1. The central question posed by Dr Elizabeth Anderson (hereinafter Anderson) in the reading is “Would people subject to such a [private] government be free?” [38]. The answer, contrary to Anderson’s numerous assertions and protestations throughout the reading, must be either A. the question fails on its own merits or B Yes.
PART II: IT IS INAPPOSITE TO COMPARE “PRIVATE” AND PUBLIC GOVERNMENT
2. Anderson’s argument proceeds from two basal propositions. The first involves unflatteringly comparing the relationship of ruler and citizen to that of boss and worker. The second involves drawing a narrow thread to expand the concept of “government” to cover certain historical uses. Neither can sustain the conclusion that “private government” is comparable to public government in respect of its tendency to limit freedom.
3. One point of comparison between government and employment that Anderson consistently emphasises is the (purportedly inadequate) remedy of “emigration” which parallels the sanction of “exile”. The first point here is that this equivalence exaggerates the costs of “exile” from a job and minimises the consequences of “exile” from a country. The second point is that tyrannical countries have often forbade emigration, meaning that there is a meaningful distinction as to the availability of the remedy between cases. And the final point is that due to Anderson’s concession that a ruler holds more power over their citizens than a boss does over their workers, it is conceivable that “emigration” may hold as a satisfactory remedy for the latter but not for the former.
4. Anderson also tries to justify the concept of “private government” by drawing upon historical characterisations of what we now term “private” arrangements as constituting government. In a sense, this does speak to a general etymological and lexical ambiguity of the term “government,” but this does not, in and of itself, establish that “private government” shares the underlying morally relevant characteristics which substantiate the conclusion that government limits freedom. Also, she likely proves too much with this idea.
5. Accordingly, and for these reasons, “private government” is not meaningfully comparable to public government, making it inapposite for Anderson to draw said comparison. Thus, the question fails on its merits.
PART III: ANDERSON MISCONSTRUES FREEDOM
6. Further to the above, in attempting to explain how “private governments” limit freedom, Anderson misconstrues the idea to such an extent that it ceases to be workable as a moral or metaphysical benchmark. The obvious limitation of her approach is that she considers it important to balance different “types” of freedom and does not recognise there is only one type of freedom, being what she calls “negative freedom”. This already deficient characterisation is further polluted by her assertion that pure negative freedom entails an anarchic arrangement where individuals are in competition not only with respect to the natural goods of the world but also with respect to their very freedom itself, which is inconsistent with a proper construction.
7. Negative freedom (or simply freedom), properly construed, is non-rivalrous, such that each person’s freedom must be consistent with every other person’s freedom. This is because it derives from the reflexivity of Kantian Categorical Imperative, which holds that the freedom of individuals must be universalizable and if it is not, it cannot be considered true freedom. Therefore, to the extent that “a kind of freedom” (such as positive freedom) violates or limits the freedom of others, it is not freedom but rather a kind of power that is held and exerted over other people, which, aside from being undesirable, is in fact a limitation on freedom.
8. What this means for employment in general is that the reciprocal constraints imposed by the employment contract do not limit the freedom of the respective parties, because they are an expression of both the boss and worker’s freedom of contract, and are specifically promised in exchange for the respective counterparty’s obligations under the contract. Accordingly, because these limitations on and obligations of the parties are the expression of their joint will and freedom, for either to exceed the limitations or ignore the obligations on the basis of their purported “freedom” would be to, as set out above, exercise a kind of power against the other party, and hence impose a limitation on the other party’s freedom which is impermissible due to its reflexivity.
9. Accordingly, and for these reasons, “private government” does not meaningfully limit people’s freed

Joe

2/10/2026 02:23:27 pm

Hi Ari. Some responses to your comments:

> due to Anderson’s concession that a ruler holds more power over their citizens than a boss does over their workers, it is conceivable that “emigration” may hold as a satisfactory remedy for the latter but not for the former.

Do you think this applies to coverture? If exit from employment is less costly, then this might make it unlike coverture (even the modified form) at the time, since the consequences of exit for the woman were much more severe. So the acceptability of such contracts may actually depend on market conditions. She actually refers in passing to "market pressures" (63) as keeping employers in line, and her criticism of coverture, oddly, is not that beating your wife disrespects her humanity, but that that would still leave "the vast majority of men" (61) with patriarchal authority over their wives, which seems to suggest that if market were gender equal, so that women do not have to accept such awful contracts, then it would be OK to have such contracts (though, presumably, not many would exist)

> this does not, in and of itself, establish that “private government” shares the underlying morally relevant characteristics which substantiate the conclusion that government limits freedom.

On this point, I think that Anderson is correct, in that, if it really is dismissal at will, then the employer has very wide discretion in what they can order their employees to do.

> does not recognise there is only one type of freedom, being what she calls “negative freedom”

It is your view that is controversial on this point. The distinction between negative freedom (=being let alone) and positive freedom (=being able to do stuff) is widespread, though her addition of republican freedom (=not dominated by anyone) is an extra wrinkle

> her assertion that pure negative freedom entails an anarchic arrangement where individuals are in competition not only with respect to the natural goods of the world but also with respect to their very freedom itself, which is inconsistent with a proper construction

Yes, I noticed that too, but maybe she is right if you distinguish between negative freedom (= Hobbes liberty in the state of nature), and negative rights. As you suggest, libertarians do not endorse negative freedom (implying we can do whatever we like to each other in the state of nature), but negative rights, and deny positive rights, and (no doubt) republican rights.

> the reciprocal constraints imposed by the employment contract do not limit the freedom of the respective parties

Depends, on the meaning. True, as I argued in my own response, the contract cannot have one person allow "too much" infringement of their default negative rights (ie, allowing themselves to be beaten etc). False, since even a legitimate contract may involve quite extensive control over the other person (think about the military), and it seems crazy to say that the day-to-day negative freedom (= being left alone) of soldiers on patrol has not been limited. At the extreme, doesn’t an (irrevocable) contract into slavery show this? It wouldn’t be a contract into slavery if slavery wasn’t the result. (I think the same argument applies, even if the contract is revocable).

Debates

Communist Dictatorships in Our Midst?

Leave a Reply.

Members