Before going any further, two clarifications are in order. First, these two kinds of motivation are not mutually exclusive. Things we do for money often have an intrinsic pull as well — in fact, they should, and their combined effect can be reinforcing. Second, this distinction implies no priority or relative importance, no hierarchy of needs as in the work of Abraham Maslow, nor an injunction, as Bertolt Brecht so elegantly put it, to “first feed the face.” Each of us values the different aspects of work in our own way, and the purpose of this short discussion is to get a lay of the land, not to elevate one personal philosophy over others.

Of course, most jobs are undertaken, initially at least, mainly for extrinsic reasons, and the extrinsic benefits we want them to offer are fairly obvious and uncontroversial. A quick list should do: a good job provides ample pay and benefits. It offers a predictable schedule and the workload the worker is looking for. It comes with enough job security to reduce stress and make it easier to plan ahead. It doesn’t expose the worker to undue risk of injury or disease. It provides opportunities to learn skills and move up, with prospects for better future benefits. Realistically, any economic system has to provide enough of these things to draw workers to the work that needs doing, and a good system will go beyond this minimum.

It’s when we get to the intrinsic attributes that more nuance is needed. A natural entry point is to distinguish between the work itself and the workplace. For the first, I think there are two dimensions, the nature of the work and how purposeful it is. Good work of the first sort is mentally and/or physically satisfying. It hones expertise and craft, making us some combination of smarter, stronger, and more confident in our abilities. An essential aspect of this is variety, doing many different things rather than the same thing over and over. Often the most important step is to break up an overly rigid division of labor, so skill isn’t smothered by routine.

Purposefulness is more complicated. There are actually two requirements for meaningful work, neither sufficient without the other. First, workers have to recognize the connection between their own input and the ultimate product. If work is highly subdivided, or if parts of it are assigned to different teams working in isolation from each other, this connection may be lost. Recognition of this issue has led to sporadic managerial interest in more integrative forms of work organization. But the second requirement is also essential: workers have to feel that the end product is genuinely valuable and lends meaning to the work they’re doing.

Of course, what counts as a meaningful outcome depends on how well the values of workers are reflected in what they produce. This is subjective, but not entirely. If a firm exploits its customers or suppliers, its workforce is less likely to take pride in their role. Think about customer service call centers. Some really do offer the sort of help they advertise, while others just go through the motions. Surely the workers wearing the headsets know the difference, and the satisfaction they might derive from their hours spent on the phone will reflect this. My underlying point is that, in this respect, meaningful work depends on whether the business has a purpose beyond just making money.

This leaves the other source of intrinsic benefit, derived less from the work than the workplace in which it occurs. Much of this is simply the joy of social connection: making friends and being part of a team. How much of this a given job offers is contingent. Sometimes the chemistry is there and sometimes it isn’t. But jobs can be organized to provide more opportunities to socialize, from the layout of the work environment to the structure of the workflow. (Policies on working from home often reduce some kinds of social connection while increasing others.)

But being part of a community is about much more than friendly chatter. The workplace is a realm of life in common. It establishes rules and expectations for what its members should do, distributing the gains and burdens of collaboration. It is organized hierarchically, and for all these reasons it is thoroughly political. If workers think the workplace is run fairly, giving them the voice they deserve, they will benefit both by getting better treatment on the job and from the process that makes it happen. (This combines what social scientists call “outcome” and “process” benefits.)

Not surprisingly, then, self-management is a key element in the intrinsic value of work, but discussion of it tends to be overgeneralized. Let’s unpack the idea, based on the recognition that what we are really talking about is self-government in a workplace setting. Using the template of politics, I suggest three dimensions, each somewhat independent of the others. Let’s call them freedom, participation, and representation.

Freedom: in a work context, this means minimizing supervision and the imposition of hierarchical authority — in a word, autonomy. Autonomy may be individual if a single worker is allowed to do their work in their own way, out of management purview, and is responsible only for delivering a product at the end of the process. University professors, especially if tenured, have a lot of autonomy in this sense, which is why these jobs are sought after despite usually paying less than alternatives requiring similar credentials. But autonomy can also apply to a team, and many workplaces are organized on the basis of autonomous team production. All else being equal, reliance on co-responsibility rather than intrusive management gives workers more freedom to control their work, although the difference may be trivial if the production goals feel excessive or pointless.

Participation: We often speak of management as a group of people, and certainly some employees are classified as managers and others not. But management can also be understood as a set of decisions that need to be made at each level of the organization, often routinely. Those decisions can be shared, depending on the skills of the workforce, the way the work process has been organized, and the willingness of those in authority to collaborate with those beneath them on the organizational charts. Management theorists like W. Edwards Deming and Stafford Beer tried to convince the top echelons that shared decision-making was not only humane and respectful, but would also benefit the bottom line. Cross-training workers and giving them opportunities to learn management skills, they both thought, would allow the firm to devolve more responsibility on them. Beer in particular stressed horizontal communication between work units, so they could coordinate without having to involve, and be subjected to, higher-ups.

Representation: Most of the theorizing and advocacy in the realm of self-management has been about systems in which workers select representatives who bring their voice to firm-level decision-making. This is what unions do when they bargain collectively, and it is also a function performed by works councils and laws, as in Germany, that require worker representatives to serve on company boards of directors. It is at the core of the democratic justification for worker cooperatives, since each worker gets one vote on key questions of policy and leadership. The argument for representative democracy is broadly the same at the workplace as it is in society at large, although there are important differences based on who is a “citizen” of the workplace polity and what other groups of “citizens” ought to have a voice in those decisions as well. That is, enterprises are far more entangled with other enterprises and the communities they operate in than governments are with other governments — although there is a case to be made that national sovereignty is also too great in an interdependent world.

I’ve spent time on the workplace governance aspect of intrinsic motivation because enthusiasts for one dimension or another tend to subsume the entire question under their chosen focus. Worker autonomy, participation in decision-making, and representation in organization-wide issues are all valuable, but none of them ensures the others. Each in its own way points to attributes of a freer and fairer economic system than the one we currently inhabit.

So now let us return to Marx and his theory of alienation, since there are important lessons to be learned for the quality of work in capitalism and beyond. I take them to be these:

1. The schema here is similar to Marx’s but differs in important ways. For Marx, workers are alienated because they lose control over the things they produce; on the account offered here, there is little meaning in work if the end product isn’t seen as valuable. The problem with Marx’s version is that workers always cede control over their work products: they create them for others to take and use. Even an artist doesn’t expect to have final say over a painting once a buyer, like a museum, takes possession. And that can lead to frustration at times! But that’s how it has to be. The meaning a worker gets from producing something, on the other hand, usually survives its sale if it’s there to begin with.

   Another of Marx’s forms of alienation involves estrangement from one’s innate physical and mental capacities to create, and is virtually identical to the absence of the intrinsic benefit work should offer in employing and developing our creative potential. The difference is that, unlike Marx, this model doesn’t postulate that work has to be this way under capitalism or that abolishing capitalism is guaranteed to fix it. The third type of alienation, from the community of coworkers, presents a greater point of disagreement, however. Social stimulation doesn’t rise or fall with changes in the mode of production, although it can be spurred by encouraging more interaction at work. The claim that labor markets, in themselves, pit workers against each other, strikes me as overdrawn (despite the nightmare vision of Donald Westlake’s The Ax).

   Meanwhile, Marx offers no analysis of the workplace as a system of governance, simply assuming it is pure autocracy under capitalist ownership. I realize that Marx was a pioneer in this field, and it’s predictable that our understanding would improve over almost two centuries of further experience. However much his work should remain a vital starting point, it need not be the final word on the problems of work we face today.

2. It’s important to recognize that workers are not just on the receiving end of job quality. They bring themselves to the job, and what that entails — the degree of care and interest they exercise, how much they push back against unnecessary or exploitive rules and expectations — makes a big difference in what ensues. That said, good work invites engagement, and the bad kind makes it seem futile. Think of workers in meat processing plants who make the same cut, over and over, as carcasses glide by, one slice or more per second. We can acknowledge worker agency while calling out the jobs that simply crush it.

3. As the preceding discussion should make clear, sweeping terms like alienation can do a disservice if they give the impression that bad jobs are bad across the board and vice versa. Empirically, we find all sorts of combinations of good and bad attributes, although, for most of us, the bad are more frequent and frustrating. An interesting question is whether there is a correlation between good extrinsic and bad intrinsic features. This was the claim of libertarian philosopher Robert Nozick, who thought that if a job were more alienating it would pay higher wages in the labor market: competition would force employers to compensate workers monetarily for especially unpleasant or meaningless tasks. This parallels a similar claim about dangerous working conditions, which I have disputed elsewhere.

   Nozick’s argument is flawed but not dumb, and there are some jobs that work that way, although it’s not the norm. At the same time, a softer version of his point is worth considering: there are unavoidable trade-offs between the different things we look for on the job. For instance, how much a firm, even one owned by its workers, can afford to pay in wages is constrained by productivity and consumer demand, and both can sometimes conflict with measures that make work more pleasant and personally rewarding.

4. Marx’s formulation implies that alienation at work is our fate as long as capitalism prevails, while doing away with that system guarantees us an unalienated future. Both are up for debate. Enormous improvement is possible right now if we understand the changes we need to make and can organize to fight for them.

   The European Union’s directive on European Works Councils is a stronger example, although it could be stronger still. If it falls short of our vision for worker self-management, it at least advances one piece of it. Meanwhile, the question of just what should replace capitalism is a bit easier to answer if we are clear on the criteria for how work should be organized. Simply asserting that a general condition of alienation settles in when capitalism takes over and is lifted when capitalism departs is not helpful. Yes, the potential for de-alienating work is greater if we replace private with social ownership, but it’s not zero under capitalism, nor unlimited in an idealized socialist future.

To go deeper into how work can be made more rewarding, we would have to look closely at specific workplaces. For instance, what would it take for work at Amazon warehouses to become good, both extrinsically and intrinsically? Or in phone support centers, where workers guide users through the complexities of computer hardware and software? Or in fields and orchards, where crops are planted and harvested? Or at airports, or on road maintenance crews? Abstract theories need to connect with detailed knowledge based on firsthand experience. Rather than fulminating at alienation in a general way, what we need are broad-based movements to make work a lot better, incrementally now and systematically in the future.