The Fall of MeaningChapter 4

Freedom

The care had made room for a person's judgment. Beyond the care, a life still had to be directed.

I followed that unfinished movement into the arrangements through which people tried to begin something of their own. The scale widened from assistance within a day to the means of changing its course. Here the catalog's entries on access, dependence, and obstruction lay close together. Their proximity was useful. Their names were less reliable as a guide to what help accomplished.

A person relying on others might have acquired room to act through that reliance. I wanted to see what the assistance changed before deciding what the dependence meant.

Among provisions for changing where people lived, practical help had made a difference that an available subsidy and information alone had not achieved. Assistance reached into the work of finding and securing a place. More people then moved, and the subsequent pattern of remaining in their new neighborhoods, together with greater satisfaction, gave reason to take the change seriously. Earlier residence had not been a sufficient guide to where those people wished to live.1

What caught my attention was the change this made to the past. The earlier conduct remained exactly as recorded. People had stayed. Yet once help made another course practicable, staying became harder to interpret as an uncomplicated preference for the existing one. The intervention had changed an opportunity in the present and weakened an explanation of conduct before it.

I could not replace that explanation with a claim that everyone had secretly wanted to leave. Some people had reasons to stay. The important loss was the observer's easy certainty. A record of what someone did could no longer carry, by itself, an account of what they had been able to do.

The work between an offer and its use now held my attention. Someone had to find a way through requirements, bring the relevant parties into contact, and keep an undertaking possible while those parties responded. In the successful provision, help had reached some of that work. The option became more than a description of a destination.

This gave the assistance a value that disappeared if I measured freedom only by the absence of other people. The person had gained practical room through a relationship. Their dependence had not vanished when they moved. What changed was what they could do within a world that still required others.

The same question reached into judgment. Before a course could be undertaken, its consequences had to become intelligible enough to consider. A person might need help to understand those consequences without needing someone else to settle their importance. That distinction drew me toward assistance which clarified the decision itself.

There, the change was sometimes visible in improved understanding and a more active part in choosing. Decision support helped people grasp benefits and risks, and probably helped more of them choose in keeping with what mattered to them. The useful result extended beyond obtaining a response. The person had more understanding with which to make it.2

I stayed with that difference. Influence was present in the explanation, in the selection of relevant information, and in the effort to make it understandable. Its presence did not tell me that judgment had been displaced. To decide that, I needed to follow what happened to the person's capacity to judge. A changed answer was too small an object to contain the whole relation.

This also made the apparent simplicity of a recorded choice less persuasive. The answer arrived at the end of conditions that helped produce it. Where assistance enlarged understanding, those conditions belonged to the achievement. Where they remained unexamined, the recorded answer could not tell me how much direction the person had exercised.

I had followed Dignity out of care expecting to examine what room remained around a protected life. The successful assistance gave that room a more definite form. Other people had helped make a course available, and sometimes helped make its consequences intelligible. I wanted to follow that possibility beyond the point of choosing a direction.

A direction has duration

Once a course began, its direction had to be sustained over time. What happened when the person later wanted to change it?

Some people chose to put resources beyond their own immediate reach so that a purpose would survive the interval before they fulfilled it. They asked an arrangement to preserve a limit they might later wish to cross. The desire for that commitment was real, even though the observed results did not establish that the restriction itself improved their lives.3

The arrangement resisted a simple measure of freedom. Making every reversal effortless would remove something the person had chosen. Making the commitment absolute would give the earlier decision power over circumstances it could not fully anticipate. The same life occupied both moments.

The earlier purpose did not become another person's imposition merely because it had acquired duration. Nor did its origin in a choice settle every later demand. A change of mind might abandon an undertaking too easily. A change of circumstances might make continued fidelity ruinous. Calling both a reversal preserved their common direction and lost the difference that mattered.

I did not need to decide in advance which moment should govern. I needed to keep both in view. What was the commitment meant to preserve? What had changed around it? What would maintaining it now require of the person whose purpose had first given it value?

The questions became harder when the undertaking included others. Their contributions and reliance did not disappear because one participant wanted different terms. Yet the existence of those contributions could not establish the necessity of every restriction attached to them. A legitimate purpose gave me something to examine; it did not complete the examination.

This was where the distance between a restriction and its justification became perceptible. The restriction could be described in a sentence. Its justification had to reach the purpose it served and the consequences it imposed. I could not make that distance vanish by calling the arrangement either voluntary or coercive before following what it did.

The wish to continue now mattered alongside the wish to change. A person might want the undertaking, but on terms they could help alter. I began to look more closely at the powers available to people who remained.

While people remain

An unused route out of an arrangement could still matter inside it.

Restrictions on changing employers brought this into view. Greater enforceability of those restrictions was associated with reduced movement and earnings, with findings consistent with diminished outside opportunities affecting bargaining. This drew my attention beyond the moment someone attempted to leave, toward the alternatives surrounding the relationship and their bearing on its terms.4

That changed what I looked for when I saw continued participation. Remaining was an action. The power to affect the conditions of remaining was another matter. A person who could realistically go elsewhere occupied a different relation to the existing terms, even if departure never occurred. The route's value could be present without appearing as a journey.

An unused opportunity did not establish effective bargaining. Remaining did not establish an inability to depart. The evidence drew my attention to a capacity that the record of departures alone left unresolved. Movement and the capacity to move were connected, but they were not interchangeable observations.

There was also something being supplied within the relationship. In separate research, greater enforceability was associated with more employer-provided training. Restriction appeared alongside an investment in people's capacities. I had to account for that investment too.5

The value did not cancel the limit. The limit did not make the value imaginary. I kept the two consequences together because separating them made either explanation too easy. Looking only at restraint left the acquired means out of view. Looking only at the investment left out the reduced power over where those means were used.

Whether the restriction was necessary remained a further question. The presence of training did not answer it. That required a comparison of ways to sustain the investment and what each demanded of those receiving it. The findings before me stopped short of that comparison. They left a gain in capacity alongside a loss of alternatives.

That combination kept my attention on the terms of remaining. Departure was one power relevant to them, but it could not exhaust the inquiry. People also had concerns they wanted an existing arrangement to answer. Acting on those concerns did not always require going elsewhere.

I followed channels through which workers brought safety concerns into a shared response. When requirements for joint committees were enforced, the committees became more active and some practical conditions improved. These gains were achieved while management retained operational authority. The channel gave concerns a way to matter within the continuing workplace.6

The distinction lay in the response. A concern could be recorded without altering the conditions that produced it. Here, some safety conditions had improved. The channel's activity had reached beyond its own meetings, though the workers had not thereby acquired control over the workplace.

The response was also uneven. The results were stronger where existing management practices were better; disappointing results elsewhere prevented me from treating the channel's formal presence as its completed function. A route to an authority still depended on what that authority could and would do when something arrived through it.

I returned to the work between an available course and its use. At the opening of the inquiry, assistance had helped a person make that passage. Here, enforcement and the capacity to respond belonged to the passage through which shared concerns became practical changes. The dependence had a different form. The need to follow it into action remained.

These channels did not establish every freedom of association. A committee operating within management's authority was not the same as people freely determining how to organize. Its success on a defined concern did not tell me how dissent within it would be received. The gain was real within its reach. I had to preserve that reach to understand the gain.

I now distinguished two powers relevant to a continuing arrangement. The employment evidence linked outside opportunities to bargaining over the terms of staying. In the workplace channels I had followed, shared concerns had reached a response that changed some conditions. Neither power had to wait for the other to fail before it mattered.

The successful channel had brought me to the edge of its assigned purpose. Someone had determined what it could receive and where it could carry a concern. I wanted to follow the capacity to act together beyond that assignment.

Beyond the assigned channel

A communications system could be controlled without every consequence of communication being contained.

The distinction first required care about what the control allowed me to infer. Evidence of selective censorship showed particular attention to the potential for collective action. It did not establish that criticism was otherwise universally tolerated. Some apparently tolerated criticism had been solicited through official consultation. An invitation to comment and the space for unsolicited dissent could not be treated as the same opening.7

The invitation told me something about the channel's intended use. It did not tell me everything people could make possible through communication. For that, I needed to follow the action that occurred, not stop at the rules for speech or the record of its removal.

I remembered Orra's ceremonial entrances. I had called them decorative until a protest found them locked. The preserved appearance still governed an expectation, and the lock still governed access. I had learned to inspect what an obstruction did. It would have been a peculiar use of that lesson to let an obstruction tell me everything about the people confronting it.

On Earth, the movement of information between cities was linked to the spread of actual protests and strikes, even within a heavily controlled communications environment. The effect extended beyond content explicitly telling people how to organize. Reports of action elsewhere also mattered to its spread.8

That was the part I wanted to follow. A report did not have to arrive as an instruction to become relevant to someone's next act. The occurrence of action elsewhere was itself something people could learn. What they made of it belonged to a further relationship, among those receiving the information and the circumstances in which they received it.

The particular pathways were not fully settled. Information might change an assessment of what was possible; an event might carry anger or encouragement beyond its location. I left those explanations open and followed the distinction the observed action supported. Information moved, and the spread of collective events was not explained simply by counting explicit instructions that survived removal.

This placed a limit on an account in which control of the medium settled the capacity of its users. The apparatus of control remained consequential. Yet people were still doing something through the environment it governed. Their actions prevented the description of the apparatus from becoming a complete description of their possibilities.

I found the incomplete fit more revealing than another example of suppression would have been. Suppression was familiar. The harder relation was that the same field contained obstruction and consequential action. I could neither erase the obstruction because people acted nor erase their action because control persisted.

The action also had to be allowed its own scale of success. A protest's occurrence did not prove that its demand was met. The spread of a strike did not establish safety from retaliation or a durable organization afterward. Those questions remained. But the capacity to undertake something together had entered the world as conduct, and an account of freedom that omitted it would be missing an event rather than merely an aspiration.

This differed from the responsive channel I had followed earlier. There, a defined concern reached an authority and some conditions improved. Here, communication contributed to action without establishing the desired final response. The first did not prove unrestricted organizing. The second did not prove successful reform. Together they widened the field in which I had to look for agency.

People within that field were doing more than selecting among existing offers. They had brought concerns into common efforts, sometimes with practical consequences. That capacity arose among them. Looking only at one person's permission to act, or only at the authority setting a channel's rules, left those relations out of view.

I had no reason to declare the whole arrangement free. I had a more exact reason to resist declaring every action within it empty.

With those shared capacities still in view, I returned to the successful help. It had addressed work that the recipient would otherwise have had to manage. Its usefulness depended on that fit. An identical offer, left at the same distance from everyone, was not necessarily an identical practical opportunity.

That did not give me a rule for distributing assistance. People had different purposes as well as different needs. But it changed the question I would have to ask of the distribution. If a person needed additional means to make a direction practicable, the difference in what they received could be part of honoring the promise.

I now wanted to know which differences an order could justify, and which differences it merely expected people to bear.

Notes

  1. Peter Bergman, Raj Chetty, Stefanie DeLuca, Nathaniel Hendren, Lawrence F. Katz, and Christopher Palmer, “Creating Moves to Opportunity: Experimental Evidence on Barriers to Neighborhood Choice,” American Economic Review 114(5), 2024. The June 2023 author version was inspected. Two randomized phases among housing-voucher recipients in Seattle and King County compared support packages. The full package outperformed information and financial assistance alone; its individual mechanisms were not independently randomized. Follow-up residence and satisfaction inform the interpretation of prior choice, without establishing universal preferences or later adult earnings for children. The chapter uses this bounded relationship, not a planetary prevalence claim. Author paper.
  2. Dawn Stacey and colleagues, “Decision aids for people facing health treatment or screening decisions,” Cochrane Database of Systematic Reviews, 2024, DOI 10.1002/14651858.CD001431.pub6. The public summary and structured abstract were inspected. The review includes randomized trials through March 2022; understanding and participation improved, while informed values-congruent choice has moderate-certainty support. This does not establish indefinite revisability, universal benefit from guidance, or a particular outcome for every recipient. Review.
  3. Nava Ashraf, Dean Karlan, and Wesley Yin, “Tying Odysseus to the Mast: Evidence from a Commitment Savings Product in the Philippines,” Quarterly Journal of Economics 121(2), 2006, 635–672. The product permitted a chosen withdrawal date or target, with specified exceptions. Offering it increased savings against no-contact controls, but the mean difference from a savings-marketing arm was not statistically significant; some other measures differed. Product components were not isolated and welfare remained ambiguous. The chapter uses voluntary demand for commitment, not a claim that restriction alone caused benefit. Its discussion of competing purposes through time is interpretive. Published paper.
  4. Matthew S. Johnson, Kurt Lavetti, and Michael Lipsitz, “The Labor Market Effects of Legal Restrictions on Worker Mobility,” Journal of Political Economy 133(9), 2025, 2735–2793. The published abstract reports panel evidence linking enforceability of noncompetes with mobility and earnings, interpreted through outside options and bargaining. Access in this research pass was at abstract level; the chapter uses no causal magnitude, detailed identification claim, or subgroup mechanism. The interpretation of an unused outside opportunity is bounded by that reported finding, not proof that every remaining worker has or lacks effective bargaining power. Published abstract.
  5. Evan Starr, “Consider This: Training, Wages, and the Enforceability of Covenants Not to Compete,” ILR Review 72(4), 2019, 783–817. The published abstract reports associations using Survey of Income and Program Participation data, including greater firm-sponsored training and lower wages with greater enforceability. It does not establish that a particular restraint was necessary or proportionate, or that investment caused an individual worker’s later dependence. This source and note 4 are separate analyses, not stages of one observed process. Published abstract.
  6. Laura Boudreau, “Multinational Enforcement of Labor Law: Experimental Evidence on Strengthening Occupational Safety and Health Committees,” Econometrica, 2024, DOI 10.3982/ECTA19408. The May 2023 author paper was inspected. Randomized rollout in 84 Bangladeshi supplier factories in 2017–18 bundled training and monitoring. Compliance and objective safety measures improved modestly; outcomes varied with baseline management practices, which were not randomized. Reporting can affect recorded injuries. The finding does not isolate association from enforcement, certify freedom from retaliation, or establish internal democracy and independent organizing. Author paper.
  7. Gary King, Jennifer Pan, and Margaret E. Roberts, “How Censorship in China Allows Government Criticism but Silences Collective Expression,” American Political Science Review 107(2), 2013, 326–343; Dimitar D. Gueorguiev and Edmund J. Malesky, “Consultation and Selective Censorship in China,” Journal of Politics 81(4), 2019, DOI 10.1086/704785. The later paper partly reanalyzes the earlier data and uses directives and a thought experiment; measurement limits prevent an ideal direct re-estimation. The chapter rejects a universal indifference-to-criticism claim. Observed removals omit some filtering and self-censorship and do not measure every effect on subsequent organizing. These sources do not establish a timeless or global censorship rule. Original paper; Reassessment.
  8. Bei Qin, David Strömberg, and Yanhui Wu, “Social Media and Collective Action in China,” Econometrica 92(6), 2024, 1993–2026. The published introduction’s design, findings, and mechanism discussion were inspected; supplements and estimator calculations were not independently audited. The city event panel spans 2006–17. Its observational causal analysis addresses correlated relationships and timing, with remaining dependence on the research design. The measured outcomes concern diffusion of protests and strikes, not concessions, safety afterward, or durable organization. Information, emotion, and perceived opportunity remain possible pathways. This analysis does not track the same participants as the censorship studies in note 7. Published paper.