There is the version of your opinion you give at work, and the version you give at your kitchen table. Most people run that calculation dozens of times a day and would struggle to say when they learned it.
Free speech is argued about far above that kitchen table: who may say what, who may stop them. Every argument about the right to free speech, however, eventually comes down to one person deciding whether to say a particular thing in a particular situation.
Psychology has studied that decision for decades, and what it describes is not really a question about speech. It is a question about audience, and about what it costs to be known.
Let’s follow that calculation outward as the audience widens. It starts with an audience of one, moves to the person you live with, then to the people who know a partial version of your life, then to a room that has already decided what it thinks, and finally to an audience you cannot see at all.
At the end it asks the question the psychology cannot answer on its own: what does a right to free speech actually change?
What Free Speech Gives and Costs
Before following the calculation outward, the sum itself. What does speaking freely actually buy a person, and what does it charge them?
Firstly, something loosens within when you finally say the thing. Not resolution, usually. Closer to a drop in pressure. Writing it down does it too, and so does saying it to an empty room, which suggests the relief is not entirely about being heard.
That effect has been tested: people randomly assigned to write or talk about something emotionally significant tend to come out slightly better than people who were not (Frattaroli, 2006).
Feeling like yourself, shows one of the clearest links to wellbeing (Sutton, 2020). Feeling authentic is an inner sense of congruence, not a policy of saying everything you think. At the same time, behaving differently in different settings does not make someone a fraud.
Speaking is also not uniformly good for all people at all times. Roughly a quarter of the experiments Frattaroli (2006) gathered came out the other way, with the people asked to disclose faring slightly worse off than the people who were not.
Experiments related to free speech have measured wellbeing rather than social fallout, so the everyday version is ours to supply: the confession that circled for weeks instead of settling, or the honest sentence you cannot take back and now have to live beside.
Which is why nobody literally speaks freely. Everybody calculates. The only question is what goes into the sum, and the largest factor in it is always who is listening.
When the Audience Is Small
Let’s begin with the smallest audience there is: another person in front of you.
What does free speech, or the lack thereof, cost here?
The flat voice. The arranged face. The “no, I’m fine” delivered so smoothly it barely registers as a choice.
Psychologists call this expressive suppression, and it tracks modestly with poorer mental health (Hu et al., 2014). At the same time, rethinking a situation so it lands differently, which is called cognitive reappraisal, tracks good mental health outcomes.
Now add one more person to the audience, and keep them there for years, like a family, or a group of friends. The same behaviour can change shape entirely. Inside a relationship it looks like a self that has quietly contracted around keeping things steady. Opinions no longer offered because offering them costs an evening. Wants that stopped being mentioned so long ago you would have to think hard to remember what they were. Mostly this comes from ordinary consideration rather than fear, which is part of why it goes unnoticed for years.
Psychologists call it self-silencing, and its association with depression is clear (Pintea & Gatea, 2021).
The silence may not be the problem. What the silence is protecting might be.
When the Audience Is Everyone
Widen it again, past the people who know you, to everyone who can possibly know you. At that scale silencing free speech stops being a habit and becomes a strategy.
The colleague who does not mention the counselling psychologist she sees on Thursdays. The person whose parents know a partial version of their life and will go on knowing it. The quick read of a room, run far more often by some people than by others.
What that costs is mostly invisible from outside. It is the background task of keeping two versions of a life from touching: remembering which story you told to whom, editing the weekend before describing it, hearing an ordinary question and checking it for edges before answering. None of it looks like distress. All of it takes something.
Concealing a stigmatised identity — the evidence here concerns sexual orientation specifically — is modestly associated with internalising difficulties such as depression and anxiety (Pachankis et al., 2020).
The same work found the association running the other way for some outcomes, and the researchers are direct about why concealment resists a simply explainable cause. It can create the strain of hiding, shield a person from discrimination, and separate them from people who would understand, sometimes all at once.
So hiding is not automatically bad for you. It is that hiding costs something, that it often buys something too, and that the exchange rate depends entirely on who is in the room.
Nobody outside that room is well placed to price it.
When the Room Decides for You
So far the audience has widened but the calculation has stayed yours. Two pressures take it out of your hands.
The first is the read of the room. You sense you are outnumbered and the sentence you had ready simply does not arrive. This has a name — the spiral of silence — and it has proved durable enough to survive fifty years of testing (Matthes et al., 2018).
On average its effect on free speech is slight. Where it concentrates is the interesting part. Among family, friends and neighbours, on the issues that touch people most directly, it becomes markedly stronger, and it holds online just as firmly as offline.
The conversations that matter most are the ones people are most likely to sit out.
The second pressure is newer — surveillance. Being made aware of surveillance makes people less comfortable sharing an opinion, looking something up, and disclosing something personal (Festic et al., 2026).
What both pressures have in common is that they do not change what you think. They change your estimate of the audience.
What Free Speech Actually Changes
Look back along the widening. An audience of one. Then the person you live with. Then everyone who might find out. Then a room that has already decided, and an audience you cannot see. Different research at every stage, and the same sum running underneath all of it: what will it cost to be known here, and what does the silence buy me instead.
That sum never switches off. So the honest question is not whether people should speak more freely. It is what a legal/constitutional right does to the arithmetic.
A right to free speech does not release anyone from the calculation. It cannot. As long as other people can hear you, there is a sum to do, and research suggests it’s better to keep doing it. Some silences protect. Some disclosures backfire. The reflex to say everything is not obviously healthier than the reflex to say nothing.
What a right changes is what goes into the sum.
Where free speech is unprotected, the calculation carries consequences you did not choose and cannot bargain with: the job, the visa, the visit from someone official. Where it is protected, those move out of the sum, and what remains is the ordinary human freight of being known. Will they think less of me? Will this cost me an evening? Is this person safe?
That is still a hard calculation. It is a different kind of hard. But the distinction matters more than it sounds, because acting from your own volition rather than under a constraint somebody else sets is consistently associated with wellbeing (Van den Broeck et al., 2016).
None of this suggests that free speech guarantees happiness.
What it does suggest is that the gap between the opinion you give at work and the one you give at your kitchen table is neither trivial nor a moral failing. It is the same sum we’ve followed outward through every audience: the small, daily, mostly unnoticed decisions about what is safe to say and to whom.
Free speech, from the inside, is mostly made of those.
And if you need help figuring out ways to build and identify safe spaces, we’re always just a call away!
References
- Festic, N., Büchi, M., Latzer, M., & Odermatt, C. (2026). Dataveillance inhibits legitimate communication: Causal evidence for chilling effects. Journal of Communication, 76(4), 277–289. https://doi.org/10.1093/joc/jqag004
- Frattaroli, J. (2006). Experimental disclosure and its moderators: A meta-analysis. Psychological Bulletin, 132(6), 823–865. https://doi.org/10.1037/0033-2909.132.6.823
- Hu, T., Zhang, D., Wang, J., Mistry, R., Ran, G., & Wang, X. (2014). Relation between emotion regulation and mental health: A meta-analysis review. Psychological Reports, 114(2), 341–362. https://doi.org/10.2466/03.20.PR0.114k22w4
- Matthes, J., Knoll, J., & von Sikorski, C. (2018). The “spiral of silence” revisited: A meta-analysis on the relationship between perceptions of opinion support and political opinion expression. Communication Research, 45(1), 3–33. https://doi.org/10.1177/0093650217745429
- Nemeth, C. J., & Kwan, J. L. (1987). Minority influence, divergent thinking and detection of correct solutions. Journal of Applied Social Psychology, 17(9), 788–799. https://doi.org/10.1111/j.1559-1816.1987.tb00339.x
- Pachankis, J. E., Mahon, C. P., Jackson, S. D., Fetzner, B. K., & Bränström, R. (2020). Sexual orientation concealment and mental health: A conceptual and meta-analytic review. Psychological Bulletin, 146(10), 831–871. https://doi.org/10.1037/bul0000271
- Pintea, S., & Gatea, A. (2021). The relationship between self-silencing and depression: A meta-analysis. Journal of Social and Clinical Psychology, 40(4), 333–358. https://doi.org/10.1521/jscp.2021.40.4.333
- Sutton, A. (2020). Living the good life: A meta-analysis of authenticity, well-being and engagement. Personality and Individual Differences, 153, 109645. https://doi.org/10.1016/j.paid.2019.109645
- Van den Broeck, A., Ferris, D. L., Chang, C.-H., & Rosen, C. C. (2016). A review of self-determination theory’s basic psychological needs at work. Journal of Management, 42(5), 1195–1229. https://doi.org/10.1177/0149206316632058