You had your own view in the meeting, watched the way it was going, and ended without saying it. You saw the positioning your peers were using and shifted your own towards it. Every time you consult someone, the direction moves. Looking back, there are hardly any occasions you remember deciding for yourself.
Being easily swayed is not only a problem of the strength of will. Falling in with other people’s choices is, in most cases, a legitimate use of material for estimating the environment. Nobody is giving an order, and nobody is yielding to pressure. Where a situation is ambiguous, what other people are doing is genuinely useful information, and there are many situations in which not referring to it would be the unreasonable move. Training yourself not to give in to pressure therefore reaches only one of the two routes by which choices align.
The subject is not whether to refer but how much weight to place on what the reference tells you. The information carried by many people making the same choice varies by orders of magnitude with the situation. Where each person judged independently and arrived at the same place, a great many judgements are stacked up in it. Where they aligned by watching one another, very little is. From outside, the shape is the same — everyone is doing the same thing — so appearance alone does not settle the weighting.
The scope narrows to a single level. How an aligned state is held in place on the side of society, and what departing from it actually involves, sit outside this article. That mechanism is handled by How social common sense is made. What is examined here is not how a norm is kept up at the level of the group but what is happening inside the judgement of one person who is present.
This way that weighting breaks down has a formalisation on the economics side. What the formalisation shows is that the size of a majority and the number of judgements inside it can, past a certain point, move separately. A going rate or an established practice looks like stronger evidence than it is because that separation is invisible from outside.
The state called by that single word divides into three, and the advertised responses reach only two of them. Referring to the majority is, in itself, a legitimate use of material for estimating the environment. But alignment runs by two routes, and one of them carries no sense of being pushed, so the act of referring leaves no memory. And as a majority grows larger, the number of judgements inside it stops growing. For someone running a business alone, this alignment concentrates in three decisions: the positioning, the price, and the form of the business.
📖 Contents
- Under “Being Swayed” Sit Three Things of Different Kinds
- The Five Offered as Responses Act on Missing Material and Suppressed Expression
- Falling In With the Majority Can Be a Legitimate Use of Material
- Alignment Occurs Even Where the Answer Is Plainly Wrong — but Not for Everyone
- The Phrase “Peer Pressure” Names Only One of the Two
- “Reading the Room” Is a Skill of Reading, Not Subordination
- The Larger the Majority Grows, the Less Information It Contains
- Alignment Surfaces in Positioning, Price and the Form of the Business
- Where a Change in How Alignment Happens Shows Up
Under “Being Swayed” Sit Three Things of Different Kinds
One word covers three states that differ in kind. Being swayed is not one thing.
The first is one where you hold no judgement at all. You have no view yet, so you take the option in front of you. This is less being swayed than receiving a default.
The second is one where the judgement exists and does not get out. Inside, a different answer has been reached, and without voicing it you fall in with the majority. The judgement is retained, so a misgiving remains afterwards.
The third is one where the judgement itself moves. On contact with someone else’s view, your own answer changes. Here no misgiving remains, because the changed answer is what gets remembered as yours.
The three arrive at one appearance. Each produces the result of having gone along with the surroundings. What happens inside, though, differs. The first is a shortfall of material, the second a suppression of expression, the third a rewriting of judgement.
The view that they amount to the same thing, since your opinion did not prevail in any of them, does not hold. The remedies that work on the second — practising how to phrase things, building an occasion to speak — do not reach the third at all. In the third, the different answer that needed conveying has already gone. Conversely, the remedy that works on the third — writing out your criterion in advance — has nothing to write out when applied to the first.
Of the three, only the third cannot be noticed. Once judgement has been rewritten, the answer from before the rewriting does not remain. Someone who can say of themselves that they are easily swayed is usually pointing at the second. Whether the third applies cannot be assessed without keeping a record.
The three can shift in sequence. As the count of suppressed expressions builds, the reason to hold on to an unvoiced answer thins, and in time the answer stops arriving first. Past a long run of the second form, the third can follow. It has to be handled while it can still be noticed, and once it cannot, it becomes indistinguishable from outside and from within. Nothing announces that the shift has occurred.
The single phrase “easily swayed by nature” takes all three on together and puts the cause on the side of strength of will. Treated as a matter of will, the remedy becomes correcting the person’s posture. And the main route by which choices align has nothing to do with strength of will. The experience of applying remedies that do not work raises, over time, the plausibility of the explanation that the will is weak.
The Five Offered as Responses Act on Missing Material and Suppressed Expression
The responses advertised for being easily swayed sort into five.
Have the courage to voice your own view. Increase your knowledge and information so you hold a criterion. Establish self-esteem so you are not shaken by other people’s views. Do not fear being disliked. Have an axis of your own.
Of the five, the second is accurate. The observation that your view does not take shape while your criterion is vague catches the first form, the missing material, correctly. Strengthening the will alone, with material and criterion missing, produces nothing to come out.
The remaining four are designed for the second form, where the judgement exists and does not get out. Courage, not fearing, an unshakeable self-assessment, an axis of your own. Each presupposes that an answer already exists inside.
What the five reach is the shortfall of material and the suppression of expression. As a remedy they mesh. What they do not reach is the route by which judgement itself gets rewritten.
That high self-esteem would also stop judgement from being rewritten does not follow. The work of referring to other people’s choices runs independently of how you assess yourself. Where a situation is ambiguous, what others are doing is genuinely useful information. The faculty that uses useful information does not switch off when self-assessment rises.
The five are not being called void here. Where expression is being suppressed, they work. The claim is only that their reach is fixed.
Setting “not shaken by other people’s views” as the goal, meanwhile, produces a side loss. Other people’s choices carry information about the environment. Block all of it and you become unshakeable at the price of one fewer route for detecting a change in the environment. Independence and sensitivity cannot both be raised by the same move.
And the “have courage” family of remedies misplaces the cost. The costs that come with departing — the load of explaining, friction in relationships, work drying up — are real, so what settles the choice is not a shortfall of courage but an estimate of cost. Since the cost is real, a remedy that ignores it can only be carried out by someone in a position to pay it.
Beyond this, the remedy treats alignment itself as an undesirable state. A temperament that is cooperative and reads the room well is often handled in these guides with the qualifier “a strength, admittedly.” Affirmed in the form of a qualifier, holding that temperament is positioned as something to be corrected. Holding something to be corrected moves you closer to a state where the assessment is held outside.
Falling In With the Majority Can Be a Legitimate Use of Material
Referring to other people’s choices is not a failure. It is a low-cost method for estimating the environment.
Judgement requires information about the situation. Is this route safe, is this price reasonable, will this positioning land. Establishing each of these from scratch has a cost. Someone already placed in the same situation has made a choice, and that choice carries the result of the checking cost they paid.
Put another way, many people making a given choice is evidence about the probability that the choice is sound. Weak evidence, and evidence nonetheless.
Suppose that in a field you are entering for the first time, the experienced people all avoid a particular method. You do not know why they avoid it. Avoiding it without knowing the reason is, here, a sound judgement.
The trouble arises when this reference gets used as a criterion rather than as material. Used as material, it gets set against other material and assessed against your own criterion. Used as a criterion, the majority’s choice is itself the answer.
The worry that referring is dangerous because the majority is not always right lands slightly off the mark. What is dangerous is not the reference but the weight placed on it. Where independent judgements happened to converge, a great deal of deliberation sits behind them. Where the convergence came from watching one another, little does.
Whether an alignment arose independently or by mutual observation is hard to tell from outside. The clue is whether the reasons that led to the choice vary from person to person. Where the reasons are aligned too, mutual reference is the likely origin.
The information also varies with whether the party you refer to sits under the same constraints. The choices of someone with a different scale of capital and a different number of clients are not evidence about your situation. Being in the same trade does not mean being under the same constraints.
And the more legitimate the reference is as material, the more readily it is promoted to a criterion. A useful source of information gets used repeatedly, and what gets used repeatedly has its checking step dropped. Usefulness itself drives the move from material to criterion.
Whether the promotion has occurred can be settled with one question. Can you explain the choice with what the majority is doing taken out? If you can, you are using it as material. If nothing remains once it is taken out, it has already become a criterion. And a reference that has become a criterion does not show up as wrong even when it is, because the result matches the majority, so a departing outcome is never observed.
Alignment Occurs Even Where the Answer Is Plainly Wrong — but Not for Everyone
There is a famous series of experiments measuring the strength of alignment: the study published in 1956 by the psychologist Solomon Asch (Asch, 1956, Psychological Monographs, 70(9), 1–70). Participants are shown one line and asked which of three comparison lines matches it in length. The answer is plain, and nobody answering alone gets it wrong. The other participants present, however, are all confederates, and they state a predetermined wrong answer without hesitation.
The result was that participants went along with the erroneous majority on roughly one third of the trials calling for a judgement.
That figure is often cited as grounds for the conclusion that people are easily swayed. The same data has another side.
Roughly one quarter of the participants never once went along with the majority. And across the whole, the correct answer was maintained on two thirds of the trials. The force of alignment is strong, and yet unaligned answers were the more common. That is the actual result.
That a third aligns on something as simple as line length, and that more would therefore align on complex judgements, does not follow. Line length is a case where, precisely because the answer is plain, alignment cannot be explained by having got it wrong. In business judgements, where the answer is ambiguous, other people’s choices genuinely do hold information, so the occasions on which referring is correct increase. A rising rate of alignment does not by itself indicate error.
This experiment is a result under specific conditions, and the numbers move sharply with the size of the group, anonymity, and whether even one person dissents. It cannot be read as a general rule that people are swayed a third of the time. What can be read is that alignment can occur even against a plain fact.
And what is measured is the answer spoken aloud, not inner conviction. Participants who did not align were not necessarily free of doubt, and those who did align did not necessarily doubt their own eyes. Agreement and disagreement seen from outside do not distinguish what happened inside. That point runs straight into how the two routes are told apart.
This experiment is also often told inside a story in which conformity is weakness. The telling supplies, to people with an experience of having aligned, an explanation in terms of a defect of temperament. The fact that a quarter never aligned gets used as the contrast that marks the fault of those who did. What the experiment showed was the effect of conditions, not the distribution of temperaments.
The Phrase “Peer Pressure” Names Only One of the Two
The routes by which alignment occurs divide in two. The distinction was set out by the psychologists Morton Deutsch and Harold Gerard in a 1955 study (Deutsch & Gerard, 1955, The Journal of Abnormal and Social Psychology, 51(3), 629–636).
The first is informational influence. Where a situation is ambiguous, other people’s judgements are used as evidence about reality. On this route, your own answer genuinely updates.
The second is normative influence. To avoid falling outside the group’s expectations, you voice an answer at odds with what you think. On this route, your own answer has not changed.
What the phrase “peer pressure” names is the second. As the word pressure indicates, it comes with the sensation of being pushed from outside. The first carries no such sensation. You are only using other people’s choices as evidence, so from inside it feels like having been persuaded on your own.
Training yourself not to give in to pressure therefore works only on the second. And the one that is hard to notice is the first.
Suppose you see that every peer around you offers a free initial consultation, and you set one up too. If you are aware that you did it because your peers do, that is the second. If it feels like you thought about it and decided it was necessary, the first may have been at work. What can be told apart is whether you examined the question of whether you would have reached the same conclusion had nobody else been doing it.
The response that using it as information is a correct judgement is right in some cases. The trouble is that on the first route, the act of referring leaves no memory. It settles as your own judgement, so it never becomes something whose premise you go back and question. The second, being unpleasant, stays in memory and becomes an occasion for questioning.
The two routes operate simultaneously in the same situation. Separating them is an analytical move, and in an actual judgement, how much each contributed cannot be identified. What is available is refusing to use the inference “I felt no pressure, so it was my own judgement.”
Learning about the first route makes all of your judgements look suspect. That they look suspect is accurate, and suspecting everything stops judgement. The realistic move is to limit the application to the small number of places that govern the judgements downstream — price and business form, for instance, which are hard to move later.
“Reading the Room” Is a Skill of Reading, Not Subordination
There is another word folding two things together: reading the room.
The phrase in fact points at two operations. The first is reading the premises and expectations running through a setting. The second is complying with what you read.
The first is a skill. The ability to detect premises that have not been put into words is useful in negotiation and in design. Not being able to read is simply a state with fewer options.
The second is a judgement. Comply with what you read, or deliberately depart from it. There is room for choice here.
Because the two are folded into one word, the prescription “stop reading the room” gets implemented as a stopping of the reading. Stopping the reading does not stop the complying. If anything, you now comply without seeing the premises, so the room for judgement shrinks.
Suppose in a meeting you can tell that the other party is steering clear of the budget. That is reading. Ending without touching the budget is complying; inferring why it is being avoided and touching it deliberately is judging. With the reading stopped, the conversation proceeds without your noticing that anything is being avoided at all.
The point that being able to read is exactly what makes you comply is well taken. When the expectation is visible, so is the cost of departing from it. Being able to see a cost, though, does not mean paying it. Where you departed without seeing it, you paid without noticing.
The accuracy of reading has limits. The “room” you read is an inference, and it can differ from the expectation actually present. Treating an inference as a fact means complying with something nobody asked for.
Having separated reading from complying, deliberate compliance is a legitimate choice. Many negotiations advance by meeting the premises of the setting, and departing is not always better. The value of the distinction lies not in raising the number of departures but in a compliance that stays in memory as a choice.
And the higher the skill of reading, the heavier the load of departing. What will follow from departing can be predicted in detail. The finer the prediction, the larger the estimated cost of departing. That people who read well comply more readily is not a matter of ability but of the precision of the estimate.
This relation has a workable side. A fine prediction also means the size of the cost is known. Where the size is known, the occasions you can pay for separate from the ones you cannot. Departing without that separation is a wager; departing having made it is an allocation. Stop the reading and the allocation becomes impossible.
The Larger the Majority Grows, the Less Information It Contains
The majority’s choice is weak evidence. Its strength as evidence can move in the opposite direction from the size of the majority.
The reversal was formalised in 1992. Bikhchandani, Hirshleifer and Welch treated what happens among people choosing in sequence (Bikhchandani, Hirshleifer & Welch, 1992, Journal of Political Economy, 100(5), 992–1026). Each person holds one private clue, and can also observe the choices made by those who went before.
What occurs is that once the preceding choices reach a certain number, later choosers stop using their own clue. Their own clue is weak, whereas the earlier choices appear to have several judgements inside them. And this is not an error but the result of a rational calculation.
The trouble lies in what happens afterwards. A choice made by someone who did not use their own clue contains no new information. From outside, though, it looks as if one more person has agreed. The next person sees that enlarged majority and runs the same calculation.
From a certain point on, in other words, the growth adds no information at all. The majority keeps getting larger while the number of judgements inside it does not. The authors called this state an informational cascade.
The consequences divide in two. The first is that size does not indicate strength of evidence. Everyone doing the same thing might be the result of everyone having checked, or of only the first few having checked. From outside, the two look identical.
The second is that this state is brittle. With few judgements stacked up, a single new piece of public information can move the whole thing to the other side. The larger an alignment looks, the faster it collapses when it does. Some cases of a long-standing practice disappearing all at once in a given year have this shape.
And the mechanism works most strongly on people who have just entered. Someone new holds the fewest clues about the environment, and the weaker the clue, the more rational it becomes to take what the earlier entrants chose. Choices made just after entry then persist for a long time as premises for later judgements.
Questioning an alignment and departing from it, meanwhile, are separate. Questioning it and concluding that continuing is sound also happens. Held as a judgement, it can be moved when conditions change. Followed without ever being questioned, it cannot. The checking is done once and does not need repeating.
Alignment Surfaces in Positioning, Price and the Form of the Business
Hold a business alone, and alignment concentrates in three decisions.
First, in the content of your positioning. You see the topics your peers are publishing on and move towards them. What you get for it is comparison in the same place as your peers. Enter the field of comparison and the axes on which you differ narrow to speed and price. This structure is handled in Choosing to stop analysing competitors.
Second, in price. You refer to the going rate to set your own figure. The going rate is the result of many choices aligning, and it is where the mechanism by which a larger number carries less judgement works most strongly. Unless each party set theirs independently, the number of judgements inside it is smaller than it appears. The going rate nonetheless operates as the strongest reference point.
Third, in the form of the business. Where those around you work by selling time, that form itself never becomes a subject of examination. Everyone doing it functions quietly as the reason it should be done. Questioning the form is handled in Why raising your hourly rate does not make you free.
The overlap is a single point: the act of referring left no memory. What you do not remember referring to does not become something you can stop referring to.
The worry that differing from your peers will simply make you undiscoverable is accurate. Shared vocabulary and shared formats genuinely function as conditions for being found. What deserves separating is the places you align to be found from the places you depart to be chosen. Align everything and you are found, with no remaining reason to be chosen.
The three cannot all be departed from at once. They support one another, so removing one breaks the coherence of the rest. Taking them one at a time, starting with whichever returns a result fastest, is the practical order.
Of the three, the reference bites hardest on price, because positioning and form are hard to compare when you depart from them, whereas prices line up in the same unit. Anything that lines up invites promotion from reference to criterion. Put the other way, holding an axis that does not line up is itself an operation that lowers the reference’s influence.
And staying aligned keeps the short-run response stable. As long as you are on the same topics, in the same format, in the same price band, you are in a place where demand has already been confirmed. Depart and the volume of response drops once. Whether you can survive the interval of the drop is the practical constraint.
Where a Change in How Alignment Happens Shows Up
Whether the route of alignment has changed shows in three places.
First, look at whether the memory of having referred to other people’s choices remains.
The goal is not to stop referring. It is to remember having referred. If you can say “I saw my peers doing this, and decided that,” the judgement can be moved later. A reference that leaves no memory is the indicator of the form in which judgement has been rewritten.
Second, look at whether you can state your answer for the case where the majority had chosen otherwise.
For a choice you currently hold, ask yourself whether you would have reached the same conclusion had nobody around you been doing it. If the answer changes, what settles that choice is the surroundings. If it does not change, the criterion sits on your side.
Third, look at whether the reasons behind the alignment vary.
Ask the people you refer to, as far as you can, why they made that choice. Where the reasons vary, the alignment is an overlap of independent judgements and carries evidential weight. Where the reasons are aligned too, it is the product of mutual reference and carries almost none.
What the assessment uses is the memory of referring, the answer under changed conditions, and the origin of the alignment. The faculty of referring itself does not switch off, so whether it has switched off is not material.
You cannot recall an occasion where you decided for yourself. That observation is accurate, and it does not indicate a weak will. The memory is thin because the route that operates as information carries no awareness with it. With no sensation of being pushed, the event of having referred is never recorded. The other route, being unpleasant, does stay, so what you can recall leans towards it. The memories that remain do not represent the distribution of what happened.
Remedies that train assertiveness reach situations where you hold an answer and cannot voice it. The view set out here treats situations where the answer itself has already aligned, and there is no different answer left to voice. Neither works outside the route it reaches, and the order is to look first at which route was at work.
The faculty of referring will not go away. It lowers the cost of estimating the environment, so there is no reason not to use it. What is within reach is writing down the act of referring in a line at the time. What has been written down can be treated as a judgement.
How convergence is held in place on society’s side, and what stepping one pace outside it consists of, is carried squarely by How society’s common sense gets made. The cognitive side of a reference settling into judgement is handled in Reading unable to change through cognitive science, and the work of holding the order of what matters on your own side in How to build a worldview.
Three articles sit alongside this one. The mechanism by which the entrance to judgement disappears is handled in Having stopped questioning things is not idleness, the line between material that may be borrowed and judgement that may not in What thinking for yourself actually contains, and how what went unsaid is booked by the room in The structure of a room where opinions cannot be voiced.






