A companion to the empty leader — the same relation, seen from the other end. That post found the leader’s mind is the shape he reproduces. This one asks about the other end of the relation — and about a literature that spent seventy years trying to name the follower as a type, and ended up dissolving the type instead.
Status. The psychometric findings, the scale critiques and the model comparisons below are measured and published; the placing of them in this series’ framework is an inference, offered as one. It names only published scholars, and no one whose conduct is not on the record; it diagnoses no population, and it takes no political side — which, as §6 shows, is the whole difficulty, because the literature it draws on has not taken no side, and the series has to say so.
The empty leader argued that the
leader’s self is externalized — made of followers — and that the group
is structurally his G, his generator of self-content. Which
raises the obvious question the post did not answer:
What is a follower?
The intuition, and the one the whole culture reaches for, is that the follower is a kind of person: a type, a disposition, something you are. The research literature agrees, superficially — it even named the type. And then, over seventy years, it took the type apart.
That is the finding this post is about, and it is unusually useful, because the takedown is more interesting than the type ever was.
The tradition starts in 1950 with Adorno and colleagues’ The Authoritarian Personality, which proposed an authoritarian character structure to explain the rise of fascism. It was enormously influential — and, by the account of the field itself, it was also criticised at the time for bias and methodology, its theory falling “into disfavor” partly through its association with Freudian psychoanalysis.1 The counterweight is on the record too: independent studies confirmed part of what the book measured, and the criticism below is not that the authors invented a construct out of nothing.1
What matters here is not that it was wrong. It is how it failed, because the failure is one this series has met before.
The instrument was the F-scale — the “F” for fascism, or for the “pre-fascist” personality it claimed to describe. Its items were written to sound moderate — the authors’ own stated aim was to measure prejudice “without appearing to have this aim” — and they were all phrased in the same direction. The check for that is to reverse each item and see whether the scale still measures the same thing. It did not: Bass found the F-scale correlating −.20 with a rewritten, content-reversed version of itself, and Messick and Jackson’s reanalysis of the same data put it at +.20 — where, if phrasing played no role, either figure should have been −1.00.1
Sit with that number, because its size is the finding and its sign is not. Whether the true value is −.20 or +.20, neither is anywhere near −1.00, and the gap is the point: a substantial part of what the F-scale measured was not the construct but acquiescence, the tendency to agree. The instrument that would sort people into authoritarians and non-authoritarians was measuring, in part, the willingness to say yes to a questionnaire.
That is the DES-T’s 54% false-positive rate in the mechanism post, met again in a different literature: the instrument that names the person is the instrument that fails. And the failure has the same structure — a self-report scale, whose variance is partly the mode of reporting rather than the thing reported.
The type was not abandoned; it was rebuilt, and the rebuilding is what makes the literature usable. Bob Altemeyer developed the RWA scale (“right-wing authoritarianism”) to measure the construct rigorously, and defined it by three attitudes: a high degree of submission to established authorities; aggression directed at targets the authorities sanction; and high adherence to social conventions.2
The repair worked better than the thing it replaced, and it did not remove the class of problem. The RWA scale carries its own version of the critique: items near-tautological with the prejudice they are meant to predict, and a unidimensional scale standing in for a three-factor construct whose factors cannot be separately analysed.3 That is not an aside: it is the post’s own thesis holding at the next step along, and it is why the argument below rests on the activation finding rather than on the scale.
And then Altemeyer did the thing that matters most for this post, and it was deliberate:
Altemeyer calls right-wing authoritarians “authoritarian followers” — the point of the phrase being that he is not speaking of authoritarian leaders, which is the more commonly understood meaning of “authoritarian.” He reserves a different term for leaders: the “social dominator.”2
The follower and the leader are not one category. The finding is a measurement’s, not a moralist’s: the literature separated them, and Duckitt and Sibley’s dual-process model gave them different engines — the authoritarian-follower path runs on threat, the dominator path on dominance.4
Read that against the empty leader. That post argued the two share one dissociation pointing opposite ways; this literature says they are two different types with two different drivers. They are not in conflict — as §5 argues, they are the same structure seen from the survey side and the model side — but it is worth noting that the empirical field reached the distinction first, and from entirely different evidence.
Here is the finding that does the work, and it comes from Karen Stenner’s The Authoritarian Dynamic (2005).
Her argument is that prior work on authoritarianism suffered from tautological assertions and had failed to produce a theory that explained how authoritarian expression fluctuates with conditions. Her correction is a single structural move: the RWA index, she argues, “actually measures not fundamental predisposition to authoritarianism, but rather expressed authoritarian attitudes (i.e., manifest expressions of intolerance of difference)” — a dynamic response to external threat, not a static disposition.5
And the distinction that follows is the whole of the move: what the scale reads is expressed authoritarianism, while a latent predisposition is what threatening circumstances convert into that response.5
Read that in this series’ own vocabulary and it says something the series has been arguing from the first post:
A latent disposition, plus a condition, produces the behaviour.
Not a kind of person. A position the conditions activate. Which is the-environment’s thesis — the levers are positions an environment supplies or withholds, not possessions of a people — and it arrives here as a survey-measured claim rather than a model. Her cross-national evidence is the same shape: she reports authoritarian intolerance running low in Eastern Europe while socialist autocracies supplied normative reassurance, and rising when that reassurance collapsed.5 Take away the condition and the disposition does not express.
Schematic — an illustration of the argument in §4–§5, not a measurement. The shape is Stenner’s activation finding; the numbers are illustrative.
That is the most usable single sentence in this whole literature for this series, and it is worth stating as a rule:
The environment does not create the follower. It decides whether the predisposition is expressed.
Which is what the environment argues about price, the Costume about position, and the Breakthrough from the case record — reached from political psychology by someone studying electorates.
She also draws a line this series should keep, because it resolves a conflation that otherwise poisons the discussion: authoritarianism is resistance to difference across space; conservatism is resistance to difference across time.5 One is aversion to diversity now; the other is aversion to change over time. They are different axes, and they are routinely merged.
Now put the psychology beside the model, because they agree, and the agreement is the point.
The empty leader established the
leader’s structural need. A self made of followers needs followers
who can be supply — which is why the group’s demand for
sameness is not decoration: “a group of a hundred is a
hundredfold larger self-content generator than one person, and none of
it is his.” The follower is recruited, in that account, into a
role — the G-unit.
And what the psychology literature adds is that the role is what explains the person, not a trait. Three lines of evidence, all in the same direction:
The last one matters most, because it upsets the comfortable story. The follower is not the person who failed to resist. The follower is a participant — and the position is one that any of us can occupy, given the conditions, because the conditions are what call it out. That is the same conclusion the whole series reaches about the levers: they are positions, they are general, and the question is never “what kind of person” but “what does this environment supply, and to whom”.
So the answer to §1 is:
A follower is someone standing in a position that a condition has activated, in relation to an authority that defines the position.
Which is why the follower cannot be diagnosed and the leader can be described: the leader needs the position filled, and will build the conditions; the follower is produced by the conditions the leader builds. One is a design; the other is a response.
The series’ rule is to describe structures and to refuse to diagnose people — and this is the subject where that rule is hardest to keep, so the difficulty goes in the body rather than the notes.
The literature is not politically neutral, and this post will not pretend it is. The dominant construct is measured as right-wing authoritarianism; it was born as an attempt to explain fascism; and it is routinely deployed in political argument as a way of diagnosing one’s opponents. Some of the scale’s own critics make the stronger charge — that the bias is built into the items, so that the political commitment arrives with the measurement.3 Altemeyer himself noted the mirror case — a “left-wing authoritarian” who submits to authorities seeking to overthrow the establishment2 — and the construct has since been measured on the left rather than merely defined there, which makes it a live and contested literature rather than a tidy asymmetry.7 Stenner, for her part, explicitly notes that the dynamic appears elsewhere on the political spectrum, citing the Nation of Islam as an example.5
That matters because the honest use of this literature dissolves the political application rather than arming it. If authoritarian expression is a response to threat, then the way to produce it is to produce the threat — which means the interesting question is never “who are the authoritarians” but “who is making people feel that their world is coming apart.” The first question is a sorting exercise that tells you about the sorter, exactly as the label post says;8 the second is a question about conditions, and it has an answer.
And the series’ own rule applies to this post with unusual force, so it goes in full: this is not a diagnostic instrument. Nothing here identifies anyone as a follower, nothing here licenses treating anyone as one, and the standing limit holds — a description of an environment is not a finding about a person’s mind. The whole finding of §4 is that the position is conditional, which means every person in it is a person in a condition, not a member of a kind. Anyone who cites this to sort people into followers and leaders has inverted it, and reproduced the F-scale’s error at the level of argument.
Every post in this series has been about a position: the witch-and-the-debt’s unfalsifiable slot, the ladder’s rank, the certified frame’s borrowed warrant, the machine’s unarguable authority. The Follower is the same finding, one level up, about the person in the position — and it is the one that closes the loop with the empty leader.
| the leader | the follower | |
|---|---|---|
| what the model says | a self externalized into others | a self-content generator recruited as supply |
| what the psychology says | a dominator: dominance-driven | a follower: threat-driven |
| what defines the position | the need to be supplied | relation to an authority |
| what activates it | opportunity | a condition (perceived threat to the world) |
| can it be diagnosed? | no | no — and the instrument that tried failed the reversal check |
The last row is the post’s contribution, and it is what makes this the series’ clearest case rather than its most dangerous one. The tradition tried to name the follower as a type for seventy years and could not — not because the research was lazy, but because the thing it was looking for is not in the person. It is in the relation, and in the condition, and in the authority that defines both.
Which is the same conclusion every lever in this series has reached, and the reason the series never names a type: a position is what a person is standing in, and a person can leave one.
Marking: measured — the F-scale’s failure on the reversal check (Bass 1955; Messick & Jackson 1957; Christie et al. 1958) and the acquiescence critique (Couch & Keniston 1960); the RWA scale and its three components; the current critiques of the RWA construct (Dunwoody & Funke 2016; Duckitt et al. 2010); the measurement of the construct on the left (Conway et al. 2018); Stenner’s activation model, her Eastern-European evidence and her expressed/latent distinction; Altemeyer’s follower/dominator distinction; Duckitt & Sibley’s dual-process model. Documented — Haslam & Reicher on tyranny as a group-identity dynamic and on followers as something other than passive. Argument — that the follower is a position activated by a condition rather than a personality type; that this is the same finding the series reaches about every lever; that the political application of this literature dissolves under its own conditional finding rather than arming it; and the boundary in §6. The reading is this post’s contribution.
Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J., & Sanford, R. N. (1950). The Authoritarian Personality (New York: Harper & Brothers), https://en.wikipedia.org/wiki/The_Authoritarian_Personality — the F-scale (the “F” glossed as fascism in the source article’s lead, and as the “pre-fascist” personality in its own body text — the same article carries both, and the secondary literature generally follows); the items’ uniformly affirmative phrasing, on a scale whose stated aim was “to construct a scale that would measure prejudice without appearing to have this aim” (quoted in Dunwoody & Funke 2016, below); and the reversal failure, on which this post’s figure turns: Bass, B. M. (1955), “Authoritarianism or acquiescence?” The Journal of Abnormal and Social Psychology 51(3): 616–623, https://doi.org/10.1037/h0042890, whose rewritten, content-reversed “G scale” correlated −.20 with the F-scale as his data are summarized by Messick, S., & Jackson, D. N. (1957), “Authoritarianism or acquiescence in Bass’s data,” The Journal of Abnormal and Social Psychology 54(3): 424–426 (reanalysis of the same data yielding +.20), https://psycnet.apa.org/record/1959-05766-001. The sign is therefore disputed in the primary literature and the draft above rests only on the size of the gap from −1.00, which both values share. The more careful reversal — preserving each item’s rationale rather than mechanically negating it — is Christie, R., Havel, J., & Seidenberg, B. (1958), “Is the F scale irreversible?” The Journal of Abnormal and Social Psychology 56(2): 143–159, https://doi.org/10.1037/h0041279; and the study that made agreeing response set a construct in its own right is Couch, A., & Keniston, K. (1960), “Yeasayers and naysayers: Agreeing response set as a personality variable,” The Journal of Abnormal and Social Psychology 60(2): 151–174, https://doi.org/10.1037/h0040372. Reception: the book “was criticized at the time for bias and methodology,” while parts of what it measured were confirmed in independent studies; the claim that its theory “fell into disfavor” partly because it was associated with Freudian psychoanalysis is stated in the Right-wing authoritarianism article, https://en.wikipedia.org/wiki/Right-wing_authoritarianism. One attribution corrected here: the phrase “superficially moderate language” appears in the source as a description of the anti-Semitism scale’s items, not the F-scale’s; the F-scale point is sourced to the construction aim instead.↩︎
Altemeyer, B. (1981). Right-Wing Authoritarianism; and (1996) The Authoritarian Specter. The three components (submission to established authorities; sanctioned aggression; adherence to social conventions) are Altemeyer’s own definitions, at Altemeyer (1996) pp. 9–10 as quoted in Dunwoody & Funke (2016), below: aggression as intentional harm the aggressor believes “proper authority approves it or that it will help preserve such authority” (p. 10), submission as “a general acceptance of the statements and actions [of those in authority] and a general willingness to comply with their instructions without further inducement” (p. 9). https://en.wikipedia.org/wiki/Right-wing_authoritarianism On the naming: the post above renders Altemeyer’s usage in reported speech rather than as a verbatim quotation, because the phrasing “sometimes refers to right-wing authoritarians as ‘authoritarian followers’ … to emphasize that he is not speaking of authoritarian leaders” is the secondary article’s sentence, not Altemeyer’s; its sources are Altemeyer (1996) p. 6 and Altemeyer (2010) The Authoritarians, ch. 1 (“Who Are the Authoritarian Followers?”). The “social dominator” term for leaders, and the definition of the “left-wing authoritarian” as one who submits to authorities who want to overthrow the establishment, are that article’s attributions to Altemeyer (2010), ch. 1: “If these authorities are the established authorities in society, that’s right-wing authoritarianism. If one submits to authorities who want to overthrow the establishment, that’s left-wing authoritarianism, as I define things.”↩︎
On the current standing of the RWA construct — the critique the post’s §3 and §6 lean on: Dunwoody, P. T., & Funke, F. (2016), “The Aggression-Submission-Conventionalism Scale: Testing a new three factor measure of authoritarianism,” Journal of Social and Political Psychology 4(2): 571–600, https://doi.org/10.5964/jspp.v4i2.168 — a unidimensional scale used for a three-factor construct, items that cannot be assigned one-to-one to the factors they are meant to measure because they are deliberately double-barrelled, and items near-tautological with the prejudice the scale is used to predict (“a host of items that measure intolerance and prejudice can predict other measures of intolerance and prejudice,” citing Stenner 2005); the same authors charge the RWA scale with political bias, and their own replacement (the ASC) is offered as the fix. See also Duckitt, J., Bizumic, B., Krauss, S. W., & Heled, E. (2010), “A tripartite approach to right-wing authoritarianism: The authoritarianism–conservatism–traditionalism model,” Political Psychology 31(5): 685–715, https://doi.org/10.1111/j.1467-9221.2010.00781.x, a further attempt to rebuild the measure factor by factor.↩︎
Duckitt, J., & Sibley, C. G. (2009). “A dual-process motivational model of ideology, politics, and prejudice,” Psychological Inquiry 20(2–3): 98–109, https://doi.org/10.1080/10478400903028540 — the two paths: authoritarian-follower attitudes driven by threat (a dangerous-world view), social-dominance attitudes by dominance (a competitive-jungle view). For the dual-process model extended into a full account, see also Duckitt, J., & Sibley, C. G. (2010), “Personality, ideology, prejudice, and politics: A dual-process motivational model,” Journal of Personality 78(6): 1861–1894, https://doi.org/10.1111/j.1467-6494.2010.00672.x.↩︎
Stenner, K. (2005). The Authoritarian Dynamic (Cambridge: Cambridge University Press) — the critique of Altemeyer’s index as measuring “not fundamental predisposition to authoritarianism, but rather expressed authoritarian attitudes (i.e., manifest expressions of intolerance of difference),” quoted verbatim from ch. 4 (pp. 52–84), which is the expressed/latent distinction this post uses; the underlying claim that RWA is a dynamic response to external threat rather than a static disposition is the secondary literature’s summary of the book’s argument (https://en.wikipedia.org/wiki/Right-wing_authoritarianism) and is paraphrased above rather than quoted, because no verbatim sentence in that wording could be located. The Eastern-European finding — authoritarian intolerance low while socialist autocracies supplied normative reassurance, rising after the collapse — is at pp. 316–319; the Nation of Islam as an example of the dynamic elsewhere on the spectrum, at pp. 141–142 (https://en.wikipedia.org/wiki/Karen_Stenner); and the analytic distinction between authoritarianism (resistance to difference across space) and conservatism (resistance to difference across time) is developed at pp. 13–27 and again in Stenner, K. (2009), “Three kinds of ‘conservatism’,” Psychological Inquiry 20(2–3): 142–159, https://doi.org/10.1080/10478400903028615.↩︎
Haslam, S. A., & Reicher, S. D. (2007). “Beyond the banality of evil: Three dynamics of an interactionist social psychology of tyranny,” Personality and Social Psychology Bulletin 33(5): 615–622, https://doi.org/10.1177/0146167206298570 — on tyranny as a group-identity dynamic, and on the point the post uses: the presence of leadership “should not be equated with the passivity of followers.” For the positive half of the claim — leaders and followers as collaborative agents producing the leadership between them rather than followers as its recipients — see Reicher, S. D., Haslam, S. A., & Hopkins, N. (2005), “Social identity and the dynamics of leadership: Leaders and followers as collaborative agents in the transformation of social reality,” The Leadership Quarterly 16(4): 547–568, https://doi.org/10.1016/j.leaqua.2005.06.007, and the book, Haslam, S. A., Reicher, S. D., & Platow, M. J. (2020), The New Psychology of Leadership: Identity, Influence and Power, 2nd ed. (London: Routledge), https://doi.org/10.4324/9781351108232. The phrase “active participants” is the paraphrase used above for that collaborative-agency position; it is not a quotation from any one of these texts.↩︎
On measuring the construct on the left rather than only defining it: Conway, L. G. III, Houck, S. C., Gornick, L. J., & Repke, M. A. (2018), “Finding the Loch Ness Monster: Left-Wing Authoritarianism in the United States,” Political Psychology 39(5): 1049–1067, https://doi.org/10.1111/pops.12470 — left-wing authoritarianism as a measurable construct in the same family. The literature is contested, and this post does not adjudicate it: the counter-argument, that the left-wing construct is being manufactured rather than found, is argued in the same journal’s pages (see Conway et al. 2023, “Is the myth of left-wing authoritarianism itself a myth?” Frontiers in Psychology 13: 1041391, https://doi.org/10.3389/fpsyg.2022.1041391). Either way, the claim in §6 is the weaker one: the construct is not the exclusive property of one side.↩︎
The label post on the difference between describing a structure and sorting people into a verdict, and the standing non-use clause.↩︎