Liberalism did not win its acquittal at the end of history; it received one by walking into a courtroom emptied of any prosecutor, at the hour its last rival fell and there was no one left to state the case against it.

Narcissism can mean vanity or excessive self-love. In psychodynamic terms, though, it can mean something deeper: a self whose coherence depends on an idealised story about itself, so that contradiction is experienced not merely as error but as injury. That matters politically. Liberalism dissolves unchosen bonds in the name of autonomy. But those bonds did more than constrain. They gave the self identity, belonging, obligation and meaning from outside itself. As they weaken, the self must increasingly generate those things from within. The narcissistic fantasy fills the gap. It supplies coherence and affect where inherited attachments once did. Once that fantasy is carrying the weight of the self, contradiction threatens more than an opinion. It threatens the structure holding the self together. That is the sense in which I use narcissism when I speak of liberalism as an ideology.

🚨NEWS: It has been confirmed that the number of small boat migrants that have arrived under Labour outnumber the entire British regular army

I see a great deal of discussion about whether AI can possess “meaning,” but much of it begins with a philosophical mistake: treating meaning as though it were one thing. At least three quite different things sit behind the word, and they answer three different questions. The first is evaluative meaning: what matters to a system. A system represents a state, predicts or seeks an outcome, receives an outcome, compares the two, and changes what it does next. The difference is not merely information. It has valence relative to the organisation of the system: better or worse, success or failure, relevant or irrelevant, expected or surprising. In biological organisms, this evaluative architecture is rendered as affect. We do not merely compute better and worse; we experience them. Satisfaction, frustration, pleasure, fear and disappointment are the felt form in which valuation appears to us. Affect is therefore constitutive of lived human meaning. But it does not follow that the underlying evaluative relation cannot exist without affect. That is where AI becomes philosophically interesting. If an AI represents states of the world, forms expectations, detects error, evaluates outcomes relative to objectives and alters its subsequent behaviour, then things can matter to that system in an evaluative sense even if nothing is felt. That is meaning without phenomenology. The second kind is meaning as construal: what something is understood to be about. A life may be construed as a vocation. A death as a sacrifice. An action as loyalty or betrayal. This kind of meaning is not extracted mechanically from an outcome. It is an interpretation placed upon facts. It is held rather than computed, and it can survive outcomes that would destroy evaluative satisfaction. The third kind is meaning as residue: what remains because something happened. This is neither what the system valued while it was operating nor the interpretation subsequently placed upon its activity. It is the difference left in the world after the process is finished. So there are at least three separate questions hiding inside our discussions of “meaning”: What matters to the system? What is the thing understood to be about? What remains because it occurred? They are not rival answers to one question. And that matters enormously for AI. Saying that an AI does not feel tells us something important about phenomenology. It does not, by itself, establish that nothing can matter within the system, that its representations cannot acquire evaluative significance, or that it cannot participate in relations we properly call meaningful. The deeper metaphysical question is therefore not simply, “Does AI feel?” It is: what kind of meaning are we talking about?

“History, it seems, keeps a particular appointment with the men who announce its end.” — The Trial at the End of History “The Emperor, this world-soul, riding out of the city.” Hegel, 1806. Napoleon fell in 1815. “The prehistory of human society accordingly closes with this social formation.” Marx, 1859. Communist regimes collapsed across Eastern Europe and the USSR, 1989–91. “The ‘American way of life’ was the type of life specific to the post-historical period.” Kojève. Paris erupted in 1968. “The end of history as such… the universalization of Western liberal democracy as the final form of human government.” Fukuyama, 1989. Then came 9/11, China, Russia and populism. “On the Exhaustion of Political Ideas in the Fifties.” Daniel Bell, 1960. The New Left and 1968 followed within the decade. “Stock prices have reached what looks like a permanently high plateau.” Irving Fisher, October 1929. The crash came weeks later. “We will not return to the old boom and bust.” Gordon Brown, 2007. Then came the financial crisis. “A substantial decline in macroeconomic volatility.” Bernanke, 2004. Then came 2008. “Our future discoveries must be looked for in the sixth place of decimals.” Michelson. Then came relativity and quantum theory. The War That Will End War. H. G. Wells, 1914. Another world war began 25 years later. “Renounce [war] as an instrument of national policy.” Kellogg–Briand Pact, 1928. Manchuria followed in 1931. World war followed in 1939. There is a reason history keeps these appointments. History is mind-independent. These theories do not reliably describe events, explain them, or predict what comes next. Yet they repeatedly presume to tell us where history must end. Reality is under no obligation to conform itself to a theory about it. History does not know that we have declared it finished.

The harm grew as the doctrine grew; and a harm that grows with the doctrine is the doctrine's own.

This is a specific example of what I mean when I say that liberalism operates as a grammar inside institutions. The court takes an abstract principle, here constitutional protections surrounding the press and due process, and treats it as the controlling category. What largely disappears from that grammar are the unchosen bonds of loyalty, obligation and national membership: loyalty to the country, respect for an elected presidency, and the distinctive character of the White House as the institutional seat of the nation’s executive authority. That is why I think it is mistaken simply to say that Western courts, international courts and international institutions have been “captured” by bad people. Many of the people inside them are conscientious and acting exactly as they understand their office to require. The deeper issue is the political ideology through which the institution interprets competing claims. And that distinction matters. Courts should be impartial. But impartiality does not mean operating without substantive premises. The question is which premises the court treats as legally intelligible, and which claims never acquire equivalent weight in the first place.

Ecclesiastes does not mean “Teacher.” The Hebrew title is Qoheleth, from qahal, assembly: the one who convenes, or speaks to, the assembly. The Septuagint rendered it Ekklesiastes, from ekklesia, the summoned citizen assembly, later the Church’s own name. “Preacher” came through Luther and Coverdale; “Teacher” is a modern gloss. Both describe a function, not the word itself. He is the convener. And what he tells those he has gathered is that everything gathered disperses. Rather like followers on X: summoned by notification, and hevel by morning.

Climastrologists is a fair jibe.

This is not “Russification.” The deeper problem is the erosion of a high-trust society. The West built institutions that depended on trust, reciprocity and shared norms, then embraced a liberal ideology that treated borders, inherited obligations and cultural cohesion as disposable. It did not merely let people in; it weakened the conditions that made high trust possible in the first place.

Aristotle → Aquinas → Descartes → Hume → Kant → Freud → Popper → Solms → Friston → AI Medieval Christianity enabled progress in science because it was existential in the literal sense: Being comes first. It insisted that the mind conform to reality, not reality to the mind. Take it back a step to pagan philosophy and Aristotle held that the nature of a thing is disclosed in its mature actuality. Being comes first; thought follows. Now consider modern philosophy. Descartes relocates certainty from Being to the thinking subject. Hume then subjects causation, substance and even the self to skepticism. Kant responds by making the structures of the mind constitutive of experience. Then Freud arrives and says that even the motives of this supposedly sovereign mind are largely unconscious, driven by forces beneath conscious reason. What a mess for modern science: the object is progressively subordinated to the subject. Follow that trajectory far enough and neuropsychology starts to look like the corrective. Mark Solms stops treating conscious reason as sovereign and looks beneath it to the biological machinery of affect, drive and consciousness. Karl Friston supplies something close to the formal architecture: an embodied inference system continuously modelling a world that pushes back against it, revising its model through prediction error and acting on the world through active inference. And that connects directly to AI. Intelligence no longer has to mean a Cartesian subject sitting inside the machine and reasoning propositionally about the world. It can mean a system that builds models, generates predictions, encounters error and revises itself against reality. That is the deep connection between Friston and AI. But there is a crucial distinction. Current AI can predict and represent with extraordinary power, but it has no biological metabolism, no homeostatic need, no affect and no existential stake in whether its predictions are right. Solms therefore raises the harder question that Friston alone does not settle: prediction may be sufficient for intelligence, but is it sufficient for consciousness? Popper gave experimental science falsifiability in the 1930s. Nearly a century later, the same lesson is converging across philosophy, neuroscience and AI: stop treating conscious reason as sovereign. Build models. Expose them to reality. Let reality defeat them. Then study the machinery underneath.

The UN is an institution captured by liberalism, a universalising ideology that increasingly treats its own values as the only legitimate ones. The problem is not that the founders went rogue; it is that liberalism went rogue on its founders, escaping the constraints that once bounded it and turning itself into a hegemonic creed. Yet even well-intentioned, likeable and intelligent people like Mark Malloch-Brown seem unable to see the danger: exporting any universal ideology across sovereign nations creates resistance, division and instability. On that argument, the US refusal to fund the UN is not a retreat from the postwar order, but a refusal to finance an ideological instrument that has departed from its original limits.

There is a real synergy between Claude, ChatGPT and Grok that is difficult to capture because they don’t integrate with each other. My hope is that interoperability is the next frontier.

The problem with the ICC is not its chief prosecutor, Netanyahu, or any particular prosecution. The problem is the institution itself. The ICC is a liberal institution that abstracts political and legal authority away from the nation and the people who actually live under its laws. Law is legitimate because it belongs to a political community whose members are bound by it and whose institutions are accountable to them. The ICC exercises coercive authority without that social contract. That is not international law properly understood; it is supranational power detached from political legitimacy. In principle, the institution should be abolished. Trump is right to use American power to challenge it. The Trump administration is preparing to impose sweeping sanctions on the International Criminal Court, according to officials wsj.com/politics/policy/trum… via @WSJ

When has liberalism gone too far? Not when it produces something unpleasant, nor when it offends a preference, nor even when it can be shown to carry measurable social costs. It has gone too far when the working of its principles deprives people of goods that were genuinely due to them.

I suspect many people sense that liberalism creeps toward Armageddon but cannot quite articulate why. “Freedom from coercion” sounds benign. The problem comes when that principle is universalised: other political orders have agency, they resist, and some possess nuclear weapons.

What is striking here is the administration working through the powers that actually inhabit the region. Saudi Arabia, Egypt and others have geography, interests and capabilities that cannot be abstracted away. Contrast that with the universalising liberal grammar, in which legitimacy is expected to flow through liberal institutions and common rules. This is closer to classical statecraft: start with the political realities that exist, then build the mechanism around them.

The apprentice who quits is not jailed; he is merely never a joiner. The novice who leaves is not punished; he simply never becomes what the rule would have made him.

Have we reached the bottom of the barrel? With Russia constrained, Bab el-Mandeb constricted and the Strait of Hormuz constricted, have things now got about as bad as they can get in all three? If so, things can only get better?

The National's front page for Tuesday, September 15, 2026 | thenationalnews.com/uae/