Lent Reader 2025

Reader’s Feast | Certainty in the Law: certainty generally is an illusion, and repose is not the destiny of man (O.W Holmes)

This is a transcript of Master Jonathan Mance’s Reading at Reader’s Feast on Wednesday 19 February 2025.

The language of judicial decision is mainly the language of logic. And the logical method and form flatter that longing for certainty and for repose which is in every human mind. But certainty generally is illusion, and repose is not the destiny of man. Oliver Wendell Holmes The Path of the Law 10 Harvard Law Review 457 (1897).

The film Conclave won big at the Baftas on Monday, though some of you may, like me, have preferred the Seed of the Sacred Fig and others, or at least their Grandchildren, Wallace and Gromit’s Vengeance Most Fowl. But whatever your preferences, you will surely remember the advice given by the Dean (Ralph Fiennes) to his fellow Cardinals about to assemble in Conclave to elect a new Pope: 

Beware of certainty – the great enemy of unity, the deadly enemy of tolerance.

This, in my view, is also very good advice for lawyers. Seeing all sides of an issue is as valuable in law as in Conclave. But both Robert Harris, the author of Conclave, and Oliver Wendell Holmes, for 30 years an Associate Justice of the US Supreme Court, were, of course, writing extra-judicially. You will find little judicial writing praising the law’s uncertainty, or even its flexibility. We associate the law with clarity and predictability. It is no surprise that the first element identified by Lord Bingham in his book The Rule of Law (Allen Lane, 2010) published just before his untimely death in 2010, is that:

The law must be accessible and so far as possible intelligible, clear and predictable.

Predictability makes the Rule of Law distinct from the rule of autocrats or the practice of demagogues. They thrive on the disruption and fear generated by unpredictability, a concept that has become of sharply increased usage and relevance even since I prepared this talk.  

Clarity and predictability in the law go, in contrast, hand in hand with the principle that the same law should apply to all alike – a fundamental difference between jurisdictions where a true rule of law exists and countries where the law is subservient to the diktats of a ruling party, elite or individual. 

No surprise, therefore, that certainty and predictability are the virtues generally emphasised in judicial writing. In Golden Straight Corporation v Nippon YKK Lord Bingham himself started his judgment by saying that ‘the quality of certainty [is] a traditional strength and major selling point of English commercial law.’  Lord Mansfield, his 17th Century predecessor, said much the same in Vallejo v Wheeler (1774) 1 Cowp 143, 153:

‘in all mercantile transactions the great object should be certainty. And therefore it is of more consequence that a rule should be certain than whether the rule is established one way or the other: because speculators in trade then know which ground to go upon’

Certainty is often assimilated with precedent and immutability. Looking back over history, there are two strands. For long periods, the dominant strand was a myth of Saxon continuity. According to this, the law had always, or at least since Saxon times, existed, to be revealed and declared by judges as occasion arose in their judgments and to be learned by lawyers through long experience. That is the declaratory theory of law. Sir Edward Coke CJ in his battles with James I seems to have deployed this theory in the Case of Prohibitions– whether as a genuine believer or to resist James I’s intervention, I do not know. The case arose when James I tried to support a contention advanced by the Archbishop of Canterbury, that the judges were simply the King’s delegates, and that the King could ‘take what causes he shall please’ to determine them himself. In answering that contention, Coke wisely agreed with James I’s claim that he, as the King, had ‘reason like the judges, excellent science, and great endowments of nature.’ This was tactful in relation to a King dubbed by some ‘the wisest fool in Christendom.’ But, Coke went on, reason was not enough. The law was different. Cases were not: 

‘to be decided by naturall reason but by the artificiall reason and judgment of Law, …. which requires long study and experience, before that a man can attain to the cognizance of it.’

Coke wrote this himself, after the event. A bystander actually records that, as Coke said, the King raised his fist, at which Coke fell on all fours to beg for mercy. But Coke did not withdraw his ruling. In the Middle Temple, we remember Coke with displeasure for his bullying prosecution, or persecution, of Sir Walter Raleigh, a Middle Templar, for treason. But The Case of Proclamations shows that Coke was at least, in Stephen Sedley’s words, ‘an exception to the maxim …. that those who trample on the helpless are disposed to cringe to the powerful.’ Coke’s refusal of the King’s attempt at direct adjudication marked an important first step in judicial independence.

Later during the 17th Century, the declaratory theory found support in the notoriously misogynistic figure of Sir Matthew Hale (I note in parenthesis that Hale was repeatedly cited by Justice Alito in Dobbs v Jackson Women’s Health Organization, when overruling Roe v Wade; originalism must be a close cousin of the declaratory theory.) The declaratory theory also found support from Sir William Blackstone in his commentaries. As late as 1892 we find its traces in a remark by Lord Esher MR, who said in Willis & Co v Baddeley:

‘This is not a case, as has been suggested, of what is sometimes called judge-made law. There is, in fact, no such thing as judge-made law, for the judges do not make the law, though they frequently have to apply existing law to circumstances as to which it has not been authoritatively laid down that such law is applicable.’

This limited view of the judicial role in the common law went hand in hand with a limited view ascribed to the judiciary in relation to statutes. Parliament was to enact, and the judges were to apply, as under Article V of Napolean’s Code civile, judges were not to gloss or elaborate. 

When I was at Oxford, Lord Denning was a hero, and I could see him from my room saying goodbye to Professor Sir Arthur Goodhart, editor of the Law Quarterly Review, after a weekend spent discussing law in the Master’s Lodgings. But the law at the highest level remained frozen by brutal reiterations of the declaratory approach. Viscount Simonds’ unremitting remarks aimed at Lord Denning in Midland Silicones Ltd v Scruttons Ltd. still chill the blood:

‘Learned Counsel for the Respondents claimed that this [that is, the principle that in no circumstances could a third party benefit from the protection of a contract to which he was not party] was the orthodox view and asked your Lordships to reject any proposition that impinged upon it. To that invitation I readily respond. For to me heterodoxy, or, as some might say, heresy, is not the more attractive because it is dignified by the name of reform. Nor will I easily be led by an undiscerning zeal for some abstract kind of justice to ignore our first duty, which is to administer justice according to law, the law which is established for us by Act of Parliament or the binding authority of precedent. The law is developed by the application of old principles to new circumstances. Therein lies its genius. Its reform by the abrogation of those principles is the task not of the courts of law but of Parliament.’ 

This was uttered in 1962, perhaps as a sort of Parthian shot, just before Lord Simond’s retirement. 

We can contrast this idea of Saxon immutability with Lord Mansfield’s approach, at least in private law matters. He was in public law a conservative and seriously over-involved, at least to modern eyes, in affairs of government as a confident of George III in the intransigence which led to the American Civil War. Lord Mansfield understood the importance of certainty, but it needed to be well thought through and justified. Where this was not the case, he was no friend of either precedent or immutability. He demonstrated his own ability to see both sides of the coin in words in Alderson v Temple, contrasting sharply with my earlier quotation from his judgment in Vallejo v Wheeler: 

‘The most desirable object in all judicial determinations, especially in mercantile ones (which ought to be determined upon natural justice, and not upon the niceties of the law) is to do justice.’ 

Not surprisingly, Mansfield, the legal innovator, became the target of those who supported the Saxon myth. One of them, in an outburst of early Euroscepticism, wrote in the anonymous Letters of Junius, in 1770:

‘In contempt of the common law of England, you have made it your study to introduce into the court where you preside, maxims of jurisprudence unknown to Englishmen. The Roman code, the law of nations, and the opinions of foreign civilians, are your perpetual theme; but whoever heard you mention Magna Carta or the Bill of Rights with approbation or respect? By such treacherous arts, the noble simplicity and spirit of our Saxon laws were first corrupted.’

Holmes was in the same camp as Mansfield. His Harvard article, The Path of the Law, written in 1897, is remarkable for treating the common law as a judge-made system, not to be seen as based on science, history, religion or even morals, but on considerations of public good – even if, he went on, ‘I think that the judges themselves have failed adequately to recognize their duty of weighing considerations of social advantage.’ There were, he continued, ‘battle grounds where the means do not exist for the determinations that shall be good for all time, and where the decision can do no more than embody the preference of a given body in a given time and place.’ Carrying this view of the judicial role to an extreme, he concluded that ‘the prophecies of what the courts will do in fact, and nothing more pretentious, are what I mean by the law.’

Holmes’ realism may have been slow to cross the Atlantic. It would have horrified Lord Simonds. But ten years after Simonds’ excoriating blast at Lord Denning, his successor as senior Law Lord, Lord Reid, could write in The Judge as Law Maker:

‘There was a time when it was thought indecent to suggest that judges make law – they only declare it …but we do not believe in fairy tales any more.’

History had moved on. Consistent with this new realism, the House of Lords also announced by Practice Direction (Judicial Precedent) [1966] 1 WLR 1234 (still applicable in the Supreme Court) that ‘it would depart from a previous decision when it appeared right to do so.’ It is now also a commonplace, even if sometimes unwelcome to politicians, that there is more to statutory interpretation than a literal reading of the text. In ex p Simms, Lord Hoffmann explained the principle of legality: 

‘Fundamental rights cannot be overridden by general or ambiguous words. This is because there is too great a risk that the full implications of their unqualified meaning may have passed unnoticed in the democratic process. In the absence of express language or necessary implication to the contrary, the courts therefore presume that even the most general words were intended to be subject to the basic rights of the individual. In this way the courts of the United Kingdom, though acknowledging the sovereignty of Parliament, apply principles of constitutionality little different from those which exist in countries where the power of the legislature is expressly limited by a constitutional document.’

The same principle of legality is found, reinforced, in section 3 of the Human Rights Act. 

We probably all agree that, on a heavenly scale, justice trumps certainty, and what is recognised as just depends on time and circumstances. Precedent can be no guarantee of correctness. Judge Richard Posner has suggested, insightfully, that, sometimes, the more caselaw there is on a point, the more dubious may be its correctness – showing the difference between law and science.  

So judges can be, and are to a degree, lawmakers. But that does not fully answer the questions of what value they should ascribe to certainty, or how far uncertainty may positively yield to other considerations. Here, we should perhaps acknowledge that there are problems without right answers. Holmes was also correct, that law cannot be worked out like mathematics from general axioms of conduct. 

‘This…is the natural error of the schools, but it is not confined to them. I once heard a very eminent judge say that he never let a decision go until he was absolutely sure that it was right. So judicial dissent often is blamed, as if it meant simply that one side or the other were not doing their sums right, and if they would take more trouble, agreement inevitably would come.’ 

Ronald Dworkin, Professor of Jurisprudence at Oxford, took issue with this. He suggested in Law’s Empire (1986) that, for a Judge of infinite intelligence and diligence – he called him Judge Hercules – there is only one correct decision on any problem as a matter of law. I sat next to Professor Dworkin at dinner once and he confirmed that belief to me. But it is not how the law feels at a less exalted practical level. Judges frequently attach differing weight to different considerations, e.g. privacy and freedom of expression. The United States and the United Kingdom attach very different values to freedom of speech and protection of individual reputation.

Even well-founded precedents will from time to time need revisiting. Societal attitudes change over time in quite fundamental ways, e.g. in relation to same sex relationships or equality of terms of employment of different sexes, or the concept of a family. The courts are expected to reflect the best thinking and instincts of society and it is their role to keep the law in touch. Judges thus, on occasion, develop the common law in the perceived interests of justice, though as a general rule they do this ‘only interstitially’, to quote OW Holmes J again. This means not only that they act generally within the confines of the doctrine of precedent, but that the change so made must be seen as a development, usually a very modest development, of existing principle and so can take its place as a congruent part of the common law as a whole. In this way, develops what has been described by Maitland as the ‘seamless web’, and by Lord Goff as the ‘mosaic’, of the common law. Oliver Wendell Holmes was therefore right to encourage courts and lawyers to look at old principles and approaches with open, modern eyes, but as participants in a continuum, not as revolutionaries or individual activists.  

The only caveat I have regarding Holmes’ approach is his provocative statement that ‘the prophecies of what the courts will do in fact, and nothing more pretentious, are what I mean by the law.’ If that were true, should not the law be held to be what most people thought it to be when a dispute arose? Yet that does not always prove the case, and in fact, at this point, we continue to deploy the declaratory theory of law. We treat any change or development of the common law as having always been in place. That is for practical reasons. If it were not so, cases which depended on advancing or changing the law would never be brought. There would be no point in bringing them, no incentive, if development or change only affected future disputes. The problem was tackled brilliantly by Lord Goff in Kleinwort Benson v Lincoln City Council. 

Let me move to a case where you may detect the competing tugs of certainty and justice. It comes from the field of AI, with which all lawyers must now be au fait, although I hope that you have not sensed the large language model that I used to draft this talk. Quoine v B2C2 is a case from the Singapore International Commercial Court, where I sit on appeals. Quoine operated a computer platform for virtual currencies on which B2C2 traded. Due to a programming error, Quoine’s computer was cut-off from other cryptocurrency exchanges, and it ceased to generate market-making orders. This eventually led to Quoine’s algorithm concluding that two other traders on the Platform (Pulsar and Mr Tomita) were in margin sell-out positions and that it was necessary to take steps to close out their positions at whatever price was available. B2C2’s algorithm was programmed to throw out very low offers from time to time on a random basis, to cover situations where there was no real market. The sell-out was as a result agreed between the two computers operating by algorithm in the middle of the night at prices for Bitcoin one two hundred and fiftieth of the actual open market price. Was there a valid and enforceable contract? No human mind had been involved or mistaken at the time of the contracts. Equally, no third party deal had intervened, and the contracts could have been unwound. The majority of four judges upheld the bargain. I thought that it could and should be set aside, because no active human observing the bargains when they were made could have thought that anything other than some major mistake was occurring. In fact, when the owner of B2C2, the beneficiary of the trade, woke up to inspect the computer print-outs in the morning, he immediately sent email at 6:15am to Quoine’s support team saying that he had detected a ‘major Quoine database breakdown.’ And what, I wondered, would have been the position if it had been a mouse or the CIA which had severed links to the worldwide markets, rather than a programming error? But certainty prevailed. 

Another example of a trade-off between justice and certainty is provided by two Privy Council cases:  Cartwright v Superintendent of HM Prison and Gibson v USA. Two suspected drug offenders, Cartwright and Gibson, wanted in the USA, were ‘astonishingly’ discharged by a Bahamian magistrate, instead of being ordered to be extradited. The question was whether Bahamian law allowed an appeal from a magistrate’s decision discharging a defendant. Cartwright took this question to the Privy Council and lost three to two, Lord Hoffmann dissenting. The magistrate’s bizarre decision was set aside, and Cartwright was extradited to the USA. Gibson had, however, prudently absconded. His recapture took some time. His appeal therefore came on separately before a different Privy Council, which now held by four to three (Lord Hoffmann again dissenting) that the magistrate’s decision was not appealable. So Cartwright’s discharge stood. Lord Hoffmann therefore dissented twice. He was in favour in the second case of upholding a decision which he had concluded in the first case was wrong in law. Considerations of consistency in his (and incidentally my) view outweighed any need to establish the correct legal position in the second case. One should however perhaps add that Bahamian law had in the meantime been changed so as positively to introduce a right of appeal in such cases – but not of course in Cartwright’s particular case, since the legislation could not be retrospective. 

The tug in two directions in these cases was obvious. But uncertainty is often endemic in more mundane situations. Issues of factual and expert evidence fall for decision, often long after the event, sometimes involving projections about the actual or hypothetical future, sometimes about hypothetical past events. Those are the counter-factuals beloved of many claimants, though not all tribunals. In some cases, the law, to avoid continuing argument at ever higher levels, simply delegates responsibility to a single decision-maker, whose decision cannot usually be challenged. Jury trial is an instance. Subject to irrationality, appellate courts commonly achieve certainty by deferring to findings made by a court or tribunal on an evaluation of admissible evidence called before that court of tribunal. This is particularly so with the evaluation of oral evidence. 

Similarly, in other cases, a concept is so broad or indeterminate as to involve a value judgment about which different reasonable decision-makers could reasonably differ. In common parlance, a word like ‘bald’ might be seen as falling into this category. Is one thinking only of a completely bald person? Or someone who is substantially bald? Or has a bald patch? The lawyer’s first answer would, I think, be to look for assistance in the context in which the question was asked. But, if that provided no answer, the issue could, if it had to be decided at all, be left to the unassailable judgment of the first instance decision maker. Or take obstruction by protestors of the highway, to which a reasonable excuse is an answer. In DPP v Ziegler [2019] UKSC 10 relating to the O2 Centre, the Supreme Court held that that was for the first instance court to decide, unless there was some error of law on the face of the decision or it was one that no reasonable court could have reached. 

As a final example of the problems of predictable but fair decision-making, take the standards by which we conventionally decide issues of fact. Most commonly, we resort to the balance of probability. We may, without departing from it, acknowledge that the more serious the charge, the less likely it is to have occurred. But that is as far as it goes. If we are satisfied with a 51% probability that the defendant was responsible for doing the claimant wrong, the claimant recovers 100% of his loss. 

But is that fair? Is it even logical? Might it not be better to reflect the evidential difficulty in some form of percentage award, in cases where the 51% standard of proof could not be met? If justice is, as Holmes suggested, the prediction of how courts may decide, should we not recognise that litigation is little different from gaming?

I should at this point make clear that the balance of probability does not mean that the side with the better story or throw necessarily wins. This may sound paradoxical. But cases exist where, at the end of the day, the Court simply cannot work out what happened. The law does not then follow Sherlock Holmes’ admonition of Dr Watson in The Sign of Four: ‘How often have I said to you that, when you have eliminated the impossible, whatever remains, however improbable, must be the truth?’ Mr Justice Bingham was held to have succumbed to Watson’s fallacy in Rhesa Shipping Co v Edmunds (The Popi M) [1985] 1 WLR 948. The Popi M sank as a result of a large and sudden entry of water into her engine room through her shell plating in calm seas and fair weather, but also in water deep enough to preclude any inspection of whatever hole there may have been. Scuttling was not positively alleged by underwriters, but it was still for the owners to prove that the loss arose from some peril of the sea – not for example from pure decrepitude. Mr Justice Bingham went through the possibilities, rejected any suggestion that the crew may have been responsible (by for example some negligence, like leaving a sea cock open), and found himself left with two possibilities, viz. some form of unseaworthiness at the start of the voyage (which would not be a peril of the sea) or collision with a submarine; a yellow submarine as it became known, though no one had seen it. Neither unseaworthiness nor collision was supported by any positive evidence or was very likely, but collision with a submarine appeared to Mr Justice Bingham and to the Court of Appeal to be, despite its inherent unlikelihood in the scale of marine events, the more probable cause. The House of Lords took a different view. The quest was not for the least improbable cause. Here, nothing was established more probably than not as the cause of the loss, even if a submarine was more likely than any other cause. Sherlock Holmes’ dictum was not law. I do not think that Oliver Wendell Holmes would have disagreed.

But there are also cases where, despite the inability of the evidence to lead to any conclusion on the balance of probabilities or even likelihood, liability is imposed.  Take the example, of which I believe there to be a real-life American example, of two huntsmen, each of whom incautiously fires a bullet into what they think to be an animal, but was in fact a third huntsman. Each of the two huntsmen’s bullets was by itself sufficient to have killed the third huntsman. Has each or neither been responsible for killing the huntsman? The law will not tolerate the answer that neither has, even though it is clear that neither shot was by itself the cause of death. Each huntsman must be seen as having caused the death. 

It goes further. Take mesothelioma, consequent on some form of exposure to asbestos, quite often by a series of enterprises acting as employers. The scientific evidence, even to this day, makes it impossible to know which or how much exposure was or is critical to the development of mesothelioma. Ambient exposure, which we all suffer, can (it seems) also lead to its development. The law, with some hesitations on the route, has identified this as a situation where all enterprises or persons exposing the victim to any extent must share the liability, and this is so even if in percentage terms ambient exposure (i.e. no-one’s fault) may, in a particular case, appear a more probable cause than other exposure. The basic authority is Fairchild v Glenhaven Funeral Services Ltd [2003] 1 AC 32. The negligent exposers and their insurers and reinsurers must work out how to share the cost between themselves. The negligence of the concerns exposing the victim to asbestos justified a radically different approach from that in The Popi M. The Popi M concerned the factual cause of the sinking, whereas Fairchild involved established and life-threatening fault. By this radical deviation from the probability principles, the Courts ensured justice, but it has to be said that the complexities to which their solution gave rise have proved a discouragement to its further extension. 

There are other ways in which courts depart from a probabilistic, all-or-nothing approach. We have been talking about proof of past events. But suppose the question is what would have happened, but for the defendant’s breach of duty, in the past or the future? If what would have happened depends on one of the parties, the balance of probability is still deployed to decide.  But suppose it depends on the choices or actions of one or more third parties?  In this counterfactual, the courts take a different approach. They look at the percentage chance that the Claimant would have been better off, and award a percentage – say, 80%, 50% or 20% based on the perceived likelihood of the claimant being better off in the particular instance. 

The classic example is of the person who was wrongly deprived of the chance of entering a competition, the outcome of which would of course have depended on the view of the competition judges. In Chaplin v Hicks the claimant had, on the basis of photographs and newspaper readers’ responses, reached the final stage of a beauty contest, in which 12 of the final 50 (out of 6,000 original entrants) would be rewarded with places in a chorus line. A notice sent late in breach of contract deprived the claimant of her chance of competing in the final and winning one of the 12 places. She was awarded damages for the loss of a chance, assessed at 25%, of winning a place in the final 12 (12 out of 50 making approximately 25%.). Delicately and sensibly, the court seems to have proceeded on the claimant’s statistical chance of winning (as if she were a lottery player), rather than by putting itself in the actual role of the beauty contest judges. Similar awards, on a percentage chance basis, are most commonly found in professional negligence cases, involving medical negligence, where the consequences of a doctor’s failure promptly to diagnose are the issue, or involving solicitors’ negligence in the conduct of litigation, where the issue is what prospects the litigation would have had, had it been pursued promptly or properly.  

How logical, how sustainable is the distinction between hypothetical situations where only the parties are involved, and others where third parties are involved?  In the House of Lords case of Gregg v Scott, the distinction survived by three to two. The negligent failure by Dr Scott to diagnose that Mr Gregg was suffering from cancer delayed the commencement of Mr Gregg’s treatment by nine months, and thereby (on the medical evidence) reduced his chances of survival for ten years from 42% to 25%. Could he recover any, and if so, what damages?  The majority held that the applicable test was the balance of probabilities. Mr Gregg had not met that standard test before the breach – i.e. even if there had been no breach, he would not have had a 51% chance of surviving for ten years. All that the breach did was make it even more unlikely that he would survive for ten years. He recovered nothing. This is so, even though he would have recovered 100% if only his prospects of survival had been 51% before the misdiagnosis and even if they had been reduced to only 49% by the negligence. In short, a reduction in prospects of 17% yielded nothing, whereas a reduction in 2% could, in other circumstances, have yielded a 100% recovery. 

It is not in the circumstances difficult to understand why a strong minority dissented, as had been the case in the Court of Appeal. Lord Nicholls in the minority acknowledged the ‘impeccable logic’ of a probabilistic approach. But he pointed out that no one could really know whether Mr Gregg fell within the cohort of the 42% likely to survive or the cohort of 58% unlikely to do so before Dr Scott’s negligence intervened. He invoked Fairchild v Glenhaven Funeral Services Ltd [2003] 1 AC 32 for the proposition that, ‘In suitable cases courts are prepared to adapt their process so as to leap an evidentiary gap when overall fairness plainly so requires,’ and he thought that the use in the present context of statistics for the purpose of evaluating a lost chance makes good sense. Nonetheless, the minority’s view would have radically affected and required reconsideration of a whole range of claims, with major implications not only for the NHS and insurers, but also entirely outside the medical negligence field. Instead, the interests of certainty have prevailed, even though the boundary lines between probabilistic and percentage chance adjudication themselves remain remarkably unexplored. 

Thank you for coming and listening to me this evening. No doubt we would prove to have many different profiles when it came to resolving the pulls of certainty and justice in difficult cases. By pure coincidence, I received notice last night, as I was about to re-read this talk, of an interesting new book, Judicial Individuality in the UK Supreme Court by Lewis Graham, and I succumbed to the familiar temptation to look at extracts on the internet, including one’s own entry. I find that I am said to embrace ‘the (contrasting) values of self-direction and security,’ while prioritising freedom and self-direction. My wife would probably suggest that this means that I sit on the fence. I would prefer to read it as saying: recognising the importance of coherence and consistency while striving to achieve a just result. Be that as it may be, I hope that I have left at least some of you more certain than before that there is some truth in Oliver Wendell Holmes’ aphorism, and that this may be of value in your legal careers. The language of judicial decision may be mainly the language of logic, and we flatter ourselves that the law is predictable. But certainty generally is illusion, and repose is not the destiny of man – and how much good a good lawyer can do by being prepared to rethink established approaches that rest in reality on foundations that have become, or perhaps always were, uncertain. Thank you for listening to me. 

Copyright reserved

Jonathan Mance

---