Metamorphoses of the Lawyer

1. Lawyers have a problem

I experienced law school as a disappointment. When I was a child, one of the things my siblings and I would look forward to was our dad coming home from his firm and sitting down for dinner with us and mum, often still wearing his suit. We’d enjoy being with our parents for all of the normal reasons, but what we were really looking forward to was the debate that would ensue. Our dinner-table debates were an ever-moving feast of literature, history, art, philosophy, law and even obscure points of etymology, for which we eventually purchased the full twenty-volume Oxford English Dictionary. Not having smartphones, personal computers, or even a television, we’d often spend our leisure time reading, storing up ammunition for the dinner-time battles to come. These formative experiences gave each of us a strong belief in the value of learning and the virtue of those who pursued it. Those memories are still vivid for me, thirty years later. Since then, I have obtained three law degrees, from leading Australian, Singaporean and American universities, and while I think of each university I attended positively, I cannot say any of them fulfilled the ideal I held as a child of what an education in law would entail. Although I was fortunate in many ways in my legal education and in my two years working as a judicial associate, the lingering sense of disappointment that the material reality of modern legal education and legal work fell short of the ideal never left me. It was no doubt a significant part of why I didn’t pursue a legal career and have spent the last decade working in technology.

My story is peculiar in its particulars, but not its theme. Disappointment, disaffection, disillusionment, and even depression are common amongst both law students and lawyers. The specific complaints around hours, specialisation, the bureaucracy of the firm and of the court, the dullness of the work, the structural gender imbalances, and many related concerns have been expressed well enough elsewhere.¹ I want to ground these symptoms in what I think is the deeper loss of identity and purpose underlying them. Faced with the pace, scale and complexity of the modern world, and the ever-expanding empire of laws it has spawned, for the last century or more a storied profession has been evolving into a mechanistic treadmill of ever-increasing velocity that will, sooner than some may think, become humanly unworkable. And now the profession faces what seems to be yet another threat in the form of artificial intelligence. Globalisation, finance and computers are each important characters in that story of the changing profession, but I think the way to start to understand it is through the ideal of what it meant to be a lawyer, the sense of virtue that played a central role in the profession’s history, and why the phrase ‘lawyerly virtue’ seems like a quaint relic of a bygone time that has little relevance to the future of the profession.

In 1993 Anthony Kronman elegised the legal profession in The Lost Lawyer: Failing Ideals of the Legal Profession.² Kronman makes the case that the American legal profession (c. 1993) is in a spiritual crisis, as it has lost its regulative ideal: the ‘lawyer-statesman’.³ He argues that forces such as postmodern academia and the reduction of legal work to purely material ends have resulted in a profound disenchantment and disillusionment among lawyers and politicians and in civic life more broadly. In the first half of the book, he attempts to articulate the philosophical framework of the lost ideal, and in the second half he attempts to situate that ideal within the debates of his day. Kronman is aware that this kind of argument can come across as a conservative reaction to liberal progress, and tries to ameliorate that sense by using a more discursive style. While he has pedagogic ends, he ends on a pessimistic note, doubting that the next generation of lawyers will recover the ideal he has sketched. Indeed, I doubt anyone would argue that they did. From the vantage of the present, the lawyer-statesman seems even more distant. And yet, we find ourselves in a time in which the future of the profession as it has been known to the last few generations of lawyers is seemingly under threat, both by the disenchantment Kronman describes and by the advent of AI potentially capable of performing at least part of a lawyer’s labour. The same commercial forces that Kronman saw himself as struggling against are now threatening the type of materialistic lawyer he despairingly saw as the likely future. As such, to consider the history of the ideal of the lawyer at this juncture is not to stand, in New Haven, athwart history and yell stop, but to look at what the profession has at its disposal in the face of seemingly existential threats, and equip the profession for its next change.

Unlike Kronman, my goal is not to argue for a revival of a particular ideal of the lawyer that flourished in the United States in the later part of the nineteenth century. Rather, I’m going to sketch the ideal and practice of the lawyer from ancient times to the present, across the places and schools that have influenced the development of both the common and civil law families, with a particular focus on legal ethics. I do not view the histories or philosophies I will sketch as evincing a lost ideal so much as a series of metamorphoses which invariably raise the question of what form the next change will take. One may give this story I’m about to sketch other readings through the lenses of competition, labour, markets, gender, or power.⁴ Those lenses have their value. However, the point of this exercise is to consider the history and philosophy of the self-conception, regulation, and ethics of lawyers. In that sense this is both an internal and external view of the past, present and future of the legal profession. Here I would echo, by analogy, Ronald Dworkin’s plea to the reader in the first chapter of Law’s Empire:

Both perspectives on law, the external and the internal, are essential, and each must embrace or take account of the other. The participant’s point of view envelops the historian’s when some claim of law depends on a matter of historical fact … The historian’s perspective includes the participant’s more pervasively, because the historian cannot understand law as an argumentative social practice, even enough to reject it as deceptive, until he has a participant’s understanding, until he has his own sense of what counts as a good or bad argument within that practice.⁵

Many of the observations and claims of this essay have been elaborated by others at more length and with greater nuance. None are intended to definitively demonstrate a particular point in philosophy, history or jurisprudence. They are intended to be taken as a whole so that the profession and its ethics can be considered as a whole, as that is the point of the essay. I have laden the text with references, not to definitively prove what I’m saying, but to give the reader avenues to read further and make up their own mind. No single essay is going to definitively sketch the future of the legal profession. This is a dialectic in which I’m attempting to engage, in the hope of producing a productive response.

1.1 Ancient to medieval lawyers

Like Kronman, we start our story with Aristotle, as his understanding of virtue is indeed foundational to the history of Western lawyerly virtue. In Book VI of the Nicomachean Ethics, Aristotle distinguishes the intellectual virtues by their objects, the kind of thing each is exercised upon.⁶ Epistēmē is the virtue of demonstrative knowledge, exercised on what is necessary and eternal, such as the objects of mathematics and astronomy.⁷ Technē is the virtue of skilled making, in which the goodness of the activity is measured against an exterior product, as the carpenter is judged by their chair.⁸ Phronēsis is the virtue of action that contains its own end, where the doing leaves behind no further product and the goodness of the act is intrinsic to it. For such action there is no end apart from itself and acting well is itself the end.⁹ Education and doctrine supply the logoi of a practice, but they do not substantiate the virtue by themselves. Virtuous action is always within a particular lived context, not in the realm of universal concepts. Therefore, phronēsis is a hexis, a settled disposition rather than a body of doctrine.¹⁰ Those with more experience than theoretical knowledge are sometimes more effective in such action. Not because theoretical knowledge is without utility, but because theoretical knowledge must be balanced with perception, habituation and judgment.¹¹ The Nicomachean Ethics is in part a portrait of this kind of person, with an example being the Athenian statesman Pericles.¹² While we’re considering Greek philosophy, it is also worth mentioning the Stoic concept of oikeiōsis, which combines a self-perception present from birth with an innate concern for others that, as our rational capacities develop, we are called to extend progressively outward, drawing wider circles of humanity in towards the self. To put it crudely, Aristotelian virtue is organised around the question of human flourishing, namely what kind of character enables a human being to live well, whereas Stoic virtue is organised around living according to nature and reason, with the doctrine of oikeiōsis explaining how the rational animal’s natural self-concern expands into duties toward family, city, and mankind.

The Roman advocate (orator) was a recognisable ‘lawyer’ in the western tradition, and Marcus Tullius Cicero its most vivid representative.¹³ Cicero was also one of the earliest conduits of Greek virtue ethics into the Western legal profession.¹⁴ In his own work dedicated to his son, De Officiis, Cicero rendered phronēsis as the Latin prudentia and defined it as the practical knowledge of what to seek and what to avoid.¹⁵ Cicero saw his work as consistent with the Peripatetics (followers of Aristotle). Drawing on Panaetius, he connected the cardinal virtues to a Stoic account of oikeiōsis and recast this natural sociability as officium, duty to the res publica (‘public affair’)public affair , where an ethics of character meets an ethics of obligation.¹⁶ Moreover, Cicero situated his account of virtue within the practice of the advocate and treated bona fides (good faith) as a touchstone of justice.¹⁷ The result of this mixture, Cicero argues, is a set of permissions any modern lawyer would still recognise as foundational, including permission to defend a guilty party, provided he is not infamously depraved, and in doing so one may in good conscience maintain what is plausible even where it is not the entirety of the truth.¹⁸ Cicero’s De Oratore, written a decade earlier, explores the ideal of the formed orator.¹⁹ Through the character of Crassus, Cicero argues that the true orator and the statesman are one figure, because the eloquence the office requires cannot be separated from the wisdom that produces it.²⁰

A fresco of the Roman Senate. Cicero stands speaking with his arm raised, senators fill the benches, and Catiline sits alone at the far right.

Through Cicero we can also see one of the core tensions at the heart of the profession between virtue, money and status. The 204 BC Lex Cincia de donis et muneribus prohibited advocates from accepting money or gifts for pleading, on the logic that the office stood outside the marketplace it served.²¹ While Augustus restated the prohibition in 17 BC,²² Claudius lifted it, permitting fees up to a limit in AD 47.²³ Nevertheless, late Republican advocates often evaded the ban. Cicero took the forbidden munera (‘rewards’)rewards in the technically lawful forms of hereditates (‘legacies’)legacies and pecunia mutua (‘money on loan’)money on loan . In the Philippics he boasts that he had been left more than twenty million sesterces in legacies, and that this was justified because the legacies were returns of friendship within the economy of gratia (‘favour’)favour rather than payment.²⁴ In Aulus Gellius’s Attic Nights a story of Cicero is told which gives a flavour of the line Roman advocates walked in this respect:

For when he wished to buy a house on the Palatine, and did not have the ready money, he received a loan of 2,000,000 sesterces privately from Publius Sulla, who was at the time under accusation. But before he bought the house, the transaction became known and reached the ears of the people, and he was charged with having received money from an accused man for the purpose of buying a house. Then Cicero, disturbed by the unexpected reproach, said that he had not received the money and also declared that he had no intention of buying a house, adding: ‘Therefore, if I buy the house, let it be considered that I did receive the money.’ But when later he had bought the house and was twitted in the senate with this falsehood by friends, he laughed heartily, saying as he did so: ‘You are men devoid of common sense, if you do not know that it is the part of a prudent and careful head of a family to get rid of rival purchasers by declaring that he does not intend to buy something that he wishes to purchase.’²⁵

In the terms of Aristotelian virtue ethics, a fee is an end outside the act of advocacy, whereas phronēsis is an act whose end is the doing of the act. Nevertheless, since the beginning of the profession, prudentia has stood in tension with the imperatives of money and status, or perhaps has always included them.

The study of Roman law and lawyers is vast.²⁶ Suffice it to say for present purposes that by the later fifth century successive imperial constitutions had formed Roman advocates into a regulated guild with fixed membership, admission to which was by examination.²⁷ When Justinian’s commissioners gathered these constitutions into the compilation later known as the Corpus Juris Civilis (‘Body of Civil Law’)Body of Civil Law in the early sixth century, they preserved a constitution of the emperors Leo and Anthemius from AD 469 on the social utility of advocates which stated:

Advocates who explain ambiguous questions which arise in the course of litigation, and who, by the ability of their defence, frequently, in both private and public matters, restore the fortunes of those who have been ruined, are not less useful to the human race than if they had preserved their country and their relatives by taking part in battles, and receiving wounds. For We do not think that those who are equipped as soldiers with swords, shields, and cuirasses should be considered the only ones who protect Our Empire, but that the advocates, also, who have charge of cases contend as soldiers, and, trusting in the glorious power of eloquence, protect the hopes, the lives, and the children of those who are distressed.²⁸

Here the Ciceronian duty of the advocate is equated to that of a miles, a soldier of the res publica, and the professional bar to a militia, a branch of military service.²⁹ Justinian’s constitution for legal education Omnem (AD 533) authorised teaching in the royal cities and at Berytus (Beirut), and condemned instruction at Alexandria and Caesarea as ‘spurious’,³⁰ and attempted to impose a strict discipline on students.³¹ The Corpus also evinced a moralistic Christian theme that would come to dominate expressions of lawyerly virtue for over a thousand years. For example advocates were required to swear on the gospels during proceedings:

[T]he advocates employed on both sides shall be sworn with their hands upon the Holy Gospels that they will endeavor to do everything for their clients which they think to be honorable and just, by every exertion of their knowledge and power, and that they will, as far as possible, neglect nothing available for this purpose.³²

In the Corpus we can see Ciceronian duty transformed into a mixture of civic and Christian duty. When the Western Roman Empire fell in the fifth century AD the administrative structures and culture that had sustained the professional Roman advocate fell with it.³³ The Roman legal tradition I’ve been describing continued in the Eastern Roman Empire, albeit on a different cultural track. When an earthquake destroyed Berytus in AD 551 Constantinople became the epicentre of legal study, and the Corpus was re-written for a Greek-speaking audience in the Ekloge ton nomon (‘selection of the laws’)selection of the laws of the Isaurian dynasty (AD 741) and the Basilika of Leo VI (c. AD 892). The Muslim conquests of the seventh century removed Syria, Egypt and most of the eastern Mediterranean from Byzantine rule, leaving the eastern imperial tradition to survive in a diminished form until the fall of Constantinople in 1453.

For five centuries disputes in Western Europe were settled under customary law by oath, ordeal and feud. These methods often involved oath-helpers and the judgment of god(s) rather than trained pleaders.³⁴ Roman law survived in a diminished form in the Catholic Church where clerical compilers copied fragments of Roman jurisprudence into their canonical collections beside religious decrees.³⁵ The term advocatus was used to describe the lay protector of a monastery or church rather than a pleader of causes, who held a rudimentary duty of protection toward widows, orphans and the poor.³⁶ As noted, the Digest retained importance in the Byzantine East, but it seems to have fallen out of circulation in Western Europe. After it was quoted in a letter of Pope Gregory the Great in 603, it was not cited again until a Tuscan court invoked it in 1076.³⁷ The recovery of the Digest in Italy did not simultaneously restore the tradition of the Roman advocatus, but it did restore some of the logos needed for any such restoration in practice. From the eleventh century law started to revive as a learned discipline capable of training, and being used by, a recognisable profession. In Bologna, Irnerius and the glossators started to practise and teach close readings of the Corpus. The glossators’ students formed guilds from which some of the first universities grew.³⁸ Around 1140 Gratian imposed a comparable order on the church’s scattered rules in his Decretum (‘Decree’)Decree , the Concordia discordantium canonum (‘Concord of discordant canons’)Concord of discordant canons . Taken together, the Roman and canon law formed into a new ius commune (‘common law’)common law of medieval Christendom.

A fourteenth-century manuscript page. A small painting shows a bishop on a throne hearing monks and clerics, the text is surrounded by a dense commentary in smaller script, and a knight rides a dragon in the margin.

As the twelfth century progressed the ius communecommon law slowly displaced customary methods of dispute resolution, litigants started to need the help of trained pleaders again, and professional advocates reappeared in the courts of the church. In the faculties at Bologna, Orléans, Paris, Padua and other early law schools, students attended lectures on the Roman and canonical texts.³⁹ As well as their book studies they participated in disputationes (‘debates’)debates in which a question was argued from both sides under the eye of a master who gave the determinatio (‘decision’)decision . The doctorate at the end of the course was conferred not by written examination but by performance, the candidate defending theses and arguing cases until their professors judged them ready. The various collegia (‘colleges’)colleges of advocates imposed similar standards, admitting only those who had passed the gauntlet of oral examinations and then amassed further years of apprenticeship in court. There were undoubtedly protectionist aspects to these barriers to entry, but it is also clear from the sources that a significant motivation was a concern for restoring the integrity of the profession, cultivating the virtue of its practitioners and upholding a wider legal order. In 1270 the Établissements de Saint Louis exhorted French advocates to fulfil their noble duties, never to plead an unjust cause, to defend the weak and defenceless, and to pronounce their pleadings in a courteous manner.⁴⁰ Similarly, the 1274 papal Properandum systematised the rules for admission of advocates to the church courts and set strict limits on legal fees:

We decree that each and every advocate in the ecclesiastical forum, whether before the apostolic see or elsewhere, is to swear on the holy gospels that in all ecclesiastical causes and others in the same forum, of which they have assumed or will assume the defence, they will do their utmost for their clients in what they judge to be true and just. They are also to swear that at whatever part of the process they find out that the cause which they had accepted in good faith is unjust, they will cease to defend it; they will rather abandon it altogether, having nothing further to do with it, and will inviolably observe the rest of the above sanction.⁴¹

The Roman advocate’s duties resurfaced in canon law through a dispersed body of procedural rules, local statutes and admission oaths. In the Liber Extra, the De postulando (‘on pleading’)on pleading title principally regulated who might plead,⁴² while the oath against calumny belonged to a procedural title and was sworn by the parties.⁴³ These rules framed advocacy as an office subject to ecclesiastical discipline rather than an ordinary trade. The conception of the advocate as a priest or religious figure has a long history, and in the High Middle Ages we see it emerging clearly in the sources, sometimes excessively so as James Brundage records:

Lawyers’ aspirations for ever greater eminence sometimes got out of hand. ‘The advocate’s tongue,’ William Durand boasted, ‘may be described as a flaming sword’ of the sort that scripture depicted in the hands of God’s angels (Gen. 3:24). Doctors of law, Durand continued, belonged in the celestial hierarchy alongside the cherubim, the angels who guarded access to the true faith.⁴⁴

In this context it is worth recalling that Thomas Aquinas, the leading church scholar of the period, treated Christian ethics as the inheritor of Aristotelian ethics and based his account of justice on the framework of ‘the Philosopher’ (Aristotle).⁴⁵ Aquinas addressed the question of an advocate’s fees directly, saying that he ‘may justly receive payment for granting what he is not bound to grant’, provided the fee is moderate, taking account of the person, the matter, the labour involved and the custom of the country, and that the advocate who extorts more sins against justice.⁴⁶

A medieval painting of a university lecture. The lecturer reads from a raised desk while rows of students sit at benches, the front rows listening and the back rows talking or asleep.

He also held that an advocate is bound to take up the cause of the poor where their necessity is evident.⁴⁷ The association between chivalric virtue and lawyers was made explicit in the medieval restatement of the Roman advocate as miles, which became the lawyer as knight:

Lawyers could even claim the title of nobility by referring to themselves as ‘knights of the law.’ As a commentator of the time stated, ‘as a knight is bound to fight by the law of his sword, so too are lawyers bound to fight by the law of their practice.’⁴⁸

It is worth leavening the self-image of the medieval lawyer as a knight or priest with some perspectives of laymen, which were often more critical. Brundage translates a verse by the poet Matheolus, written around 1290, as follows:

What can I tell you about a lawyer?
He’s a bit like a filthy whore,
But nastier.
If a whore just rents out her ass,
He sells his tongue.
That’s even more degrading,
Since the tongue is more precious
Than the ass.⁴⁹

By the thirteenth century, lawyers commonly charged fees, and complaints about their greed were commonplace.⁵⁰ Ivo of Kermartin, who practised in the church courts of Rennes and Tréguier in Brittany, pleaded without fee for the poor and was canonised in 1347. However, a common rhyme of the time makes him out as the exception: Sanctus Ivo erat Brito, Advocatus et non latro, Res miranda populo (‘St Ivo was a Breton, a lawyer and not a thief, a thing of wonder to the people’)St Ivo was a Breton, a lawyer and not a thief, a thing of wonder to the people .⁵¹

Across the Channel the English crown had begun to control entry to the profession by royal order. In 1292 Edward I directed the justices of the Common Pleas to choose the attorneys and apprentices who should follow the court.⁵² The Serjeants at Law, the senior advocates who enjoyed exclusive audience in the Court of Common Pleas, are documented from the end of the thirteenth century, soon followed by the four Inns of Court, which coalesced into societies that combined training, residence and disciplinary jurisdiction over their members under a single roof.⁵³ Geoffrey Chaucer put a serjeant among his pilgrims in The Canterbury Tales (c. 1387):

A SERGEANT OF THE LAWE, war and wys,
That often hadde been at the Parvys,
Ther was also, ful riche of excellence.
Discreet he was and of greet reverence –
He semed swich, his wordes weren so wise.
Justice he was ful often in assise,
By patente and by pleyn commissioun.
For his science and for his heigh renoun,
Of fees and robes hadde he many oon.
So greet a purchasour was nowher noon:
Al was fee symple to hym in effect;
His purchasyng myghte nat been infect.
Nowher so bisy a man as he ther nas,
And yet he semed bisier than he was.⁵⁴

[‘A SERGEANT OF THE LAW (high-ranking attorney), prudent and wise, Who often had been at the Porch of St. Paul's (where lawyers gather) Was also there, very rich in superior qualities. He was judicious and of great dignity – He seemed such, his words were so wise. He was very often a judge in the court of assizes, By royal appointment and with full jurisdiction. For his knowledge and for his excellent reputation, He had many grants of yearly income. There was nowhere so great a land-buyer: In fact, all was unrestricted possession to him; His purchasing could not be invalidated. There was nowhere so busy a man as he, And yet he seemed busier than he was.’]A SERGEANT OF THE LAW (high-ranking attorney), prudent and wise, Who often had been at the Porch of St. Paul's (where lawyers gather) Was also there, very rich in superior qualities. He was judicious and of great dignity – He seemed such, his words were so wise. He was very often a judge in the court of assizes, By royal appointment and with full jurisdiction. For his knowledge and for his excellent reputation, He had many grants of yearly income. There was nowhere so great a land-buyer: In fact, all was unrestricted possession to him; His purchasing could not be invalidated. There was nowhere so busy a man as he, And yet he seemed busier than he was.

William Langland put the serjeants in Piers Plowman in their silk caps ‘Pleading the law for pennies and pounds’, and wrote that you might better measure the mist on Malvern Hills ‘Than get a sound out of their mouth unless money were showed’.⁵⁵ Indeed, it is evident that the same Roman tension between the virtuous and virtueless lawyer was alive and well across medieval Christendom.

Nevertheless, like the early universities, the pedagogy of the English Inns could be said to echo Aristotelian phronēsis. Senior members delivered readings on statutes, the whole hall disputed the cases put at the benchers' table after dinner, and junior members argued moots before them after supper.⁵⁶ This could be seen as a hexis formed through repeated participation in the practice under the eye of those already accomplished in it. John Fortescue, writing in the late 1460s for the instruction of the young Prince Edward, described the Inns as a studium publicum (‘place of study’)place of study better suited to study the laws of England than any university, and devoted several chapters of De Laudibus Legum Anglie to their exercises and to the cultivated common life that surrounded them. The members of the Inns:

employ themselves in the study of sacred and prophane history: here every thing which is good and virtuous is to be learned: all vice is discouraged and banished. So that knights, barons, and the greatest nobility of the kingdom, often place their children in those Inns of Court; not so much to make the laws their study, much less to live by the profession (having large patrimonies of their own) but to form their manners and to preserve them from the contagion of vice.⁵⁷

Fortescue’s aside about patrimony alludes to the fact that advocates in this period were typically of independent means, and often aristocratic. He also recorded that the feast at a serjeant’s induction ran for seven days, ‘like that at a Coronation’.⁵⁸ No new serjeant spent less than two hundred and sixty pounds on the other serjeants and nobles present.⁵⁹

1.2 Early-modern to modern lawyers

It would be a mistake to think that the medieval and early modern profession was simply an expression of aristocratic power. Indeed, the profession was self-consciously becoming its own locus of power and advancement. When a papal bull of 1292 gave doctors licensed at Bologna the right to teach throughout Christendom, the doctorate became ‘an order of intellectual nobility with as distinct and definite a place in the hierarchical system of medieval Christendom as the priesthood or the knighthood’.⁶⁰ Among sixteenth-century legal humanists, Jacques Cujas, the son of a fuller in Toulouse, rose to become the leading textual critic of Roman law. His editions of Roman legal sources and his commentaries on Justinian’s compilations joined legal reasoning to the characteristic disciplines of Renaissance humanism. He compared manuscript readings and drew on his command of Greek and Latin and his knowledge of ancient history to place particular texts within the periods and jurists from which they came. Cujas treated Justinian’s compilations as the product of centuries of legal and humanistic development rather than as a single, timeless statement of Roman law or aristocratic power.⁶¹ In France legal learning was known to carry commoners into the aristocracy itself. An aspirant to the position of magistrate in the parlements (the high courts of Paris and the provinces) could, through a combination of study and payoffs, enter the noblesse de robe (‘‘nobility of the robe’)‘nobility of the robe .⁶² These courts used their control over the registration and dissemination of royal edicts to exercise a certain degree of leverage vis-à-vis the king and other nobles.⁶³ Over in the British Isles, James VI of Scotland’s (later James I of England) Trew Law of Free Monarchies (1598) set out a doctrine of divine right of kingship in which the king stood as both author and giver of his realm’s law.⁶⁴ In Prohibitions del Roy of 1607 James applied this doctrine and tried to claim a right to adjudicate personally in matters touching the inheritances and lives of his subjects. He argued that the law was founded in reason, and that he, like the judges, possessed reason and could therefore decide as well as they. Edward Coke, Chief Justice of Common Pleas, answered as follows:

[T]rue it was, that God had endowed his Majesty with excellent science, and great endowments of nature; but his Majesty was not learned in the laws of his realm of England, and causes which concern the life, or inheritance, or goods, or fortunes of his subjects, are not to be decided by natural reason but by the artificial reason and judgment of law, which law is an art which requires long study and experience, before that a man can attain to the cognizance of it: that the law was the golden met-wand and measure to try the causes of the subjects…⁶⁵

Coke is distinguishing here between natural reason, with which God has endowed the King, and artificial reason, which is an art attained only through long study and experience. It is possible to read Prohibitions del Roy through the lens of power, but read through the history we have been tracing, one can also see the lawyerly self-conception of a distinct culture and cultivation.

A painting of a lawyer in a fur-trimmed red gown at a green table, one hand raised as he speaks, a clerk writing beside him and clients leaning in with coins.

Coke’s contemporary Alberico Gentili had read law at Perugia in the Bartolist tradition, and defended the jurists’ commentaries against Cujas and the humanists, before the Inquisition drove him from Italy in 1579 for his Protestantism. By 1587 the Italian exile was the Regius Professor of Civil Law at Oxford. In De Iure Belli he exhorted theologians to ‘keep silence about a matter which is outside of their province’, insisting that the law of war belonged to the jurists, not the clergy.⁶⁶ In 1584 the English crown asked his advice on Bernardino de Mendoza, the Spanish ambassador caught in conspiracy against Elizabeth, and Gentili held that an ambassador must be expelled rather than tried, the law of nations shielding another nation’s envoy.⁶⁷ Nevertheless, two years before Prohibitions del Roy Gentili had defended monarchical primacy in Regales Disputationes and treated the subject’s violence against the king as unjust.⁶⁸ Moreover, De Iure Belli did not merely restrain monarchs from waging war. Rather, it allowed for preventive war where a just fear of a growing power made an anticipatory strike a form of defence, and treated pirates as common enemies of humankind.⁶⁹ Gentili also became the salaried advocate of Spain in the English Admiralty. A Protestant exile from Catholic Italy represented Catholic Spain’s prize causes in a Protestant realm, arguing to the court that the English and Dutch captains who took Spanish ships were committing piracy under letters of marque.⁷⁰ In each of these debates, Gentili typically justifies his position with references to Rome, either to follow or (more rarely) distinguish their example, his discussion of ambassadors being typical:

So in the type of offense which we are considering, the Romans judged that the ambassadors of the Tarquins ought not to be put to death. Livy says: ‘Although their actions seemed to justify their being regarded as enemies, international law prevailed.’ We read in Appian’s work on the Punic wars that the ambassador of Syphax, who plotted the assassination of Masinissa, suffered no harm other than being charged with the crime. I should speak of the ambassadors of the Allobrogians who, according to the accounts of Plutarch and Sallust, were implicated in the conspiracy of Catiline, except that it might be urged in rebuttal that the reason why the Romans pardoned them was that they afterwards made amends for the crime which they had committed. Besides, they were provincials.⁷¹

Secularising jurists of this early modern period still saw themselves in dialogue with their Roman forebears, both in the substance and practice of law. Down in Salamanca, Francisco de Vitoria denied that conquest of the Indies could rest either on the Pope’s donation or on the Holy Roman Emperor’s dominion over the world, yet reopened a path to coercion through the ‘just titles’ of trade and preaching, resting the Spaniard’s right to travel and trade on the Roman law of nations.⁷² Grotius, much indebted to Gentili, recast the law of war as a law binding princes and peoples alike. In the prolegomena to De Iure Belli Ac Pacis he insisted that natural law would retain force even if there were no God, a distinctively modern formulation whose practical antecedents lay in his earlier De Iure Praedae, written at the Dutch East India Company’s request to justify Jacob van Heemskerck’s 1603 seizure of the Portuguese carrack Santa Catarina.⁷³ The same artificial reason that freed itself from religion, shielded an enemy envoy, and defied a monarch could also protect that monarch, license his wars, and sanction imperial avarice.

While James I had dismissed Coke from the bench in 1616, the constitutional struggles of the seventeenth century culminated with the Act of Settlement of 1701, which provided that judges’ commissions be made quam diu se bene gesserint (‘during good behaviour’)during good behaviour and that a judge be removed on the address of both Houses of Parliament.⁷⁴ In 1753 William Blackstone began lecturing on the laws of England at Oxford, the first university lectures on English law per se, and in 1758 he became the first professor of English Law. His introductory lecture, printed at the start of his Commentaries on the Laws of England as ‘On the Study of the Law’, traces some of the history we’ve covered, traversing Cicero, Roman law, Justinian’s codes, Anglo-Saxon practice, medieval canon law, monks, universities, the Inns of Court, Fortescue and Coke, weaving them all together to establish the provenance of English law and legal practice.⁷⁵ He lingers in particular on the decline of the Inns in the last hundred or so years, both in numbers and in quality, for which he suggests three reasons:

first, because the inns of chancery being now almost totally filled by the inferior branch of the profession, they are neither commodious nor proper for the resort of gentlemen of any rank or figure; so that there are now very rarely any young students entered at the inns of chancery: secondly, because in the inns of court all sorts of regimen and academical superintendance, either with regard to morals or studies, are found impracticable and therefore entirely neglected: lastly, because persons of birth and fortune, after having finished their usual courses at the universities, have seldom leisure or resolution sufficient to enter upon a new scheme of study at a new place of instruction.⁷⁶

While Fortescue saw the Inns as the studium publicumplace of study, Blackstone saw his task as restoring the study of English law to the universities, above all Oxford and Cambridge, from which the English common law had long been excluded. In Blackstone’s view, the universities, unlike the Inns of Court as they had become, offered a formation in which a gentleman’s ‘conduct and studies’ were ‘regulated by a discipline so wise and exact, yet so liberal, so sensible, and manly’.⁷⁷ Practical legal judgment, in Blackstone’s account, cannot be formed by practice alone but requires a clear logos. He describes law as:

a science, which distinguishes the criterions of right and wrong; which teaches to establish the one, and prevent, punish, or redress the other; which employs in it’s theory the noblest faculties of the soul, and exerts in it’s practice the cardinal virtues of the heart; a science, which is universal in it’s use and extent, accommodated to each individual, yet comprehending the whole community⁷⁸

Blackstone then invokes Aristotle, writing that ‘jurisprudence or the knowlege of those laws is the principal and most perfect branch of ethics’.⁷⁹ In this story we’ve been telling, one might see Blackstone as bringing the tradition of lawyerly virtue into the law of a single nation state, rehabilitating law’s intellectual discipline for the secular, and increasingly national, inclinations of his age.

A small unfinished oil painting of four judges in full wigs, seated close together behind a bench under a dark canopy. The judge at the centre, in a scarlet robe, frowns and holds a quill over a sheet of paper; of the three beside him one leans on his hand with his eyes shut, one has his head sunk on his chest, and the fourth looks away out of the picture.

In the North American and Australasian colonies, Blackstone’s Commentaries became the canonical primer of common law doctrine, and a model for the aspiring gentleman lawyer. John Marshall, James Wilson, Joseph Story and a generation of American lawyers learned their trade from Blackstone.⁸⁰ Sir Francis Forbes, who began his legal career at Lincoln’s Inn and served in various colonial roles before being appointed first Chief Justice of New South Wales in 1823, staked out his judicial independence from Governor Darling and refused to certify legislation he held repugnant to the laws of England, echoing Coke via Blackstone:

Your Excellency will perceive that my first duty is to take care that any proposed Law is not repugnant to the law of England in pari materia, and then to see that it is nearly consistent with or Conformable to such Law, as the local differences of the parent State and Colony may admit.⁸¹

In the Australian colonies, many leading early barristers had been at one of the four English Inns before emigrating. The earliest recorded barristers admitted in what is now Victoria included practitioners already admitted to Inns in London or Dublin.⁸² As to new admissions, we can get a sense of the type of education that was required from the 1853 rules concerning admission to practice in the Supreme Court of the Colony of Victoria which, in addition to standard legal texts, required prospective barristers to be tested on:

Mathematics.

Euclid, Book I.; Hutton or Butler’s Algebra, to Quadratic Equations.

Classics.

Greek.—Any one Book of the Evangelists, or any one Book of Homer’s Illiad, or any one Tragedy of Euripides or Sophocles.

Latin.—Any three Orations of Cicero, or any three Books of Virgil, or any one Book of Horace.

History.

Arnold or Hook’s Rome; Thelwall or Grote, Greece; Hume and Smollett, with continuation, England; Hallam’s Constitutional History; Alison’s History of Europe.⁸³

We also see the inheritance of some of the traditions we’ve been tracing in the early articulations of legal professional ethics in the American Republic. David Hoffman’s ‘Fifty Resolutions in Regard to Professional Deportment’ are Ciceronian, a portrait of the formed practitioner rather than a set of rules.⁸⁴ For example, resolution XXXIII declares that ‘what is morally wrong, cannot be professionally right, however it may be sanctioned by time or custom’.⁸⁵ George Sharswood, professor at the University of Pennsylvania from 1850 and Chief Justice of the Supreme Court of Pennsylvania from 1879, gave an influential treatment in An Essay on Professional Ethics, which would have been at home in many of the other earlier periods we’ve mentioned.⁸⁶

At the same time, we also see some of the same concerns about the profession we saw earlier. Speaking against the amalgamation of the profession in Queensland (hitherto split between solicitors and barristers), Samuel Griffith, the future first Chief Justice of the High Court of Australia, argued that ‘if the House was legislating for a Utopia, or a Commune, and wished to reduce all to a dead level of mediocrity, it might be found a useful measure, but not otherwise’, and then proceeded to paraphrase the proponents of the bill as arguing that ‘amongst the numerous mercenaries of the colony, the advocates who disgraced civilization the worst of all, were the highest in talent, as they were the most hungry for riches, and required the highest remuneration’.⁸⁷ To give some illustrative context for this opprobrium toward certain lawyers, some of the first solicitors of the Australian colonies were convicts. George Crossley, convicted of perjury in England and transported for seven years, arrived in 1799 and by 1803 was one of only two or three attorneys practising in Sydney with legal training.⁸⁸ Mr Justice Field reported to a royal commission that Crossley ‘transacted more business and with more activity than any one of the regular solicitors’.⁸⁹ One newly arrived English attorney received an annuity of four hundred pounds and half the fees under what appears to have been an arrangement allowing Crossley’s practice to operate in his name.⁹⁰ When the new Supreme Court was established in 1824, its Charter preferred the admission of practitioners already enrolled in Britain or Ireland, and excluded anyone whose conviction would have disqualified him from practice in the Westminster courts.⁹¹ In the new American Republic a significant resentment against lawyers grew as suits for debt began to rise after the Revolution.⁹² In the Massachusetts town of Braintree they voted for laws to ‘crush or at least put a proper check or restraint’ on ‘that order of Gentlemen denominated Lawyers’.⁹³ Benjamin Austin Jr, writing as Honestus, indicted the order’s ‘malpractices, delays and extravagant fees’, and argued that it should be ‘abolished as not only a useless but a dangerous body to the public’.⁹⁴ When Abraham Lincoln was making notes for a lecture on legal practice in 1850 he observed that ‘[a]n exorbitant fee should never be claimed’ and that as a general rule the whole fee should never be taken in advance, for ‘[w]hen fully paid beforehand, you are more than a common mortal if you can feel the same interest in the case’.⁹⁵ As to the tension between the lawyer’s and the layman’s conceptions of the profession Lincoln observed:

There is a vague popular belief that lawyers are necessarily dishonest. I say vague, because when we consider to what extent confidence and honors are reposed in and conferred upon lawyers by the people, it appears improbable that their impression of dishonesty is very distinct and vivid. Yet the impression is common, almost universal. Let no young man choosing the law for a calling for a moment yield to the popular belief — resolve to be honest at all events; and if in your own judgment you cannot be an honest lawyer, resolve to be honest without being a lawyer.⁹⁶

Over in France, the eighteenth century Barreau (bar) had carried an attitude toward its inheritance similar to Blackstone’s. Henri François d’Aguesseau, then avocat général (‘Advocate General’)Advocate General and later Chancellor of France, described the profession as ‘an order as ancient as the magistracy, as noble as virtue, as necessary as justice’.⁹⁷ Armand-Gaston Camus’s Lettres sur la profession d’avocat of 1772 restated d’Aguesseau, treating the office as a vocation whose obligations followed from formation rather than from regulation.⁹⁸ Needless to say, the French Revolution and Napoleon, who referred to lawyers as ‘factionists, artisans of crime and treason’, significantly disrupted the practice of law, as they did all aspects of French society.⁹⁹ However, by the 1810s the culture of the Barreau was showing distinct signs of continuity with its ancien inheritance.¹⁰⁰ François-Étienne Mollot’s Règles sur la profession d’avocat evinced the rule that an avocat (‘advocate’)advocate could not sue for his fees as a corollary of the office.¹⁰¹ As Trai Le notes in her study of the period:

The traditional reasoning underlying this prohibition is based upon the notion of the independence of the lawyer. Any public debate on fees would be beneath the dignity of any member of the profession and would tarnish the image of the lawyer. This notion is regarded as a rule ‘of professional elegance’.¹⁰²

The avocat could be paid but could not sue for his fee, or bargain over it, or let it become a matter of public debate. To do so would be to admit that the fee was the object of the work rather than its epiphenomenon. The same instinct ran through the English rule that a barrister could not sue for his fees, settled in Kennedy v Broun in 1863, explicitly as a guard against externalising the ends of a counsel’s practice such that their adherence to their duties is undermined:

if the law allowed the advocate to make a contract of hiring and service, it may be that his mind would be lowered, and that his performance would be guided by the words of his contract rather than by principles of duty,— that words sold and delivered according to contract, for the purpose of earning hire, would fail of creating sympathy and persuasion in proportion as they were suggestive of effrontery and selfishness; and that the standard of duty throughout the whole class of advocates would be degraded.¹⁰³

This position was upheld in Wells v Wells in 1914, where Swinfen Eady LJ stated ‘[i]t is settled beyond all question that counsel’s fee is not a debt but an honorarium, the fees are payable as a matter of honour and not of legal obligation’.¹⁰⁴ It is worth noticing that the bar of the nineteenth century, at its most confident, was also the bar in which the mechanisms for keeping the fee outside the act of advocacy itself were most fully developed.¹⁰⁵

A lithograph of a courtroom. An advocate in robes stands at the right with his mouth open and one arm flung out to point across the room; a second advocate on the left watches him sourly, two judges sit at the bench behind, and two clients wait in front with their hats on the table. The printed caption below says the lawyer is filled with the most intimate conviction that his client will pay him well.

The inheritance of Rome was perhaps the clearest in the German profession of the nineteenth century. Following Napoleon, Anton Thibaut had called for the immediate codification of German civil law on the French model in 1814.¹⁰⁶ Friedrich Carl von Savigny answered him later that year in Vom Beruf unsrer Zeit für Gesetzgebung und Rechtswissenschaft,¹⁰⁷ articulating a position that held significant influence across the German legal establishment of the nineteenth century. Savigny argued that law was not a product that could be made on demand by an act of legislative will, but only through the slow accrual of a society’s legal life, and to attempt codification before the science of law had matured was to mistake the artificial reason of the jurists for the natural reason of the legislator:

What has here been wished for from the study of the law, is not to be preserved in books, not even to be entrusted to individual lawyers, but to be common to all jurists who will labour in their calling with energy and spirit. A living school ought therefore to be formed, as the aggregate body of Roman jurists, not merely the followers of Sabinius or Proculeius by themselves, actually formed one large school.¹⁰⁸

Savigny is echoing many of the figures we’ve already mentioned, now in the idiom of German historicism. Like Blackstone, Savigny linked his historical perspective on the foundations of legal science to the development of legal reason within a culture. Savigny’s later System attempted to organise contemporary Roman law as a coherent body of jural relations.¹⁰⁹ Rechtswissenschaft (‘Jurisprudence’)Jurisprudence split into different strands, a number of which pushed for greater systemisation, most notably those within the Pandektenwissenschaft (‘Pandectism’)Pandectism movement. Reimann describes the movement’s endpoint as follows:

The concepts on which it rested were abstract entities, distilled from actual legal rules, but also detached from reality. From these abstract concepts it logically deduced legal rules to arrive at a system that was presented as the incarnation of timeless legal truth. But in the beauty of its logic, it paid scant attention to the realities of late nineteenth century German society.¹¹⁰

Nevertheless, the intensity of the German dedication to legal science was attractive, and, arguably, internally productive. American lawyers were much enamoured of the perceived German synthesis of history, reason and order, as evinced in accounts like those of James Morgan Hart who documented his experience studying at German universities, describing them as institutions where reason and intellect were prized above all else, logos in the pure sense, intellect pursued for intellect’s sake:

Many of the students who attend the university do so simply with a view to becoming in time professors. The entire personnel of the faculty is thus a close corporation, a spiritual order perpetuating itself after the fashion of the Roman Catholic hierarchy.¹¹¹

This religious devotion to the intellect, Morgan Hart says, produces a:

sovereign contempt that all German students evince for everything that savors of ‘bread-and-butter.’ … [e]ven [those that] have no intention of becoming [academics] pass three and four years of their life in generous devotion to study pure and simple, without casting a single forward glance to future ‘business.’ All thought of practical life is kept in abeyance. The future practitioners and the future theoreticians sit side by side on the same bench … hear the same lectures, use the same books, have every sentiment in common; hence the perfect rapport that exists in Germany between the lawyer and the jurist, the pastor and the theologian, the practicing doctor and the speculative pathologist, the gymnasial teacher of Latin and Greek and the professed philologist.¹¹²

On Morgan Hart’s account, German universities had achieved what Blackstone had articulated as his goal for English universities, and in doing so had seemed to create a culture around law almost as epistēmē, pure scientific knowledge, as opposed to phronēsis or technē. Not scientific knowledge in the dry formal sense, but the free pursuit of ideas for their own sake, which stands as the example of freedom throughout the student’s life:

We Americans, who live in a surfeit of freedom, as it were, can dispense perhaps with the libertas academica; but the Germans know too well that it is the only phase in the life of the educated classes that prevents that life from becoming an intolerable monotony.¹¹³

If we take the English Inns of Fortescue’s time as a time when the tradition of a place leaned towards law as pure practice, we might take PandektenwissenschaftPandectism as a time when the traditions of a place leaned towards law as pure intellect. Such instances of the legal profession were both the products of, and internally productive of, distinct cultures.

1.3 Post-modern lawyers

The rise of codification, the Industrial Revolution, globalisation and global cataclysms of the first half of the twentieth century fundamentally changed the conditions under which the legal profession operated. By the middle of the twentieth century, the nineteenth century’s faith in the slow ripening of legal science amongst a common community of jurists, in the bar as a self-regulating moral community, and even in the rule of law was contested across multiple axes. Karl Llewellyn’s The Common Law Tradition (1960) reflects this shift in interesting ways.¹¹⁴ Llewellyn begins from a ‘crisis in confidence’ about the reckonability of appellate decisions and answers it by describing the ‘craft’ of law. He describes a ‘Grand Style’ and a ‘situation-sense’ of advocates and judges, through which inherited doctrine is made responsive to the case at hand. Read in light of the history we’ve been tracing, these formulations conjure Aristotelian phronēsis, Ciceronian prudentia and Coke’s ‘artificial reason’. Llewellyn presents mature American legal realism as continuous with those older ideas while remaining practical, hard headed, and ‘modern’ about the law in action.

By the 1970s, sociological accounts were recasting professionalisation as a market and status project rather than an inherited vocation.¹¹⁵ The advent of the billable hour is particularly important in this story as it is seemingly harder to cast an activity whose every moment is organised around an external financial end as involving Aristotelian phronēsis or Ciceronian prudentia. Around the same time, competition law had begun to recast professional self-regulation as a potential restraint on trade. In the United States, Goldfarb v Virginia State Bar held that a minimum-fee schedule for legal services could amount to price fixing under federal antitrust law and rejected any sweeping ‘learned profession’ immunity.¹¹⁶ In Australia, the Hilmer Report on National Competition Policy and the Trade Practices Commission’s review of the professions led to parallel state and territory reforms.¹¹⁷ On the Continent, a similar mood can be seen in Wouters v Algemene Raad van de Nederlandse Orde van Advocaten.¹¹⁸ Wouters held that lawyers were undertakings for the purposes of competition law, and that bar rules could be decisions of associations of undertakings, even though particular restraints might not be unlawful if they are necessary and proportionate to proper practice. In the academy a similar ‘deconstruction’ of the profession ran through law-and-economics and critical legal studies in the United States, both of which had roots in American realism,¹¹⁹ and through sociology more broadly, particularly in France. Pierre Bourdieu cast law as a juridical field in which professional authority was a form of symbolic power,¹²⁰ while Lucien Karpik cast lawyers as brokers between state, public and capital.¹²¹ Each of these intellectual movements redescribed the life of the law under material metrics, ‘disenchanted realities’ if you will, and treated the practitioner’s formed judgment as the residue of a discipline whose claimed autonomy was an illusion.¹²² In 1986 Owen Fiss argued that the convergence of these movements could mean the death of law as an internally normative practice and as a ‘generative force’ in public life,¹²³ and Kronman essentially does the same in the second part of The Lost Lawyer.

A black-and-white photograph of three lawyers at a table with microphones: a man with a cigarette, a woman speaking, and an older man with his hands pressed together.

If we restrict ourselves to comparing the American legal profession from the 1970s to the 1990s (or the 2020s) with the American legal profession of the nineteenth century, then a narrative of loss of an ideal may ring true. But when we increase the aperture to consider the history of legal ethics in the form of the brief tale I have told, or some other version of it, then that ‘loss’ looks more like one amongst a number of metamorphoses. I have intimated that certain themes run through two thousand years of history, but needless to say that the lawyers of any half century frame within that time would have also thought, acted and looked quite differently from many other lawyers in other frames. The Industrial Revolution, globalisation and the lifeworld of twenty first century modernity have no doubt fundamentally changed the circumstances in which the profession operates. So too did the fall of the Roman Empire, the rise of Christianity, the Dark Ages, the Reformation, and the Age of Exploration, the Scientific Revolution and the Enlightenment. This is not to say that Kronman or those who feel similarly to him in the present are thereby proven wrong. Rather, it is to say that it is relatively easy for a single ‘grand’ era of a specific profession in a single country to seem quaint in the face of the exciting possibilities of a global future, but much harder to show our present and imagined futures are the categorical exceptions against the weight of a many-threaded two thousand year history. Many societies have seen value in a profession that cultivates jurisprudence, whether as a practice, a science, a humanist pursuit, or an expression of Aristotelian phronēsis or Ciceronian prudentia. Those same societies have also often seen the limitations, vices and hypocrisies of those that engage in that cultivation. Academics of the 1970s are far from the first to notice that the realities of practice don’t always match the rhetoric of virtue, and lawyers of the twenty-first century are hardly the first of their kind to attract substantial wealth and power. Restoring one form in those cycles, one metamorphosis, is, of course, impossible. But learning from those many changes such that they inform the next change is not only possible, but necessary.

If there was one key mistake I’d charge Kronman with, it is that he attempted to reconstruct his lawyer-statesman philosophically.¹²⁴ While philosophy can provide some of the structure and syntax, the tale of the lawyer’s future ideal must be grounded in the use of history as a regulative ideal, not a grand past to restore. Deborah Hussey Freeland’s ‘Recovering the Lost Lawyer’ attempts to contextualise Kronman’s story for a reader of 2014, arguing that the lawyer-statesman ideal could be recontextualised through gendered social formation and thereby recover a role for phronēsis.¹²⁵ Whether or not one finds Hussey Freeland’s attempt compelling, any study of the history of virtue ethics will demonstrate how different cultures and periods have re-conceptualised the virtues to a syntax and structure suited to the material realities of their moment. From the vantage point of the present, I would again urge the reader not to consider this history of lawyerly virtue as a history of an old candle in the dark, long since drowned out by the bright lights of modernity. Rather, consider what it meant to the people who defined it, believed in it, and criticised it, what roles it played in the societies in which it flourished, and the many changes it has undergone. In any event, before I offer my own suggestions on how that history might be synthesised into a new thesis for the future, there are two more histories, two more parts of any such synthesis, we need to explore.

2. The role of duty

A colour wood engraving of an eighteenth-century dining room. Kant, in a powdered wig, holds up a letter at the head of the table while eight guests in wigs and frock coats talk and lean in to listen, and his servant, in a yellow coat, stands by the door wiping a plate.

There is a common anecdote told about Immanuel Kant, that he was so regular in his habits his neighbours would set their clocks by his afternoon walks, with the intended evocation of a highly structured, austere life. However, the caricature of Kant as a dry, clockwork-like man is largely an invention of Heinrich Heine.¹²⁶ If the invocation of Kant’s name triggers an instinctive dread of difficult prose on difficult subjects, I’d suggest the palate cleanser of Thomas De Quincey’s translation of ‘The Last Days of Kant’, which paints a portrait of a man who loved company, didn’t stand on ceremony, and would have been a lively conversationalist.¹²⁷ One of the works he likely discussed at his regular convivial luncheons was Cicero. The De Officiis was among the Latin Kant read as a schoolboy at the Collegium Fridericianum, where Cicero was a standard part of the curriculum, and it is hard to overstate the extent to which Roman intellectual culture influenced German intellectual culture of the late eighteenth and early nineteenth centuries.¹²⁸ In 1784 Hamann reported that Kant was at work on a reply to a recent commentary on Cicero’s De Officiis, and a year later this became the Groundwork of the Metaphysics of Morals, the first of a number of Kant’s works on ethics.¹²⁹

Kant’s ethics took shape across the Groundwork (1785), the Critique of Practical Reason (1788), and the Metaphysics of Morals (1797), the last of which is divided into the Doctrine of Right and the Doctrine of Virtue. The moral duties Kant treated in these works were similar to those Cicero had treated in De Officiis, such as keeping promises, sparing the innocent, not betraying friends, and telling the truth. The difference lay in the ground on which the duties rested. As we noted, Cicero drew on a Stoic account of natural sociability, oikeiōsis, and translated it into the civic language of officium, res publicapublic affair and bona fides. For Kant a duty binds because reason itself requires it. The binding force comes from an a priori law underlying rational will. This move, from the Stoic underpinnings of civic life to the a priori reasons underlying morality for any rational being, ran with the current of Kant’s age, and still does with ours. Late eighteenth century European society was undergoing shifts from character to reason, from superstition to science, from the traditional to the rational, from the local to the global, from the communal to the individual and from anciens régimes to revolutionary republics. Kant’s philosophy drew its power, both then and now, by articulating a philosophy that typified the intellectual and social trends of the Age of Enlightenment.¹³⁰ Kant saw himself as articulating a new philosophy of ethics for what he saw as a newly enlightened age, what would later be described as ‘deontological ethics’ from the Greek for ‘the study of’ (‘-logia’) ‘duty’ (déon).

For Kant the moral valence of an action done from a sense of duty inheres in the will behind the action, not in the act itself. Kant separates an action done in conformance with duty to one done from duty. The shopkeeper who keeps fair prices only because cheating his customers would cost him in some way may meet the formal requirements of a duty of fairness without acting from that duty.¹³¹ His conduct has legality, outward conformity to the law, but not morality, which arises only where duty itself supplies the incentive. The same distinction applies to the making of laws, or ‘lawgiving’. Lawgiving that secures compliance through an incentive other than duty, such as habit, interest, or fear of the penalty, is juridical, whereas lawgiving in which duty itself is the incentive is ethical.¹³² So a code, a tribunal, an audit, any apparatus that enforces conduct from outside, can only secure legality. What it can’t do is guarantee that the practitioner acts from the duty the rule expresses. A related distinction applies to the form of the duties themselves. A ‘duty of right’ is narrow. It fixes the act required, so it can be enforced externally and is the proper focus of juridical lawgiving. A ‘duty of virtue’, on the other hand, is wide. It fixes an end, but leaves a latitude that the actor must fill through Urteilskraft.¹³³ To apply a duty of virtue is to judge that a particular case falls under that broad latitude. The judgment required to fulfil a broad duty of virtue cannot itself be supplied by a rule, because a further rule of judgment would in turn have to be applied, and so on, ad infinitum. Judgment is a capacity that is developed by practice, and its absence admits of no remedy.¹³⁴ Moreover, in Kant’s account, the norms of a legal system cannot provide the criterion of right, just like the precepts of logic cannot provide the criterion of truth. A lawyer can state exactly what the law holds and still not have answered whether what it holds is right. To reach what is right he has to leave laws and facts behind and look for the criterion of right in reason itself.¹³⁵ A jurisprudence that does not move beyond the law to include reason is, in Kant’s image, ‘[l]ike the wooden head in Phaedrus’s fable … a head that may be beautiful but unfortunately it has no brain’.¹³⁶ In Kant’s view, only the jurist who leverages the a priori criterion of right can supply the rational principles that positive lawmaking presupposes.¹³⁷ That criterion is provided by practical reason, above all in the idea that right consists in the conditions under which each person’s free choice can coexist with the free choice of all others under a universal law.¹³⁸

2.1 How duties became rules

The nineteenth century, the time Kronman saw as the climax of a form of lawyerly virtue, was also when the first deontological articulations of legal ethics in the Kantian sense began to emerge. Hoffman’s Resolutions and Sharswood’s lectures were discursive monographs addressed to a young man preparing for legal office. In contrast, the Code of Ethics adopted by the Alabama State Bar Association in 1887 was drafted under the mandate of a committee,¹³⁹ and its 56 rules were addressed to any member, regardless of their background.¹⁴⁰ Substantively the Code drew heavily on Sharswood, but a code addressed to every admitted member and focused on discipline is a different kind of document from a discursive portrait of the formed practitioner.¹⁴¹ Within two decades, versions of the Alabama Code had been adopted by ten other state bars¹⁴² and had become the principal textual material for the American Bar Association’s (ABA) 32 Canons of Professional Ethics, adopted in 1908 against what the drafting Committee called the tendency ‘to reduce our high calling to the level of a trade’.¹⁴³ The Canons were aspirational rather than directly enforceable in their own right, but state-by-state adoption gave them the force of law in the jurisdictions that took them up, and they remained the American profession’s normative instrument until they were superseded in 1969 and again in 1983.¹⁴⁴ In England the deontological move came more gradually, and was preceded by institutional development, particularly the Law Society of England and Wales.¹⁴⁵ Formal rule making power for the Law Society didn’t come until the Solicitors Act 1933, which the Society first exercised to produce the Solicitors’ Practice Rules of 1936. The 1936 Rules were relatively narrow, being confined to bans on touting, undercutting scale fees, fee sharing, and accepting referrals from claims farmers, and were largely promulgated to forestall direct government regulation of the profession in the wake of a client account embezzlement crisis.¹⁴⁶ The Society’s Guide to the Professional Conduct and Etiquette of Solicitors, first published in 1960, still did not claim code status. The first English instrument to do so was the Bar Code of Conduct of 1981, adopted on the recommendation of the 1979 Royal Commission on Legal Services, and the first solicitors’ code didn’t come until 2007.¹⁴⁷ Australia followed roughly the same arc, with its representative institutions dating from the middle of the nineteenth century and its codified conduct rules from the end of the twentieth.¹⁴⁸

The Civilian profession underwent a similar change, albeit by a different route and more gradually. In France the rules of the profession were developed through the disciplinary jurisprudence of each Barreau's Conseil de l’Ordre (‘Bar Association Council’)Bar Association Council . Mollot’s Règles of 1842 drew those dispersed précédents into a single text, however that was a synthesis, not a code.¹⁴⁹ National codification didn’t really come until the Règlement Intérieur National (RIN) of 2005.¹⁵⁰ Germany’s Lawyers’ Code (Rechtsanwaltsordnung) of 1878, enacted alongside the Reich Justice Acts, abolished the existing state concession system in favour of free advocacy, regulated admission, and disciplinary proceedings, but left substantive duties to professional understanding and precedent.¹⁵¹ The Bundesrechtsanwaltsordnung (BRAO) of 1959 retained that structure, continuing to delegate substantive conduct rules to the Standesrichtlinien (‘professional guidelines’) promulgated by the Federal Bar Chamber.¹⁵² Only after the Federal Constitutional Court struck down those guidelines on the ground that fundamental rules of conduct required a statutory basis was a body empowered to issue a binding code of conduct, the Berufsordnung of 1996.¹⁵³ The Civilian statutes carried the older vocabulary of duty into modernity more explicitly than their Anglo-Saxon counterparts. The BRAO opens by naming the lawyer ‘ein unabhängiges Organ der Rechtspflege (‘an independent agent of the administration of justice’)an independent agent of the administration of justice’.¹⁵⁴ Similarly, a French reform in 1971 merged various types of lawyers into a unified profession and declared the resulting avocat as an independent auxiliaire de justice (‘auxiliary of justice’)auxiliary of justice .¹⁵⁵ Nevertheless, taken in the longue durée, the moves toward deontological codification have been similar across the Western legal world, albeit at different speeds and with different emphases.

These moves toward deontological codification changed the form of lawyers’ professional duties. The eighteenth and nineteenth century monographs on lawyerly virtue were discursive. The Alabama Code, the RIN, the Solicitors Regulation Authority’s (SRA) Standards, and similar codes set rules of general application. Core duties such as competence, candour and independence were still there, but they had become rules that could be enforced. Initially, these rules retained the structural breadth Kant had suggested was necessary for duties of this kind. They were, initially, ‘wide’ duties in Kant’s sense. The competent discharge of a wide duty is, on Kant’s account, an exercise of judgment. The initial codifiers did not seek to, nor did they, take judgment out of legal professional ethics. They relocated it, from the discursively formed character to broad standards in rules of general application.

2.2 How rules became regulations

Over the course of the twentieth century legal professional ethics rules increasingly became systems for the specification of general duties. There are multiple ways a wide duty can be specified ranging from comments, opinions, guidance and policies to internal firm practices. Taken together, these individual components grew into a regulatory ecology that attempts to reduce the costs of resolving recurring questions and make lawyers’ conduct more predictable. We can see this in the history of the legal profession in the United States. The 1908 ABA Canons had been aspirational and had assumed a homogeneous profession that shared common standards, albeit that was a somewhat circumspect proposition even then.¹⁵⁶ The 1969 Model Code kept the canons, but distinguished aspirational ethical considerations from mandatory rules.¹⁵⁷ The 1983 Model Rules collapsed those three tiers into two, just black letter rules and interpretive comments, responding to demands for better guidance in an increasingly complex practice environment.¹⁵⁸ Further amendments, opinions, and then the American Law Institute’s Restatement (Third) of the Law Governing Lawyers of 2000 consolidated this increasingly specified body of law.¹⁵⁹ We can see similar developments in England and Wales, albeit they are not linear. Successive editions of the Law Society’s Guide added rules, principles, rulings, and other regulatory material, and by its eighth edition it had become increasingly unwieldy. The UK SRA’s 2007 Code reorganised this material, its 2011 Handbook adopted outcomes-focused regulation, and then the 2019 Standards and Regulations attempted to simplify things by specifying two shorter codes and relied more expressly on professional judgment.¹⁶⁰ Nevertheless, within weeks of the new code's commencement it had issued guidance on what one of its paragraphs required, and its standards have since been supplemented by enforcement policy and a continuing series of warning notices.¹⁶¹ Similar developments can be seen in the modern operation of professional ethics in private practice. Firms’ conflicts databases made it easier to have broader and more mobile partnerships. A conflicts database replaces part of the judgment and duty of the lawyer, narrowing the space in which judgment must then operate. Internal information barriers can address the risk of conflicted disclosure between different members of the same firm, if such a barrier is structural.¹⁶² Such barriers similarly replace an individual lawyer’s judgment of who can properly act with architecture and procedure.¹⁶³ It is this broader ethics ecosystem in which the ongoing pressure for specification now resides. The headline professional conduct rules still contain broad standards like ‘best interests’ and ‘reasonably necessary’, but in practice a practitioner will look to the ever-growing secondary material and industry practice that specifies what a duty requires in her circumstances. Indeed, there are strong arguments for increasing specification, including that both the market and the modern regulatory state prefer it. Both lawyers and clients prefer it as it makes conduct, cost and risk easier to manage. The state prefers it because it allows for an ever-growing and ever-changing profession to be governed efficiently. These incentives have a ratchet-like effect as they militate in favour of the parts of professional duties that can more easily be specified and audited over the parts that cannot. Specification does not remove the need for judgment, but it does relocate a fair bit of it to rule drafters, regulators, courts, ethics advisers, compliance specialists, and the internal structures and procedures of firms.

At the same time the discursive material of legal ethics has been largely set aside because, from the perspective of the ever-evolving system of rules and procedures, it has lost its utility. Texts like Hoffman’s Resolutions require a latitude in action broad enough for a lawyer’s character to matter. As those rules have been increasingly detailed and specified, discursive moral precepts seem to no longer have a role. For example, in 1983 the ABA’s Kutak Commission dropped the aspirational ethical considerations almost entirely with the rationale being that they were unenforceable before a tribunal,¹⁶⁴ and awkward vis-à-vis the antitrust position the bar had taken after Goldfarb.¹⁶⁵ In England and Wales, while the 1960 Guide to the Professional Conduct and Etiquette of Solicitors still carried discursive material on the bearing of the practitioner, the 1981 Bar Code of Conduct moved away from it. The solicitors’ Code of 2007 continued in that direction, as did the SRA’s 2019 Standards and Regulations. In Australia, the Law Council’s 2002 Model Rules of Professional Conduct and Practice had retained some discursive practice material alongside the conduct material, but the 2011 Australian Solicitors’ Conduct Rules dropped most of it. What remains of the older approach can be found in the practical legal training and supervised practice requirements that precede admission in Australian states, not in the text of the rules themselves.¹⁶⁶ The Civilian codes kept more of the discursive vocabulary, with the avocat (‘advocate’)advocate still an auxiliaire de justice (‘auxiliary of justice’)auxiliary of justice and the Rechtsanwalt (‘lawyer’)lawyer still an unabhängiges Organ der Rechtspflege (‘independent agent of the administration of justice’)independent agent of the administration of justice , but there too the operative provisions are directed at specified conduct rather than at character formation. France may be the most resistant case. The ‘collective honour’ tradition of the Ordre has survived as the barreau’s constitutive conception of itself through to the present.¹⁶⁷ The RIN of 2005 stands at the beginning rather than the completion of a specification process. In contrast, admission to practice in the United States is perhaps at the opposite end of the spectrum. The Multistate Professional Responsibility Examination is a two hour multiple choice test required for admission in most US jurisdictions.¹⁶⁸ It is common for those sitting it to simply learn its material by rote. It illustrates the extent to which professional standards have been rendered as specifications that can be tested independently of the cultivation of any practice. The wide form of the rules survives in the texts, preserving the resemblance to Kant’s wide duty, but the latitude those duties had once given has been substantially filled by the regulatory ecology that has grown around them.

2.3 How regulations affect judgment

Kant’s distinctions are useful heuristics in understanding what these changes mean. An enforceable code is juridical lawgiving in Kant’s sense. It secures compliance by attaching a sanction to the breach of its rules, which is an incentive other than duty per se. Lawgiving of that kind is a form of legality, the conformity of the act to the rule. It is not morality, the doing of the act from duty. Nevertheless, the early professional societies and codifiers were not trying to produce mere legality. They took themselves to be stating the existing duties clearly, binding them generally, and making them enforceable. But the core duties qua duties depend on circumstances and an environment beyond a code, because they are not duties to produce a particular result. In Kant’s eyes independence is not the condition of being unpressured. It is the disposition to decide a matter on its merits, regardless of any pressures to do otherwise. A practitioner who is independent only because the rule requires it has the outward form of independence while acting from a motive divorced from duty. Candour to the court works in the same way. If you’re only ever honest out of the fear of being caught lying, it is not the quality the rule was relying on. A code can require independence and candour, but, as far as Kant is concerned, the moral worth of compliance depends on the maxim or incentive from which the act is done. It is unlikely the early codifiers were diligently reading Kant when writing the early codes, but the point is that they were taking an approach that can broadly be described as deontological, and that Kant can give us a lens through which to view this transition.

The various professional societies and codifiers transformed a virtue-based practice into a rule-based practice the modern market and regulatory state required. In doing so they also established the circumstances of ever-increasing specification that arguably militates against the integrity of legal professional duties when viewed from a deontological perspective. Duties often can be found somewhere in their general form in the modern codes, and still include ideas of competence, diligence, candour, loyalty, and independence. However, viewed from a deontological perspective, none of those duties can be applied without judgment, Kant’s Urteilskraft, Aristotle’s phronēsis, or Cicero’s prudentia. Whether that judgment can survive once the duty is further specified and articulated, once the community of practitioners changes, or once the agent to whom the duty applies changes, is another question.

3. The role of consequences

A tall arched painting of five men in early nineteenth-century dress before a classical portico still in scaffolding. Bentham, elderly, in a broad-brimmed hat and long dark coat, stands at the centre holding a sheet of paper; two men stand at the left, and at the right a man in a red coat and a kneeling architect hold out the plans.

More than a century before the modern codes of legal ethics were drafted, Jeremy Bentham had already offered what could be seen as a more appropriate philosophical grounding to the increasing specification of modern professional ethics. Bentham was born in London in 1748 to an attorney who intended him for the bar, sending him at the age of twelve to Queen’s College Oxford and at fifteen to Lincoln’s Inn. He refused to practise and turned to writing instead, enjoying a position of inherited independence and remaining an outsider and critic of the profession for the rest of his life. The Oxford course in those years closed with Blackstone’s lectures, soon to be published as the Commentaries. Bentham heard the lectures in 1763 and (rightly) concluded that they contained a conservative defence of the system Blackstone described. The progressive polemic Bentham produced in response to Blackstone began with A Comment on the Commentaries, composed in 1774–76 but left in manuscript until CW Everett published it in 1928, and with A Fragment on Government, which Bentham extracted from the Comment and issued anonymously in 1776.¹⁶⁹ The Fragment’s Preface states the ‘fundamental axiom’ that ‘it is the greatest happiness of the greatest number that is the measure of right and wrong’.¹⁷⁰ Bentham’s primary complaint is essentially one of process. English judges had made the common law, and Blackstone, by deriving what they had made from natural law, had concealed that fact. Bentham argued that the law should be made openly, by the legislator, on the principle of utility, and known publicly to those it governed. Bentham spent the next six decades working that argument through general treatises, manuscripts on the limits of penal law, his Civil and Penal Codes, his procedural reforms and his Constitutional Code of 1830, while working alongside a circle of younger collaborators such as James Mill, Étienne Dumont, John Stuart Mill and Edwin Chadwick, who edited, translated and propagated his work across Britain, Europe and the Americas.

3.1 A consequentialist view of law

Bentham’s more systematic exposition of his views relevant to legal ethics first came in An Introduction to the Principles of Morals and Legislation.¹⁷¹ Bentham conceived the work as the introduction to a complete law code, and from the very first pages it declares a framework meant to dispense with the existing moral and legal vocabulary I’ve been canvassing. The first sentence declares that ‘Nature has placed mankind under the governance of two sovereign masters, pain and pleasure. It is for them alone to point out what we ought to do, as well as to determine what we shall do’.¹⁷² The principle of utility follows, defined as ‘that principle which approves or disapproves of every action whatsoever, according to the tendency which it appears to have to augment or diminish the happiness of the party whose interest is in question’.¹⁷³ Virtue, duty and natural rights are treated as obscure proxies for the calculable consequences in pleasure and pain, weighed across all affected parties, and adopted where the net outcome is pleasure. Applied to legal ethics, the moral assessment of the practitioner’s conduct is treated as a calculation of the outcome of the lawyer’s conduct, aggregated across every party whose interest the conduct touches. Lawyers’ professional duties are reformulated under the principle of utility as duties whose weights are distributed across all affected interests according to what they stand to gain or lose.

Bentham continued this project in a separate manuscript focused on a conceptual account of law itself, which HLA Hart edited in 1970 under the title Of Laws in General.¹⁷⁴ According to Bentham, a law is an expression of the sovereign’s will in an imperative statement directed at a class of subjects, and backed by a sanction.¹⁷⁵ The imperative may command, prohibit or permit.¹⁷⁶ The class of subjects may be defined broadly or narrowly, and the sanction is what gives the subject a motive to act as required.¹⁷⁷ To put it in more explicit utilitarian terms, we obey laws to avoid pain. A complete law is one in which the imperative is articulated with the precision needed for any subject to know whether his conduct falls within it.¹⁷⁸ A law that is incomplete leaves the application to a judge. A system of laws that systematically requires judicial ‘completion’ requires the judge to silently make the law the judge claims, in Bentham’s time at least, merely to ‘discover’. Bentham was under no illusions that no extant laws met his criterion of completeness, indeed this awareness was one of his motivations in setting out his theories:

This being the description of a complete law, where then it may naturally be asked is there a specimen of such a law to be met with? I answer—nowhere. Before any such specimen can be found, a perfect plan of legislation must first have been produced: perfect in point of method at least, whatever it be in point of matter.¹⁷⁹

Bentham would later seek to write a ‘complete’ code of law he called the Pannomion.¹⁸⁰ His ambition was to make the entire body of legal duty knowable in advance and to leave the judge with nothing to do but apply the rules to the facts. In Bentham’s eyes, a system of laws that cannot be fully specified in advance is one whose duties no one can know in advance, and the principle of utility cannot endorse a legal system whose duties cannot be known by those they apply to.¹⁸¹

The implications of Bentham’s arguments for the legal profession are drawn out, with a fair dose of his somewhat acerbic humour, in Truth versus Ashhurst; or Law as it is, contrasted with what it is said to be.¹⁸² Bentham writes:

It is the judges (as we have seen) that make the common law. Do you know how they make it? Just as a man makes laws for his dog. When your dog does anything you want to break him of, you wait till he does it, and then beat him for it.¹⁸³

Bentham’s case against the common law is that it is unknown to those it governs, retrospective in its operation, and discoverable only by a class that lives by its mysteries. This structure is inconsistent with the demand of utility which only works when people can foresee what their actions bring about. On this reading, professional self-regulation serves the profession’s own interest rather than the public good. Bentham’s remedy to this is threefold. First, as we have seen, the substantive law should be codified in a Pannomion, a ‘complete’ legal code. Second, judicial procedure should be codified into a complete procedural code, to which Bentham devoted a substantial body of his later work, including the Principles of Judicial Procedure and the five volume Rationale of Judicial Evidence edited by the young JS Mill in 1827. Third, the storied practitioner should be displaced by a body of rules a layperson can read, understand and apply. Once a duty is fixed in these ways, the agent who fulfils it becomes fungible, and the ‘lawyer’, traditionally understood, is no longer needed. Duties of this form can extend to any agent capable of acting, because they are not anchored on the nature of any particular agent.

3.2 Consequentialism and legal ethics

Bentham combines a criterion of rightness with a particular account of the good. His criterion of right makes the moral quality of an act, rule or institution depend on its consequences. His account of the good equates value with pleasure and requires the aggregate good (pleasure) of all to be maximised. ‘Consequentialism’ is the name modern philosophy gives to his criterion of rightness. Bentham is a consequentialist because consequences are decisive for him, and a utilitarian because he identifies the good with the aggregate balance of pleasure. Utilitarianism is one form of consequentialism, but not the only one. Applied to legal ethics, consequentialism asks which arrangements most reliably promote ends like clients’ interests or the proper administration of justice. The desired consequences are set by law and professional principles rather than by a calculus of pleasure and pain.

The modern history of regulation of the legal profession can be seen in consequentialist terms, particularly in the changing relationship to the agent performing legal services. Goldfarb rejected the use of the ‘learned profession’ to exempt lawyers from antitrust scrutiny.¹⁸⁴ Some legal work began to migrate to outsourcing centres,¹⁸⁵ and the regulations bound responsibility for the output to the supervising lawyer.¹⁸⁶ The corporate accounting scandals involving Enron and WorldCom led to Section 307 of the Sarbanes-Oxley Act 2002, which gave the Securities and Exchange Commission the power to set minimum standards of professional conduct for attorneys appearing before it and required ‘up-the-ladder’ reporting.¹⁸⁷ John Coffee’s Gatekeepers (2006) set out a conceptual framework for this, describing lawyers, auditors and rating agencies as reputational intermediaries whose duties run to the various stakeholders in relationship with them.¹⁸⁸ More recently, the Independent Review of Legal Services Regulation proposed that the regulatory perimeter be redrawn around the activities themselves on the basis of risk.¹⁸⁹ The place of AI systems in professional ethics should be seen as part of this consequentialist story. The ABA’s Generative Artificial Intelligence Tools discusses how a lawyer using AI remains ‘fully responsible’ for its outputs, and that the lawyer’s duty of competent representation extends to whatever outputs are produced by AI.¹⁹⁰ The Law Society’s recent report Shaping the Future of Agentic AI in Legal Practice describes firms using random samples from work product to estimate the risk that an AI system has missed relevant data. The report describes this as a shift from checking individual documents to a ‘system-level assurance’.¹⁹¹ Members of the legal profession who were interviewed for the report questioned whether lawyers could remain solely responsible for systems they could not understand, and discussed liability, audits and accreditation of systems as possible responses.¹⁹² These initial experiences with AI in legal practice suggest an emerging division. Some professional duties will likely remain with the lawyers that deploy or use a system, while the system’s outputs are assessed globally and consequentially. This fits with the consequentialist approach to professional ethics. An AI system cannot form a character (in the human sense) and cannot act from duty. The relevant questions are who bears responsibility for its outputs, what the evidence about its performance shows and how it should factor into professional ethics.

The past half century of consequentialist professional regulation could be seen as minimising the salience of the gap between the rule and the disposition its duty presumes. A deontological code needs a disposition in the agent, but a consequentialist code just asks for an outcome. This approach has its uses. Fee schedules, monopolies, and business structure barriers were anticompetitive arrangements whose costs were borne by the public. The same outcome orientated logic was salient in response to the accounting scandals of the early 2000s. Moreover, the focus on outcomes addresses a structural problem with legal ethics inherent in the modern economy. Legal work has migrated to agents that the older duty orientated approaches do not work with, including paralegals, in-house compliance teams, outsourcing centres, and now, AI systems.

3.3 The limits of consequentialism

There are also clear downsides to the consequentialist approach when it is taken as a complete account of both legal ethics and law. Bentham’s complaint against the common law was not simply that it was technically imprecise, but that it made legal obligation depend on an obscure and retroactive form of judicial lawmaking. Judges, on his account, make law while only pretending to discover it, leaving the subject to learn the rule they should be following only by breaking it.¹⁹³ Codification was meant to remedy this by relocating all lawmaking to the legislature. Read charitably, Bentham’s Pannomion is an institutional proposal for a public and prospective body of law, where defects found by judges would be corrected openly through legislation rather than ‘concealed’ inside judicial discretion. This is seemingly attractive from a democratic perspective. However, I would suggest that it is an incomplete account of how law actually works in a pluralistic liberal democracy. Legislative correction can repair a general gap after it has appeared, but it cannot abolish the need for judgment in the specific case in which the gap appears. The judge cannot suspend or decline to hear every hard case until the legislature has spoken, and if the legislature answers each hard case individually it ceases to legislate by general, prospective rules at all. Nor does amendment solve the deeper problem, because the amended rule will be, and should be, a general rule cast in language for future cases not yet known. Hart’s point about the open texture of law is not merely that some words are vague, or that some codes are poorly drafted. It is that rules capable of governing future cases will, by design, have both a settled core and a margin in which the rule must be applied by judgment, rather than lexically.¹⁹⁴ Bentham’s approach can perhaps make law more public and accountable than the common law of the eighteenth century, but it cannot remove judgment from law.

Moreover, hard cases do not arise only because legal language is sometimes penumbral. They often arise because people reasonably disagree about the goods and forms of life the rule is attempting to govern. John Rawls’s account of ‘burdens of judgment’ is apposite here, as he elaborates why such disagreement is not simply error, bad faith or a case of insufficient information.¹⁹⁵ A liberal democracy does not tolerate plurality as an unfortunate residue to be chipped away at. Citizens in a liberal democracy hold different, and often incompatible, conceptions of the good life, the purposes of cooperation, and the ends for which law should be used. Many will not accept Bentham’s premise that laws should be enacted according to the principle of utility, just as others will not accept Kant’s ethics, Aristotle's virtues, religious obligation, or any other single ethical doctrine as the overarching principle of law. The danger of a Benthamite legal ethics is not that it is consequentialist, which modern regulation must be in some form or another, but that it risks treating one contestable set of ethics as though it were the neutral grammar of a legal order. If the space for judgment is closed by evermore specification, the law will fail when an unanticipated case arrives. It will have to smuggle a contestable settlement into its rules as though it were the only rational settlement, or otherwise force the case into a rule whose purpose it no longer serves. Nor is the difficulty solved by handing the ‘residue’ to a solitary judicial Hercules, human or artificial. Even if a Herculean reasoner could produce a coherent answer in every case, the legitimacy of legal judgment cannot come from cognitive power alone. It comes from judgment exercised by institutions and professionals who can both participate in public discourse, and be held accountable in a meaningful sense. I suggest, therefore, that legal ethics in a liberal democracy cannot seek the abolition of judgment. It must instead ask how judgment can be cultivated, disciplined and made accountable as the law, and the agents who practice it, change.

Another fundamental issue with the utilitarian approach to legal ethics is the effect it has on the agent tasked with engaging in legal dialogue and applying legal judgment. This was one of Kronman’s central concerns. Indeed, one may say that it is his motivating concern. The lawyer-statesman was not merely a technician of legal outcomes. He was a person whose practical judgment had been formed by law as a way of life, a way of life Kronman wanted to revive, or at least give its swansong. More squarely within the philosophy of ethics, Bernard Williams considered the cost of the utilitarian indifference toward the agent in ‘A Critique of Utilitarianism’ (1973).¹⁹⁶ Williams observed that focusing entirely on the consequences of an agent’s action:

is to alienate him in a real sense from his actions and the source of his action in his own convictions. It is to make him into a channel between the input of everyone’s projects, including his own, and an output of optimific decision; but this is to neglect the extent to which his actions and his decisions have to be seen as the actions and decisions which flow from the projects and attitudes with which he is most closely identified. It is thus, in the most literal sense, an attack on his integrity.¹⁹⁷

This degradation of the human condition is an inherent structural risk of a consequentialist ethics. Gerald Postema made a similar point in the context of legal ethics, describing the ‘moral distance’ that role-based and outcome-focused frameworks open between the lawyer’s professional self and her private one.¹⁹⁸ The lawyer can come to do as a professional what she would not do as a person, treating the role’s demands as a partition behind which her ordinary moral commitments need not be brought to bear. The result is the morally disintegrated practitioner, not corrupt in any straightforward sense, but unable to bring the parts of her life into a coherent whole.¹⁹⁹ This personal risk manifests directly in the lawyer’s public actions. The public risk is not the risk of any one act going wrong. It is a slow erosion of her capacity to act from her own convictions in a coherent way.

This erosion becomes particularly pertinent in the face of rapid technological change. It is the unknowable future changes to society, the market and the law itself that pose perhaps the most significant ethical risk to the actions of lawyers. It is often the newest technologies that are the most susceptible to abuse and misuse, before they can be metabolised by the institutions of both the liberal democratic state and the profession itself. If a lawyer’s ability to exercise phronēsis, or prudentia, or Urteilskraft, has been eroded by years of Benthamite indifference to the disposition and character guiding her actions then she will be incapable of acting ethically in any sense of that term in the face of such change. Indeed, we might say that the amoral behaviour often underlying the various banking, accounting and corporate crises over the last fifty years can only ever be systemically addressed by taking the character of the agent seriously in the first place. Moreover, thinking we could eventually discard the human agent, and thereby remove this risk, by relying on a sufficiently advanced artificial agent in some future legal ethics, doesn’t solve the problem so much as move it. The question then becomes whether and how you can create an AI to have a certain ethical disposition it consistently applies when handling novel cases, and how a human or machine building or auditing such an agent would be able to judge such a disposition. Moreover, to actually remove this risk would require such a system to be applied in all cases without the need for human intervention. Indeed, it would require an artificial judicial Hercules, which gets us back into the same territory as when we considered whether AI provided any ‘solution’ to the democratic pluralism inherent in the rule of law, and the necessary role of judgment. To consider this as a goal one moves toward incrementally is to misunderstand both the nature of the risk, and the inherent tension between the putative goal and the rule of law in a liberal democratic system. There is an irreducible dispositional quality to any system of liberal legal ethics that a consequentialist approach will always be incapable of accommodating in a systematic fashion, regardless of how sophisticated AI becomes.

4. Post-modern legal ethics

A black-and-white photograph outside the US Supreme Court. A woman holding a wire coat hanger draped with a lace garment shouts at a man in glasses and a jacket, who shouts back with a raised hand; demonstrators with placards stand behind them, one placard reading Keep Murder Legal.

The previous sections traced three streams in legal ethics, taking Aristotle, Cicero, Kant and Bentham as their touchstones. These traditions did not succeed one another so much as settle into layers, each laid over the last and collectively forming the palimpsest of legal ethics as it exists today. The virtues of independence, integrity, candour and loyalty are still held up to be aspired to. The rules that attempt to encode them are deontological in form, duties of general application owed by every member of the profession. And the regulation enforcing them is often consequentialist, testing the lawyer, or their agent, by what the conduct delivers and to whom. This layering is not peculiar to law. In After Virtue Alasdair MacIntyre argues that it is the condition of all moral language in advanced modernity. According to MacIntyre, what we now possess are:

the fragments of a conceptual scheme, parts which now lack those contexts from which their significance derived. We possess indeed simulacra of morality, we continue to use many of the key expressions. But we have — very largely, if not entirely — lost our comprehension, both theoretical and practical, of morality.²⁰⁰

The result is what MacIntyre calls emotivism. Emotivism is a state of affairs in which moral argument keeps the forms of older frameworks while having lost the shared substance that once gave those forms their meaning. This is why modern moral debates are interminable. The arguments from these different approaches do not share premises, while they each claim some form of objectivity, and their own intellectual genealogy.²⁰¹ If interlocutors have no agreed premises, or shared understanding of history, they will simply talk past one another and end up in a kind of unacknowledged meta-relativism. Despite any protestations that one’s approach is ‘rational’, ‘fair’, or even historically grounded, ‘all moral judgments are nothing but expressions of preference, expressions of attitude or feeling, insofar as they are moral or evaluative in character’.²⁰²

MacIntyre gives us a way to think about where the history of legal ethics has left us. As viewed by MacIntyre, Aristotle, Cicero and Kant have not disappeared from legal ethics to be replaced by pure consequentialism. Those older approaches have survived in fragments, layered over one another in the language of virtue and duty, while the shared teleologies, education and cultures that once gave those fragments their substantive meaning no longer command authority. The virtues of independence, candour, loyalty and integrity remain, but as mere words and phrases in codes, rules and regulatory instruments. We are left with a language of an ethics denuded of the social context that once made it intelligible. MacIntyre would probably say that emotivism manifests in legal ethics as a legalism, relying solely on semantic reflections of virtues and duties that once had substantive meaning in lives and communities. He would probably agree that consequentialist regulation is perhaps most true to modern legal ethics, because it judges the outcome and not the agent, their character, or their community. He would also likely argue that such consequentialist legalism is an insufficient foundation for a lawyer trying to assimilate their professional work into the unity of a life, or one trying to exercise the phronēsis, prudentia, or Urteilskraft law sometimes calls for. He would argue that there remains an irreducibly dispositional, and therefore irreducibly human, quality to liberal legal ethics that consequentialist philosophies, however refined, cannot fully accommodate.

4.1 The lawyer is now a manager

On MacIntyre’s account this post-traditional moral condition is embodied in stock social roles he calls characters, the ‘moral representatives of their culture’ and ‘the masks worn by moral philosophies’.²⁰³ The role that matters to our present analysis is what MacIntyre calls the Manager. The Manager treats the ends of the organisation as given and concerns himself only with technical implementation, with ‘the rationality of matching means to ends economically and efficiently’.²⁰⁴ He is accepted by society as an essentially uncontested figure who keeps to the realm of measurable effectiveness, where rational agreement is supposed to be possible, and for that reason he does not take part in moral debate. He has an indifference to the difference between manipulating people and treating them as ends in themselves. Following Weber, MacIntyre argues that managers claim to serve given ends neutrally but can only justify themselves by their effectiveness, essentially meaning that ‘bureaucratic authority is nothing other than successful power’.²⁰⁵ The Manager does not have a ‘practice’, which MacIntyre defines as:

any coherent and complex form of socially established cooperative human activity through which goods internal to that form of activity are realized in the course of trying to achieve those standards of excellence which are appropriate to, and partially definitive of, that form of activity, with the result that human powers to achieve excellence, and human conceptions of the ends and goods involved, are systematically extended.²⁰⁶

The goods internal to a practice ‘can only be identified and recognized by the experience of participating in the practice in question’.²⁰⁷ The Manager stands outside the work, counts only what can be counted, and sees only the external goods of money, power and status. In MacIntyre’s formulation, ‘when achieved they are always some individual’s property and possession’; they are won in zero sum competition.²⁰⁸

MacIntyre argues that virtues, particularly justice, courage and truthfulness, are needed by every practice, because practices cannot survive among people who will not deal with one another honestly, fairly and at some cost to themselves. Similarly, the virtues also need practices, because practices make the virtues intelligible. Practices are the activities in which people learn to recognise goods internal to a shared form of work, submit themselves to standards of excellence, and become answerable to their peers. A virtue is ‘an acquired human quality the possession and exercise of which tends to enable us to achieve those goods which are internal to practices’.²⁰⁹ A virtue requires narrative unity. It is not a skill switched on in one part of life and off in another. It is a disposition that runs through a life judged as a whole.²¹⁰ Furthermore, the virtues depend on a living tradition. To take up a practice which leads to virtue is to join those who have gone before and learn from them, such that: ‘the story of my life is always embedded in the story of those communities from which I derive my identity’.²¹¹ Practice, life and tradition are what the virtues require, and the Manager has none of them. He belongs instead to institutions that are meant to carry practices, while in reality they endanger them.²¹² Modern institutions deal in external goods such as money, power and status, and since no practice can survive without an institution the ‘ideals and the creativity of the practice are always vulnerable to the acquisitiveness of the institution’.²¹³ MacIntyre would argue that the lawyer as Manager (herein the ‘lawyer-manager’) is the role the modern law firm asks of its members. The lawyer-manager meters the day against a target, runs the matter strictly to budget and a realisation rate, and treats the client’s instructions as a service bought at a price. A lawyer can play that role and still care about his or her work, but it is through that performance of that role that the acquisitiveness of the legal profession dominates.

4.2 Legal ethics has lost its virtue

Read in MacIntyre’s terms, the previous half century of regulatory deconstruction of duty and virtue is the smothering of the practice and its internal goods by institutions and external goods. The profession’s self regulation became suspect. A profession historically resistant to fees became an industry of paid service providers. The lawyer-statesman gave way to the lawyer-manager. The narrative unity of a legal life was broken up by the billable hour, specialisation and by the legal talent market. The practitioner’s working life, MacIntyre would argue, has become a run of separately metered tasks rather than a whole in which a disposition could form. The formative culture accumulated over two thousand years was set aside, leaving the practitioner with the form but not the substance of her professional virtue. MacIntyre’s prediction for a society in which external goods have become dominant is that ‘the concept of the virtues might suffer first attrition and then perhaps something near total effacement, although simulacra might abound’.²¹⁴ The present legal ethics codes, MacIntyre might argue, are such simulacra, retaining the words ‘candour’, ‘independence’ and ‘loyalty’ while the environment that gave the words their substance has gone. His account of justice takes this further. Where a society no longer shares a conception of the good, it loses any agreed measure of desert, because ‘the notion of desert is at home only in the context of a community whose primary bond is a shared understanding … of the good of that community’.²¹⁵ What is left is the management of conflict between incommensurable ideas of justice. The legal order becomes merely a way of keeping the peace among groups who do not agree, such that ‘modern politics is civil war carried on by other means’.²¹⁶ MacIntyre would say that the legal profession has lost the practice, the formed life and the tradition that gave it meaning, and that it is presiding over a legal order suffering from the same affliction. MacIntyre’s own answer to this state of affairs is the recovery of community, ‘the construction of local forms of community within which civility and the intellectual and moral life can be sustained through the new dark ages which are already upon us’.²¹⁷ In short, MacIntyre’s answer to this affliction is to return to communities and cultures sufficiently small and homogeneous such that practices, culture and, ultimately, virtues, can be constituted and truly inhabited. Albeit he is pessimistic about the prospects of doing so in our current (in his words) ‘dark ages’.²¹⁸

While MacIntyre’s account of the genealogy of modern ethics provides a compelling way to think about the history and current state of legal ethics, I would argue that both his account of justice and his answer to his diagnosis of modern ethics cannot be accepted. Interestingly, the primary issue I have with MacIntyre is similar to the issue I had with Bentham, in some ways MacIntyre’s polar opposite. I would argue that MacIntyre’s account of justice and his solution to the condition of emotivism are incompatible with pluralistic liberal democracy. MacIntyre would probably agree with a version of that claim, insofar as ‘pluralistic liberal democracy’ is taken to be something akin to that outlined by Rawls in A Theory of Justice,²¹⁹ or by Robert Nozick in Anarchy, State, and Utopia,²²⁰ both of whom MacIntyre uses as foils to distinguish his neo-Aristotelian account of justice.²²¹ To put it simply, MacIntyre does not find the move from tradition to universals, perhaps best typified by Kant, an intellectual ancestor of both Rawls and Nozick, compelling. In Whose Justice? Which Rationality? MacIntyre argues that rationality is internal to traditions, and that the moral schemes are so bound up with their own languages and forms of life that there is no neutral, independent standpoint from which their claims can be adjudicated. To come over to a tradition is less a matter of being persuaded than of undergoing something like a conversion, the remaking of an identity formed in one tradition into an identity formed in another.²²² He does not think that you can have a coherent account of justice outside of the context of a particular way of life lived in a particular community with a particular history and culture. MacIntyre is keen to point out that this is a different approach from most modern conservatives.²²³ This is why MacIntyre uses the figure of St Benedict, a quiet, pious, traditional monk, as a reference for surviving our current ‘dark age’.²²⁴ I would argue that the problem with this approach is that it, perhaps somewhat ironically given MacIntyre’s mode of critiquing other philosophies, has few practical outcomes for the particular lives, in particular communities, with particular histories and cultures of people living now. We now live in pluralistic liberal democracies with globalised economies and cultures. Indeed, we have fought bloody wars, and sacrificed and built much, to do so. Nothing is stopping specific individuals from taking a Benedictine approach in response to modernity, but I would suggest that whether or not modern pluralistic liberal democracy is suited for cultivating justice and virtue is the wrong question for the real individuals and societies of modernity. The question is how justice, virtue, and a substantive legal ethics are possible within pluralistic liberal democracies in the lifeworld of modernity.

4.3 Legal ethics can find new virtue

A black-and-white photograph. Habermas, in glasses and a dark jacket, holds his notes and raises a pen as he speaks into a standing microphone; young men and women stand close around him, listening.

Jürgen Habermas offers a potential way forward, particularly in Between Facts and Norms.²²⁵ Habermas begins from a somewhat similar diagnosis to MacIntyre’s, observing that ‘positive law and postconventional morality emerge co-originally from the crumbling edifice of substantial ethical life’, such that the shared ethos which once held communities together has been broken into its constituent elements.²²⁶ In an earlier text Habermas had rejected MacIntyre’s solution, arguing his account of rationality is both self-contradictory and ill-suited to the way human understanding works. It is self-contradictory, Habermas argues, because MacIntyre states his communitarian thesis in the supposedly community-neutral idiom of rationality. It is ill-suited because the barriers between different communities are more permeable than MacIntyre makes out. Habermas argues that many people do in fact move between traditions and acquire a ‘flexibly maintained identity’.²²⁷ Habermas argues there is a shared core beneath this plurality, since ‘concepts such as truth, rationality, and justification play the same role in every language community, even if they are interpreted differently’.²²⁸ Were MacIntyre correct that rival traditions share no rational ground, a plural society could never make legitimate law together, and a retreat into communities small enough to maintain a single moral order would indeed be the only option. However, Habermas argues, the presuppositions of argument are common to any tradition that lays claim to truth, such that a society need not agree on a conception of the good to agree on the procedures by which it makes its law.

Habermas’s answer to MacIntyre’s diagnosis of post-traditional morality is to try to show that law can perform the role that a shared ethical life once did, not by restoring lost agreement about the good but by providing a procedure through which people who disagree about the good can make law together. Modern law is coercive, a body of commands backed by sanction and alterable at the will of a legislator. It also claims to be legitimate, to bind its subjects as free and equal persons who could have rationally agreed to it. It is the one medium that holds these two things together, binding people by both ‘the threat of external sanctions’ and by ‘the supposition of a rationally motivated agreement’.²²⁹ This double character is why the attitude a person takes towards the law matters. Treated strategically, law is mere coercion, a set of penalties to be weighed in deciding whether to comply. Treated as binding among free and equal individuals, it is a body of legitimate norms which are obeyed ‘out of respect for the law’, and it retains that legitimacy only so far as it maintains that respect.²³⁰ Underpinning that legitimating attitude toward the law is what Habermas calls the discourse principle: ‘Just those action norms are valid to which all possibly affected persons could agree as participants in rational discourses’.²³¹ In such discourse law is the core principle of democracy, insofar as ‘only those statutes may claim legitimacy that can meet with the assent of all citizens in a discursive process of legislation that in turn has been legally constituted’.²³² Therefore, the legitimacy of a norm lies not in the good it produces but in whether those bound by it could have agreed to it. Nor does it descend from a higher moral law, as it did for Kant, who subordinated the juridical law to the moral law, at least on Habermas’s reading of him.²³³ Habermas treats law and morality as co-original and complementary, ‘two different but mutually complementary kinds of action norms’, with neither subordinating the other.²³⁴ At the core of his account is the requirement that the addressees of the law be able to understand themselves as its authors, that ‘the addressees of law are simultaneously the authors of their rights’.²³⁵ This means that private autonomy, the rights that preserve individual freedom, and public autonomy, the procedures by which citizens make laws, are not rivals but conditions of one another.

Habermas thinks that this account of law and procedure in a pluralistic liberal democracy is what gives both the substance to replace the traditional underpinnings of ethics and justice, and thereby escape emotivism and legalism. MacIntyre has a relatively easy riposte to this, namely that the actual state of modern moral discourse and deportment shows that law and procedure have plainly failed to perform that role. One can see the contours of this disagreement in how each of them handles Ronald Dworkin. Habermas offers a critique of Dworkin’s Hercules, the judge who reconstructs the community’s principles single handedly.²³⁶ Habermas argues that the work Dworkin imagines Hercules to do alone is, in truth, a ‘common undertaking supported by the public communication of citizens’.²³⁷ Even Hercules, who can reconstruct the whole institutional history of a legal order and settle on the single appropriate norm in every case, would still be deciding alone. On Habermas’s account it is Hercules’ solitude, not any shortfall in his capacity, that is his flaw. Legitimacy is not a property of the answer but of the process by which the answer is reached. I would briefly note that while this helps to illustrate Habermas’s point, his reading of Dworkin is perhaps uncharitable. Law’s Empire is opaque at times, but I would suggest that, read charitably, Dworkin does not introduce Hercules as a model of legal reasoning in the sense Habermas suggests, but rather as a regulative ideal, not to show how judicial reasoning should work in practice, but to illuminate how judges might think about integrating positive law and moral precepts when exercising roles in specific procedures.²³⁸ This leads us to MacIntyre’s critique of Dworkin, which is that modern judges cannot integrate law and morality in the way Dworkin suggests, rather they can only play the role of peacemakers between incompatible worldviews.²³⁹ At times, MacIntyre’s vision of law in a modern liberal democracy seems to be that it is a bare scaffold in place to prevent and resolve conflicts, a kind of legalism to put it in the terms of legal theory. But to view MacIntyre as having any fleshed out theory of law is probably misleading, as MacIntyre’s real point is that morality can only be properly understood within a specific culture and time. MacIntyre takes his theory of morality to suggest Dworkin’s theory of law is incorrect. And he would likely give a similar response to Habermas, namely that it is not possible for the law and procedure of liberal democracies to replace the role played by practice and virtue in traditional societies, as law and procedure cannot play the cultivating role previously played by practice and virtue in either an individual life, or that of a community.

It is perhaps worth stepping back and re-contextualising Habermas’s arguments within the story of legal ethics I have been telling. I am saying that both MacIntyre and Habermas agree that the current palimpsest of legal ethics is a somewhat confused mixture of virtue, duty and consequence, largely the product of a history of the severance of legal ethics from the circumstances that gave it meaning. I am also suggesting that, unlike MacIntyre, Habermas would seek to ground contemporary legal ethics in the processes of pluralistic liberal democracy, which can mediate coercion and liberty in such a way as to allow for a substantive legal ethics to exist. Habermas is offering law as a modern system of legitimation to prevent legal ethics from descending into a state of pure legalism. As to how an ethics is actually cultivated in real discourse, in real procedures, led by real people, Habermas grants that law ‘must draw on sources of legitimation that are not at its disposal’, on a spontaneity that ‘cannot be compelled simply through law’ but ‘is regenerated from traditions and preserved in the associations of a liberal political culture’.²⁴⁰ Furthermore, Habermas makes truthfulness and sincerity presuppositions of genuine discourse, but says little about how such, for want of a better word, virtues, can be fostered in a pluralistic liberal political culture.²⁴¹ While I think that Habermas makes a good case for how post-traditional norms can be legitimated, I’m not so sure he makes a good case for how post-traditional ethics can be cultivated, and thereby produce a sustaining environment for the legitimating procedures at the centre of his approach, which he concedes is necessary. One is tempted to take MacIntyre’s approach to cultivation and combine it somehow with Habermas’s legitimation in a pluralistic liberal democracy and thereby try to render a coherent philosophy for legal ethics for our present and future. But, regardless of whether such an attempt would result in Pygmalion’s Galatea or in Frankenstein’s monster, I think it would be a mistake for the same reason I think Kronman erred in trying to recreate the lawyer-statesman for the lifeworld of the 1990s through the medium of philosophy. However deftly it achieves its synthesis, a philosophy in dense prose, unmoored from practical realities, can only hope, at best, to convince other philosophers. Instead, I will try to use these philosophies to try to illuminate the possible futures of the legal profession in the face of AI.

5. How to address lawyers’ problem

No form is fixed. Nature, the innovator
of things, refashions new shapes out of old.
And, trust me, nothing in creation dies.
It varies and transforms. What we call ‘birth’
is just the start of being something different,
while ‘dying’ is the end.²⁴²

5.1 New forms of legal ethics

As I discussed in the essay ‘Artificial General Intelligence for Law’ (AGIL), the expansion of law’s empire has put the law beyond what practitioners, either individually or collectively, can master at the speed and scope modern society demands.²⁴³ Artificial systems performing real legal reasoning are the only approach that can realistically match the scale of that problem. Habermas’s account of law provides a useful complement to that argument, particularly for the purposes of legal ethics. For Habermas, positive law is not merely a body of valid norms, but simultaneously a system of knowledge and a system of action. For Habermas, law is symbiotic with post-traditional morality by relieving individuals of the cognitive burden of determining what abstract norms require in particular cases, the motivational burden of acting upon moral insight, and the organisational burden of coordinating action across a complex society.²⁴⁴ The scale and complexity of modern law now subjects legal institutions to an analogous burden. A norm can be democratically legitimate and publicly promulgated but still fail to guide conduct if actors cannot reliably determine what it requires. Consistent with this Habermasian vision, an AGIL can be understood as a functional complement to positive law. It does not supply legitimation to law, which remains grounded in discursive and democratic processes. It supplies part of the cognitive and organisational capacity through which legitimate law can become accessible, applicable and effective at scale. The increased compliance made possible by AGIL cannot make a norm legitimate, but it can give a legitimate norm the practical efficacy needed to stabilise expectations, coordinate action and sustain a functioning legal order. An AGIL could stand in relation to positive law as positive law stands in relation to post-traditional morality by providing the practical means by which normatively grounded demands can operate in a complex society.

In this Habermasian synthesis, the consequentialist provisions of codes of legal ethics could be translated into safety requirements for an AGIL. The emerging regulation of autonomous driving systems (ADS) provides a useful case study here. As I discussed in my essay on AGIL, ADS safety is treated as a property of the whole sociotechnical system, encompassing neural networks, interfaces, operators, the deploying entity, monitoring arrangements and the institutional culture.²⁴⁵ An analogous regime for AGIL might identify harms, specify controls and fallback procedures, monitor performance and trace responsibility across the persons and components involved. ‘Safety’ would remain a normative concept because identifying harms, fixing levels of risk and resolving conflicts within such a system requires judgment. The Habermasian synthesis supplies the division of labour within such a system. Law and legally institutionalised discourse determine the norms and interests to which the system must be responsive, and a consequentialist analysis tests whether the system reliably protects them. The pluralism objection to utilitarianism would be circumscribed by the fact that utility is not supplying the ends of the system. Moreover, the risk to the character of the actor does not arise in the same form for an artificial component, which has no character from which it can be alienated. Conversely, the concerns of character remain fully applicable to the lawyers and institutions that design, deploy and act upon the system, indeed even more so. This kind of systems analysis would make the division of human and machine responsibility clearer. Moreover, if my argument about a systems approach to legal alignment from the AGIL essay is combined with the argument I am making here, consequentialist legal ethics becomes an aspect of AI alignment.

5.2 New forms of lawyers

I would suggest that a version of that Habermasian vision is necessary in some form for the consequentialist aspects of legal ethics to have both efficacy and legitimacy in the age of AI. It is necessary, but not sufficient. I think Habermas’s failure to compellingly articulate how the resources needed to sustain his ‘discourse’ can be maintained without structural cultivation of the ‘traditions’ and ‘culture’ underpinning them needs a remedy. At this point, I would bring back into focus the brief history of legal ethics that I sketched in the first section of this essay. The first thing I would observe is that there is no single tradition of legal ethics to revive or return to. Part of the reason I have been using the terms phronēsis, prudentia, and Urteilskraft, instead of grouping them all within the English word ‘judgment’, is because, in the history of the legal profession, there is no single concept of judgment with a crisp definition. Each term which we might translate to the English word ‘judgment’ has to be viewed in the context of its place and time. Similarly, the Inns of Fortescue’s time, or the German university system James Morgan Hart attended, or any other locus of the cultivation of lawyerly judgment has to be seen in the context of their place and time. It does not follow from this that different streams of traditional ethics are incommensurable, or that legal ethics is inherently relative. In the semantic register it means that the different instances of terms like ‘judgment’, or ‘guild’, or ‘university’, have a ‘family resemblance’ in the way Wittgenstein discusses in the Philosophical Investigations.²⁴⁶ In the historical register it means that each time, place and phenomenon needs to be treated seriously within its own context, while at the same time recognising that there are similar behaviours, practices, and ideas that run through those contexts. Indeed, MacIntyre’s description of a living tradition as an ‘historically extended, socially embodied argument’ supplies a useful philosophy of history,²⁴⁷ provided that the argument is not confined within impermeable borders, or tied to MacIntyre’s own Thomistic worldview. The soul of the legal profession is not to be found in specific forms, but in recurrent questions such as how judgment is cultivated, how lawyers are disciplined, and how public duty is reconciled with money and power. The past does not offer a model to restore. It offers a substance from which the profession’s next form may be shaped.

What then of the thought that the material realities of our time have fundamentally changed the profession’s relationship with its inherited traditions and virtues? The relationship between the profession and the world has indeed changed, but successive forms of the Western legal profession have always been shaped by the tension between money, power and ideals of virtue. The ascendancy of the lawyer-manager since the 1970s is novel not because lawyers have begun to use their legal knowledge and institutional position to acquire wealth and status, but because the pursuit of external goods has become systematically embedded in the organisation and the measurement of professional legal work. The profession’s history of metamorphoses supports the presumption that this current arrangement will also give way to another form. The presumption that the current arrangement will continue indefinitely mistakes a historically contingent response to the scale, specialisation and competitive pressures of global capitalism for their necessary consequence. Kronman’s The Lost Lawyer appeared in 1993, one year after Francis Fukuyama’s The End of History and the Last Man.²⁴⁸ Fukuyama was read as giving expression to a post-Cold War mentality in which the combination of global capitalism and liberal democracy appeared to have outlasted its ideological rivals and become the final horizon of human development. Fukuyama’s argument entered the shared public imagination through the phrase ‘the end of history’, which became shorthand for the supposed inevitability of this seemingly stable settlement. Kronman’s lawyer-manager can be read as the last lawyer at the end of history, as it was popularly imagined at the time, materially successful but deprived of the ideal that once gave his work meaning. I suggest that Kronman’s pessimism about the lawyer-manager’s apparent permanence should be seen in the context of the then-apparent permanence of the wider political and economic order that had produced him. From the vantage point of 2026 however, the post-Cold War neo-liberal settlement no longer carries the same aura of inevitability. The sense that the lawyer-manager represents the final form of the profession is perhaps likewise better understood as a product of its time and place. The stronger historical presumption is that the profession will undergo another metamorphosis and that the lawyer-manager is not the last lawyer.

As to what changes for lawyers may be coming, AI may shape the lawyer’s next metamorphosis by making the economic distinction between technē and phronēsis more visible. Let’s assume for a moment that an AI becomes consistently as capable as the best lawyer at legal analysis, so much so that some neo-Aristotelian is willing to call technē. But technical excellence does not evolve by accretion into phronēsis. Phronēsis is not excellent expert reasoning reserved for especially difficult problems. It is the practical wisdom an individual gains through habituation, and whose outcomes are oriented towards the goods of a human life. An artificial system may reproduce the reasons in which such judgment is expressed, but it does not thereby acquire a character relevant to the cultivation of virtue. If AI makes the automatable components of legal analysis abundant, it follows that the scarce complement will be the capacity to decide what should be done with the outcomes of those components. That scarcity has economic significance because a liberal legal order cannot treat legal output as self-legitimating. To give an example by way of contrast, experienced software developers can now command a premium in the technology talent marketplace by building and maintaining sophisticated agentic development workflows. Even if that work were subsumed by a future advanced AI in the technology industry, there are reasons to think the same would not apply mutatis mutandis to the legal profession. There often aren’t ‘right answers’ in law like there often are in technology. A future AI system may be able to perform senior roles in software development, perhaps because many engineering goals can be tested against functional criteria. In contested legal matters however, the criteria of a correct resolution are often central to the case. The legal profession’s opportunity in the age of AI is not merely to preserve a residue of work that machines cannot yet perform. It is to breathe new life into the traditions and institutions through which practitioners learn to cultivate judgment, and deliberate between contestable goods. In short, AI may make cultivated judgment the scarce complement to machine competence.

The same distinction between machine competence and cultivated judgment places the billable hour under real pressure. Since hourly billing became the organising principle of commercial practice in the second half of the twentieth century, it has been continually criticised.²⁴⁹ Fixed fees, retainers and other alternatives have long existed, but they have not displaced hourly billing as the dominant medium of valuing legal work.²⁵⁰ Nevertheless, if legal analysis can indeed be decomposed into auditable components and some of those components can be mastered by an artificial system, the production of legal value no longer requires human time in rough proportion to its quantity. Under hourly billing, every human hour an AI system saves is an hour that can’t be billed for that task. Under fixed, subscription or outcome based pricing, the firm or organisation can pass part of that efficiency gain to the client or its own bottom line. Where machine and human outputs are genuinely substitutable, no hourly rate can make human labour competitive with a system operating continuously at scale. This does not mean that the billable hour will disappear. Rather, it would mean that it is confined to work in which the lawyer’s attention, judgment or assumption of responsibility is itself what the client is buying. As I write this essay, an agentic software development system I built is writing, testing and integrating the components of an AGIL based on instructions I gave it, and it will continue after I leave my desk, often for days at a time without my input. The relevant economic unit is not each hour I spend, but the value created by the system which produces the outputs. A software product can serve additional users at low marginal cost, as such returns need not track the engineering time embodied in each use. A law firm able to scale the technical parts of its work in the same way would need a business model that captures the value of the system without treating each output as though it represented a new block of professional time. Its lawyers would create their distinctive value elsewhere, through the phronēsis, prudentia or Urteilskraft by which the system is governed. Regrounding legal ethics in cultivated practice need not entail a sacrifice of economic value.

Treating cultivated judgment as a defensible value proposition may appear to subordinate cultivated judgment to the external good of money. That appearance rests on conflating payment for a practice with allowing payment to define its end. A lawyer may be paid for judging well without payment supplying the only standard of what counts as good judgment. Moreover, the profession has never possessed the innocence or purity for which its elegists sometimes seem to mourn. As we saw, Cicero indirectly compensated his services with legacies and loans, medieval guilds joined duties to the defence of professional privilege, and the so-called lawyer-statesman of the nineteenth century joined civic virtue to wealth and political power. What has changed is not the presence of money, power or hypocrisy, but the scale and consistency with which professional institutions organise work around external ends. Accounts of lost virtue can sometimes mistake an age’s account of its ideals for an account of how people actually operated. Lawyers have always professed standards that their livelihoods gave them incentives to evade. Moreover, hypocrisy has always been native to the philosophy of virtue. Aristotle’s treatment of truthfulness, boastfulness and irony can be extended mutatis mutandis to hypocrisy.²⁵¹ Judith Shklar warns against making hypocrisy the primus inter pares of the vices, because anti-hypocrisy can itself become cruel and domineering.²⁵² David Runciman offers an institutional account of the same point.²⁵³ The exercise of public authority requires some form of representation. Politicians speak for the people, judges for the law and lawyers for client and court. Each role necessarily places a fallible person behind a public mask. For Runciman, the mask is not itself corrupt, rather corruption begins when the mask is used to evade the standards that make the role legitimate. Legal ethics should therefore not imagine a profession without hypocrisy. Indeed, such a profession has never existed. Rather, the profession should cultivate offices, habits and institutions that make hypocrisy visible and criticisable before it hardens into fraud or domination.

5.3 New forms of legal institutions

The institutions in which lawyerly judgment was cultivated also marked lawyers off as an elite, so any appeal to those traditions inevitably raises the spectre of elitism, particularly in tertiary education. Nevertheless, I would observe that top law schools have largely abandoned the language and curriculum of virtue and judgment without thereby becoming less elite. Elite status rests on an institution’s power to select, credential and distribute access to money and power. The postwar expansion of tertiary education widened participation in the sector, but it also enlarged the credentialing system and moved competition into hierarchies among institutions and qualifications, leaving the most selective law schools with a disproportionate role in distributing access to the most lucrative and powerful parts of the profession.²⁵⁴ Modern law schools attempt to combine wider participation with a meritocratic approach to selection and an increasingly vocational justification for legal education. However a meritocratic, progressive or anti-elitist self-conception does not alter the structural position of these institutions. Indeed, it can sharpen the impression of hypocrisy when an institution presents the status and opportunities it distributes as purely the products of merit.²⁵⁵ Moreover, AI may now unsettle the vocational aspects of that settlement. If a legal credential is justified principally by the certification of technical competence, then AI systems capable of performing an increasing share of technical legal tasks will make that credential less valuable. The Habermasian symbiosis between human and machine sketched above suggests an alternative basis for that vocational value. The practitioner’s distinctive contribution may not be technical competence, but the cultivated judgment by which the system is governed. In such a future, a renewed concern with virtue and judgment would not amount to restoring the social hierarchies once associated with them. Their study and cultivation are not what made law schools elite, and moving away from them did not make legal education egalitarian. The cultivation of judgment and virtue might instead become central to legal education precisely because the technical competence that has underpinned the modern law degree’s vocational value will increasingly be supplied by AI systems.

The spectre of elitism also hangs over professional associations, since the Inns, bars and guilds through which lawyerly judgment was historically cultivated were also institutions of status, exclusion and privilege. A revival that simply restored old traditions, new leather upholstery on old Chesterfields, would revive that spectre. As I hope I have shown, the cultivation of judgment has never belonged to a single institution or taken a single form. Blackstone’s lecture ‘On the Study of the Law’ is a paradigmatic example in this respect, as it signalled an attempt to bring university learning into a profession that was, at that time, formed principally through the Inns. Different jurisdictions have continued to allocate the formation of lawyers differently between universities, professional bodies and practice. A living tradition can survive through such changes rather than through the reproduction of any one of them. Hussey Freeland’s brief attempt to reimagine phronēsis through gendered social formation shows how an inherited ideal can be continued through a reimagining of the social form in which it was received. The words ‘virtue’, ‘judgment’ and ‘tradition’ may evoke nineteenth century ideas of Christian moralism, romantic historicism or aristocratic patriarchy, but those associations belong to specific historical embodiments rather than to the concepts themselves. The histories of phronēsis, prudentia and Urteilskraft and their broader semantic brethren reveal repeated efforts to give judgment and its kin forms suited to different times and places. The history of the legal profession is not a single tradition with a fixed institutional home, but several traditions and several socially embodied arguments about the purposes of legal practice and the judgment its different forms require. The locus of that formation will depend upon place. In one jurisdiction it may lie primarily within universities, in another within professional associations, and in another within firms, chambers, or some combination of them. The question is not whether a particular institution should recover its ‘grand’ past, but whether key institutions in each place can continue these socially embodied arguments in forms suited to our age.

Similarly, what we have been calling the ‘lawyer’ has never embodied a single tradition of legal practice. Roman legal practice had never centred on a single figure. Pontifical jurists gave way to secular iurisconsulti (‘jurisconsults’)jurisconsults , while orators such as Cicero remained advocates.²⁵⁶ The institutions that sustain legal practice have always cultivated several forms of lawyerly excellence. Bruce Ackerman’s Reconstructing American Law envisaged an interdisciplinary reconstruction of law in which economics and other bodies of knowledge entered legal argument.²⁵⁷ Kronman uses Ackerman’s multidisciplinary social engineer as a foil and argues that the lawyer’s distinctive expertise is the art of handling cases.²⁵⁸ I would argue that Ackerman’s and Kronman’s different approaches are better understood as different disciplines of legal practice rather than as rival ideals for the whole profession. The lawyer handling cases needs the social, historical and imaginative breadth Kronman describes to perceive what a particular dispute contains, or to work symbiotically with an AGIL in a Habermasian synthesis like I’ve described above. Such a lawyer would probably benefit from the case method of education, as Kronman suggests, at least in common law jurisdictions. They may also benefit from other methods and institutions pertinent to the traditions of advocacy in other cultures and places. The interdisciplinary lawyer, in contrast, needs a sufficiently deep formation in law for knowledge drawn from economics, computer science or another discipline to become part of legal judgment, or vice versa. I would note that neither Kronman’s advocate nor Ackerman’s social engineer can be formed through a general technical education in ‘law’ that plods through black letter subject by black letter subject, like chapters in a textbook. That’s how you might program a computer, not how you cultivate a lawyer. Both the advocate and the social engineer depend upon a broad education, the mixture of which would likely be different in different places. To offer a few sparks to light the imagination in this respect, James Boyd White approaches legal education through comparisons between law and literature,²⁵⁹ and Maksymilian Del Mar’s recent Artefacts of Legal Inquiry traces the role of imagination in adjudication.²⁶⁰ As for the multidisciplinary social engineer, there is nothing more topical than the legal alignment of AI, which requires an understanding of law and legal systems, and enough working knowledge of technology and AI to translate between legal and computational descriptions, without being overly reductive one way or the other. Legal alignment requires a balanced intellectual and practical approach across two disciplines, which inherently involves questions of judgment. If legal alignment is to be realised, it won’t be by turning lawyers into computer scientists or by turning computer scientists into lawyers, it will be by balancing empiricism with idealism, theory and practice, jurisprudence and specification, law and technology. Such balancing requires both depth and breadth. It requires judgment. The institutional question is not which form of legal practice should displace the others, but what intellectual and practical grounding would allow these different forms of lawyerly excellence to develop.

6. Conclusion

My own cultivation of practice is more of the interdisciplinary variety, studying and working in law for a decade, then working in technology for a decade, and now working on the combination of both. In my experience, the intellectual and practical crossover between the two domains is less of a discontinuity than the social and institutional distance between them might suggest, as the same habits of mind are at work in both. There is the systems level grasp of how parts work within a whole, how a clause sits inside a contract, how a provision sits in statute, how a statute sits in a regulatory regime, how a function sits inside a program, and how a program sits inside an architecture. There is also a syntactical precision required by both. Both require careful reading of what sentences or functions express, reasoned inference from premises to consequences, and comprehension of how a technical specification works and applies. These are capacities that a good legal education cultivates and that a lawyer who is both grounded and sufficiently flexible can apply to another discipline. Indeed, lawyers often pride themselves on their ability to comprehend non-legal domains, as their work sometimes requires it of them. If that mode of thinking is taken one step further, with appropriate flexibility and humility, they might find many opportunities arise to apply the skills of their practice in non-legal domains, while retaining a fidelity to that practice. I emphasise ‘humility’, not to make myself out as virtuous, but as a reflection on the many humbling experiences I’ve had navigating the technology industry. Being cultivated in a learned tradition that still attracts significant wealth and power naturally militates in favour of a degree of pride. Aristotle does not see pride as an inherent fault, nor humility as an inherent virtue. In Book IV of the Nicomachean Ethics he treats greatness of soul (megalopsychia), which we now may call a kind of pride, as a kind of crown or ornament of the virtues, the disposition of one who claims great honours and deserves them, while undue humility (mikropsychia), the claiming of less than one’s deserts, is a vice both commoner and worse than vanity.²⁶¹ What makes pride or humility virtuous is neither its greatness nor its smallness but its truth, the fit between what one claims and what one deserves. The lawyer who carries the honours of an ancient profession into a discipline in which she has yet to earn any is not proud in Aristotle’s sense but vain, which is why humility is necessary, not for its own ends, but to build a real practice that will, one day, make her proud.

The technology industry can be somewhat solipsistic in its outlook on other domains of the economy and society. Preparing ourselves for the future of the legal profession must involve understanding its social, political and intellectual history. You cannot build the future of something without knowing what it is in the first place. The legal profession has its own myopias. A respect for the wisdom of tradition and a preference for incrementalism should be grounded in an understanding of that incremental tradition, how it has changed, and what it has preserved through its changes, lest it become a reactionary and shallow conservatism. When any tradition is appropriately contextualised in its history and motivating philosophies it is invariably revealed to be more flexible and nuanced than it may superficially seem. Moreover, such an intellectual archaeology often reveals what initially seemed to be tensions as opportunities in disguise. In public discourse, artificial intelligence is often framed as a threat to the legal profession. I have not set out to prove the converse, but when considering the history of legal ethics, whether in the way I have done so here, or in the myriad other ways the philosophy and history of the legal profession can be constructed, artificial intelligence can plausibly be seen as an opportunity to resolve some of the tensions the profession has borne for some time. This is not to diminish the risks artificial intelligence poses, both to the practice of law and to legal ethics. What such an exercise does attempt to do is engage in a discourse aimed at reshaping the profession from the clay of its traditions so it can pursue new opportunities in whatever change may come.