Introduction to Law & Legal Reasoning

Law is a system of rules created and enforced by a recognized public authority to govern the conduct of a community. Many kinds of rules shape our behavior - manners, house rules, the rules of a game, moral principles - but what sets legal rules apart is that they are backed by the organized power of the state and applied through official institutions such as legislatures, courts, and the police. If you break a rule of etiquette, you may be thought rude; if you break a legal rule, you may be sued, fined, or imprisoned through a formal public process.

Each word in that definition is doing work. "Created" signals that law is deliberately made and changed by institutions, not merely inherited habit. "Enforced" signals that the rules have teeth: official consequences follow violation. "Recognized public authority" signals that the rules come from bodies the community accepts as entitled to make them, such as a legislature or a court. And "community" signals that law is social: it coordinates the conduct of many people who may never meet, which is something no private arrangement can do on its own.

A sorting exercise

Consider four rules you might encounter in a single week. First, a chess club's bylaw that members must arrive ten minutes early. Second, a family's rule that phones go away at dinner. Third, a posted speed limit of 30 miles per hour. Fourth, a national requirement that residents file a tax return each year. All four are genuine rules: they guide conduct, and breaking them brings consequences of some kind.

Yet only the third and fourth are law. The club and the family made their rules privately, and the worst official consequence of breaking them is private displeasure or expulsion from a voluntary group. The speed limit and the tax rule were made by public bodies with recognized lawmaking power, and breaking them triggers public machinery: a citation, a fine, an audit, ultimately a court. The difference is not seriousness or even importance - a family rule may matter more to you than a parking rule - but source and enforcement.

Three features of a legal rule

Scholars often point to a few features that make a rule genuinely legal rather than merely social:

• Authority. The rule comes from a source recognized as entitled to make law, such as a constitution, a legislature, or a court, rather than from private preference.
• Generality. Law typically applies to categories of people and situations ("drivers," "sellers of goods") rather than naming one person, and it is meant to be applied consistently.
• Enforcement. Legal rules carry official consequences - remedies, penalties, or coercion - that public institutions will actually impose.

Authority is a question of pedigree. When someone claims a rule is law, a lawyer asks where it came from: was it enacted by the legislature, adopted by an agency acting under a statute, or announced by a court with power to decide the question? A rule with the right pedigree counts as law even if it is unpopular; a rule without it is not law no matter how widely admired. Later lessons trace these pedigrees in detail, from constitutions down to individual court decisions.

Generality connects law to fairness. Because legal rules speak in categories, they commit the state in advance to treating like cases alike. A rule that said "Jordan Smith shall pay a special tax" would strike most people as an abuse precisely because it singles out a named person rather than defining a class. Many constitutions restrict such targeted enactments. Generality also makes law learnable: you can know your obligations as "a driver" or "an employer" without knowing which official will apply them to you.

Enforcement is what makes legal duties more than advice. But enforcement takes many forms beyond punishment. A court may order a contract breacher to pay damages, an agency may revoke a license, a registry may refuse to record a defective deed. Sometimes the "sanction" is simply nullity: if you sign a will without the required witnesses, no one fines you, but the document fails to do what you wanted. The common thread is that public institutions stand ready to give the rule practical effect.

Rules that empower rather than command

The nullity example points to something important. Much of law does not order anyone to do anything; it confers powers. The law of contracts lets you create binding obligations that did not exist before. The law of wills lets you direct what happens to your property after death. The law of corporations lets people create a new legal person that can own assets and sue and be sued. These are facilitative rules: recipes that, if followed, change your legal position.

The legal philosopher H. L. A. Hart made this distinction central. He called duty-imposing rules primary rules and rules about rules - rules for making, changing, and applying other rules - secondary rules. A society with only primary rules, he argued, would face constant uncertainty about what the rules are, no way to change them deliberately, and endless disputes about violations. Legal systems solve these problems with a rule of recognition that identifies valid law, rules of change such as legislative procedure, and rules of adjudication that empower courts.

On this view, what marks a mature legal system is not merely that it issues commands, but that it contains this union of primary and secondary rules. That is why the question "is this rule law?" usually becomes a question about sources: does the rule pass the system's own tests for validity? Keep this idea in hand; it explains why Module 2, on sources of law, matters so much.

Law and coercion, law and consent

Because law can be enforced by force, some thinkers describe it mainly in terms of commands backed by sanctions. That captures part of the picture, but it is incomplete. In stable legal systems most people obey most law most of the time not because they fear punishment in each instance, but because they accept the system as legitimate - they treat the law as giving them genuine reasons to act. A legal system that rested on fear alone would be fragile and costly to run. Legitimacy, the widely shared belief that the law has a rightful claim to be obeyed, is what lets law function at low cost.

The command picture is associated with the nineteenth-century theorist John Austin, who defined law as the general commands of a sovereign, backed by threats, that people habitually obey. The theory has real strengths: it insists that law is a human creation and that enforcement matters. But it struggles to explain facilitative rules, which threaten nothing, and it struggles to explain why law persists when one lawmaker replaces another, since habits of obedience attach to persons.

Hart pressed a famous objection: a gunman who says "your money or your life" also issues a command backed by a threat, yet we would not say his victim has a legal obligation. Being obliged by force is not the same as being obligated by a rule. What the gunman lacks is the standing that comes from a system of accepted rules. Officials in a functioning legal system take what Hart called the internal point of view: they treat the rules as standards that justify decisions and criticism, not merely as predictions of what enforcers will do.

The rule of law

A related ideal deserves its own name. The rule of law is the principle that government itself is bound by law: officials may act only with legal authority, and the same courts that judge citizens can judge the state. The contrast is rule by law, where rulers use legal forms as tools while remaining above them. A classic English decision, Entick v. Carrington (1765), held that officers who ransacked a writer's home without lawful authority were ordinary trespassers; their government positions gave them no immunity. The case is remembered because it makes the ideal concrete: power needs a legal warrant.

Legal philosopher Lon Fuller usefully listed what the ideal demands of the rules themselves. Laws should be general rather than ad hoc; publicly available; prospective rather than retroactive; clear enough to follow; consistent with one another; possible to comply with; reasonably stable; and actually administered as written. A system that failed all of these - secret, retroactive, contradictory, ever-shifting decrees - would not merely be bad law; it would fail at being law at all, because no one could guide conduct by it.

These requirements have daily payoffs. Prospectivity means you cannot be punished for conduct that was lawful when you did it, a principle known as legality. Publicity and clarity let people plan: businesses sign contracts and families buy homes because the governing rules are knowable in advance. And congruence between the rules on paper and official practice is what makes legal rights worth having. When people say a country has a strong or weak rule of law, they are usually measuring these features.

Positive law and its limits

The rules actually laid down and enforced in a given place and time are called positive law - the law that "is posited," or put in place, by human authorities. Positive law can be written (a statute) or unwritten (a judge-made rule), but in each case it is identifiable: you can point to the source.

This is different from asking whether a law is good or just, which is a separate question we take up next. For now, the key idea is that law is a special kind of social ordering: authoritative, general, and enforceable, and it works best when people regard it as legitimate rather than merely as a threat.

Keeping the two questions apart is a professional habit worth forming now. A lawyer advising a client must first report what the positive law is, however she evaluates it; a citizen or legislator deciding what the law should be must argue about justice and policy, not just recite the rulebook. The two inquiries feed each other - criticism of existing law drives reform through legislation and litigation - but collapsing them produces confusion in both directions.

A day inside the legal system

It is easy to picture law as courtroom drama, but most law works silently in the background of an ordinary day. The milk you pour at breakfast met food-safety regulations. Your commute was structured by traffic law and vehicle-inspection rules. Your job exists inside employment law, wage law, and an enforceable contract. Buying lunch formed a small sales contract; your apartment rests on a lease and property law; the app you used tonight runs on licenses and data rules. None of this required a lawsuit. Law succeeded precisely because no dispute arose.

This background quality explains a common statistic of legal life: the great majority of legal relationships never see a courtroom. Litigation is the visible failure mode of arrangements that usually work. When you study cases in later modules, remember that each dispute sits atop millions of transactions that went smoothly because the rules were known and followed.

Common misconceptions

First misconception: law means criminal law. Television makes arrests and trials vivid, but crimes are one province of a much larger map that includes contracts, property, torts, family law, administrative regulation, and constitutional structure. Most lawyers never handle a criminal case.

Second misconception: if it is not written in a statute, it is not law. In common-law systems, binding rules also come from judicial decisions accumulated over centuries, and enforceable obligations arise from private instruments such as contracts and deeds that the law stands behind. Module 2 sorts these sources.

Third misconception: saying "there ought to be a law against that" tells you what the law is. It states a moral or policy view. The lawyer's first question is always the descriptive one - what do the recognized sources currently provide? - and only then the evaluative one. Confusing the two leads people to assume that anything unfair must already be illegal, which is often not true.

Try it: which of these rules is law?

Work through this fact pattern. A city ordinance requires owners to clear snow from the sidewalk fronting their property within 24 hours, on pain of a fine. A homeowners association's recorded covenant requires houses in the neighborhood to be painted in approved colors. A student film society requires members to wear black to screenings. A widespread local custom holds that you should tip food servers around 20 percent. Which of these rules are law, and in what sense?

The ordinance is law in the fullest sense: enacted by a public authority under delegated power, general in application, and enforced by fines through public process. The film society rule is not law: privately made, privately enforced, and the state is indifferent to it; the worst outcome is losing membership in a voluntary association.

The covenant is the instructive middle case. A homeowners association is not a legislature, yet covenants recorded against land can be enforced in court against buyers who never personally agreed to them, because property and contract law give such instruments legal effect. The rule is privately authored but publicly enforceable - law lends the association its machinery. The tipping custom, by contrast, is social practice only; it becomes law only if some rule incorporates it, as when a statute regulates how employers may count tips toward wages.

Notice what the exercise trains: you classified rules by asking about source and enforcement, not popularity or importance. That is exactly the analytical move this lesson has been building, and it is the first reflex of legal reasoning.

In sum, law is the subset of a community's rules that carries official pedigree, speaks in general terms, and can call on public institutions for effect. It commands, but it also empowers; it coerces, but it depends on legitimacy; and it includes rules for making and changing rules. With this working definition in place, the next lesson asks why societies build such a system at all, and what justice demands of it.

Sources

1. Legal Information Institute. (n.d.). Law. Cornell Law School. law.cornell.edu
2. Atiq, E., & Marmor, A. (2025). The nature of law. In The Stanford Encyclopedia of Philosophy. Stanford University. plato.stanford.edu
3. Green, L., & Adams, T. (2025). Legal positivism. In The Stanford Encyclopedia of Philosophy. Stanford University. plato.stanford.edu
4. Rundle, K. (2026). The rule of law. In The Stanford Encyclopedia of Philosophy. Stanford University. plato.stanford.edu
5. Hart, H. L. A. (1958). Positivism and the separation of law and morals. Harvard Law Review, 71(4), 593. doi.org/10.2307/1338225
6. Fuller, L. L. (1958). Positivism and fidelity to law: A reply to Professor Hart. Harvard Law Review, 71(4), 630. doi.org/10.2307/1338226
7. United Nations. (n.d.). What is the rule of law? United Nations and the Rule of Law. un.org
8. Entick v. Carrington, Court of King's Bench (1765). (Case discussed in the lesson; cited by name, court, and year only.) find source ↗

Why do communities create law at all? Imagine a group with no shared rules that anyone will enforce. Disputes would be settled by whoever is stronger, agreements would be unreliable, and cooperation among strangers would be risky. Law addresses these problems. Political and legal theorists usually identify several overlapping functions of law.

The thought experiment has a long pedigree. Thomas Hobbes argued in the seventeenth century that life without a common power to keep order would collapse into conflict, because even people of good will cannot trust strangers to hold up their end of a bargain. John Locke added that even a mostly peaceful society without law lacks three things: a settled, known rule to judge disputes by, an impartial judge to apply it, and a reliable power to enforce the result. On this social-contract picture, legal institutions are the machinery a community builds to supply exactly those missing pieces.

What law does

• Keeping order and resolving disputes. Law provides peaceful, public procedures - courts and rules of evidence - so that conflicts are settled by argument and authority rather than by force or private revenge.
• Protecting people and property. Criminal law and tort law define wrongs and provide protection and redress for harm to person, property, and reputation.
• Enabling cooperation. Contract, property, and commercial law let strangers make reliable promises, own and transfer things, and plan for the future, which underpins trade and everyday life.
• Allocating and limiting power. Constitutional and administrative law set up government, distribute authority among institutions, and constrain officials so power is not exercised arbitrarily.
• Expressing and channeling values. Law declares which conduct a society treats as acceptable or forbidden and can steer behavior toward shared goals such as safety or fairness.

Each function repays a closer look. Dispute resolution replaces the feud. Where legal remedies are unavailable or distrusted, people resort to self-help, and grievances escalate; a working court system converts a fight into a file. Notice that the point is not that courts always reach perfect answers. It is that they reach final, public, reasoned answers by a method both sides accepted in advance, which lets the parties and everyone watching move on.

The enabling function is the least visible and perhaps the most important. A farmer sells a harvest that does not yet exist to a buyer she has never met; a stranger lends money to another stranger for thirty years to buy a house. These transactions are rational only because contract and property law make the promises enforceable and the titles secure. Economists describe this as reducing transaction costs: law substitutes for the personal trust that strangers cannot have.

The power-allocating function turns inward, on government itself. Constitutions divide authority among institutions and levels of government precisely so that no official's will is the last word. This is law regulating the lawmakers, and it is what separates constitutional government from rule by decree. Module 2 and Module 3 examine this architecture in detail.

Coordination and the problems law solves

Some legal rules solve moral problems, but many solve coordination problems, where what matters is not which rule we pick but that we all pick the same one. Whether traffic keeps right or left is morally arbitrary; a collision is not. Standard weights and measures, the date taxes are due, the formalities for a valid deed - these rules create a shared grid that lets millions of independent plans mesh. Once the rule exists, each person has a reason to follow it simply because everyone else does.

Law also manages collective-action problems, where individually sensible choices add up to group harm. No single factory's discharge ruins a river, and no single fisher empties a fishery, yet together they can. Environmental statutes, licensing schemes, and tax-funded public goods exist because voluntary restraint tends to unravel when others cannot be counted on to match it. Seeing these structures helps you predict where law appears: wherever trust among strangers, standardization, or shared resources are at stake.

Law and morality

Law and morality overlap but are not identical. Much law reflects moral judgments - prohibitions on killing, stealing, and fraud track widely shared ethics. Yet the two can diverge. Some conduct is immoral but not illegal (breaking a private promise that is not a contract), and some conduct is illegal but not obviously immoral in itself (parking on the wrong side of the street).

Lawyers even have vocabulary for the second category. Conduct wrong in itself, such as murder, is called malum in se; conduct wrong only because a rule forbids it, such as driving on the wrong side, is called malum prohibitum. The distinction matters practically: for regulatory offenses, fair notice does most of the moral work, which is one reason such offenses are usually punished lightly and sometimes without proof of bad intent.

A long-running debate contrasts legal positivism, which holds that whether something is law depends on its social sources rather than its moral merit, with natural law theories, which hold that law is connected to morality and that grossly unjust rules lack full authority as law. You do not have to resolve this debate to see the practical point: identifying what the law is and judging whether it is good are two different tasks, and a careful thinker keeps them apart.

The debate sharpened after the Second World War, when courts and scholars confronted the legal systems of defeated regimes that had followed formally valid but deeply wicked rules. Positivists such as H. L. A. Hart argued that clarity is served by saying "this was law, and it was too evil to obey." Natural lawyers such as Lon Fuller replied that systems flouting law's inner requirements of generality, publicity, and non-retroactivity forfeit the name of law. Both sides agreed on the judgment of evil; they disagreed about the most honest way to describe it.

Unjust laws and legal change

History supplies unmistakable examples of valid-but-unjust law. In the United States, statutes once enforced racial segregation in schools, transport, and public places. In Plessy v. Ferguson (1896) the Supreme Court upheld such a statute under the formula "separate but equal"; in Brown v. Board of Education (1954) the Court repudiated that formula for public education, holding that separate educational facilities are inherently unequal. The pair shows both that legality and justice can come apart and that legal systems contain tools for correcting themselves, though often slowly and under pressure.

What may a person do about an unjust law while it stands? Legal systems channel disagreement into lawful routes: campaign to repeal, sue to invalidate, vote to replace the lawmakers. Some reformers have gone further, practicing civil disobedience: openly breaking the law, without violence, and accepting the penalty, in order to force public attention to injustice. Martin Luther King Jr., writing from a Birmingham jail, defended that practice while insisting that disobedience be open and its consequences accepted, precisely to show respect for law as such while condemning a particular law.

The takeaway for this course is analytical, not political. When you evaluate a rule, keep three questions separate. Is it valid law under the system's sources? Is it just? And what lawful mechanisms exist to change it? A great deal of confused argument collapses these into one.

Justice: procedure and substance

Law aspires to justice, which has two dimensions. Procedural justice concerns fair process: like cases treated alike, notice of the charges or claims, an impartial decision-maker, and a genuine chance to be heard. Substantive justice concerns fair outcomes and the content of the rules themselves - whether the law distributes benefits, burdens, and punishments fairly. A legal system can have fair procedures yet unjust rules, or good rules applied through unfair procedures. Because reasonable people disagree about what substantive justice requires, mature legal systems place great weight on procedural justice: fair, predictable, and public procedures that everyone can accept even when they dislike a particular result.

Procedural justice has a recognizable core across systems. The decision-maker should not be a judge in her own cause and should have no stake in the outcome. The person affected should receive notice: a statement of what is claimed or charged, early enough to prepare. There should be a genuine opportunity to be heard, to present evidence, and to answer the other side's evidence. And the decision should rest on stated reasons applied from pre-announced rules, so it can be reviewed. In American law these ideas travel under the name due process of law.

A famous illustration is Goldberg v. Kelly (1970), where the United States Supreme Court held that a state could not cut off a person's welfare benefits without first providing notice and an evidentiary hearing. The benefits were the recipients' means of survival, so the timing of the hearing - before termination rather than after - was itself part of what fairness required. The case shows that procedure is not an ornament; when process fails, the substantive right becomes worthless in practice.

Why fair process earns acceptance

There is a practical reason mature systems invest so heavily in procedure. Every dispute produces a loser, and a legal system survives only if losers, most of the time, comply anyway. Research on procedural fairness consistently finds that people's willingness to accept an adverse decision depends heavily on whether they believe the process was fair: whether they were heard, treated with respect, and judged by neutral rules. Fair procedure is thus not merely an ethical nicety; it is the mechanism by which a society converts raw disagreement into stable, accepted outcomes.

This also explains why courts explain themselves. A judgment that says "the plaintiff loses" settles the case; a judgment that states reasons submits itself to professional criticism, appellate review, and public evaluation. Reason-giving disciplines the decider and reassures the loser that the result came from the rule, not from favoritism.

Common misconceptions

One misconception holds that whatever is legal is therefore moral. The segregation statutes refute this in one direction; in the other, plenty of unkind, dishonest, or disloyal conduct remains perfectly lawful. Law sets a floor for conduct in a large, plural society; it does not exhaust ethics.

A second misconception holds that every moral wrong should be made illegal. Enforcement has costs: surveillance, error, selective application, and the loss of privacy and autonomy. Many systems deliberately leave large zones of private life to conscience and social norms, judging legal coercion too blunt for them. Deciding what to leave unregulated is itself a design choice about the proper reach of the state.

A third misconception treats procedure as a technicality, as when a case is said to be won or lost "on a technicality." Notice, an impartial judge, and a chance to respond are not technicalities; they are what distinguishes adjudication from decree. When you hear the phrase, ask which procedural safeguard was at stake and what its purpose is. The answer is usually more substantive than the phrase suggests.

Try it: the midnight park ordinance

Work this fact pattern with the lesson's tools. A city council, worried about vandalism, enacts an ordinance closing all public parks from midnight to 5 a.m. and imposing a modest fine for violations. The ordinance was published in advance and applies to everyone. Mara, walking home through a park at 1 a.m., is cited by an officer who tells her, incorrectly, that she has no right to contest the fine. Analyze the ordinance's functions, and then analyze the procedure applied to Mara.

Function analysis first. The ordinance serves order maintenance (deterring nighttime vandalism), protects public property, and coordinates use of a shared resource. It also expresses a community judgment about park use. It is general, prospective, and published, so it satisfies rule-of-law basics. Whether it is substantively wise is debatable - it burdens night-shift workers who cut through the park - and that debate belongs in the council chamber, where the rule can be amended.

Now procedure. The citation itself is lawful process, but the officer's statement misdescribes it. Procedural justice requires that Mara receive notice of the violation and a genuine opportunity to be heard by a neutral decision-maker before the fine becomes final. If the city in fact provides a hearing, the officer's misstatement is an error to be corrected, not a defect in the system; if the city provides none, the scheme fails procedural justice even though the underlying rule is valid and sensible. Substance and procedure must each pass inspection separately.

This lesson's map - functions, the law-morality relation, and the two faces of justice - underlies everything that follows. When later modules present doctrines of contract, tort, property, and crime, ask of each rule: what function does it serve, what moral judgment does it embody, and what procedures give it effect? Those three questions turn memorization into understanding.

Sources

1. Tyler, T. R. (2003). Procedural justice, legitimacy, and the effective rule of law. Crime and Justice, 30, 283-357. doi.org ↗
2. Miller, D. (2026). Justice. In The Stanford Encyclopedia of Philosophy. Stanford University. plato.stanford.edu
3. Goldberg v. Kelly, 397 U.S. 254 (1970). Legal Information Institute, Cornell Law School. law.cornell.edu
4. Plessy v. Ferguson, 163 U.S. 537 (1896). Legal Information Institute, Cornell Law School. law.cornell.edu
5. National Archives. (n.d.). Brown v. Board of Education (1954). Milestone Documents. archives.gov
6. Legal Information Institute. (n.d.). Procedural due process. Cornell Law School. law.cornell.edu
7. Legal Information Institute. (n.d.). Natural law. Cornell Law School. law.cornell.edu

Law does not come from a single place. In most modern legal systems it flows from several sources of law, which fit together in a ranked order. The two we cover first are the highest and the most familiar: constitutions and statutes.

Learning to ask "what is the source of this rule?" is the first professional habit of legal research. The source determines the rule's rank, who can change it, and how courts will read it. A right guaranteed by a constitution is sheltered from ordinary politics; the same right created only by statute lasts exactly as long as the legislature leaves it in place. Keep that difference in view throughout this lesson.

Constitutions

A constitution is the fundamental law that establishes a government, defines and distributes its powers, and usually protects the basic rights of individuals. It sits at the top of the legal order: ordinary laws that conflict with the constitution can be held invalid.

Some countries have a single written constitution (as in the United States or Germany); others have an uncodified constitution drawn from many statutes, judicial decisions, and long-standing conventions (as in the United Kingdom). Either way, the constitution does two big jobs. It empowers government by creating institutions and granting them authority, and it limits government by setting boundaries those institutions may not cross, such as guarantees of free expression, fair trials, or equal treatment.

Constitutions are also deliberately hard to change, a feature called entrenchment. In the United States, an amendment must be proposed by two-thirds of both houses of Congress (or by a convention sought by two-thirds of the state legislatures) and then ratified by three-fourths of the states; only 27 amendments have made it through. Entrenchment protects the ground rules from momentary majorities, at the price of making even popular corrections slow. Other systems entrench more lightly, requiring special majorities or referendums.

The United States example

Because this course draws many illustrations from American law, sketch the design once. The Constitution of 1787 allocates national power among three branches: Article I vests legislative power in Congress, Article II vests executive power in the President, and Article III vests judicial power in the Supreme Court and lower federal courts. Separating the power to make, enforce, and interpret law is itself a liberty-protecting device: each branch can check the others, and no one body holds the whole chain.

Two further features matter for the hierarchy of sources. The Supremacy Clause of Article VI declares the Constitution, federal statutes made in pursuance of it, and treaties to be the supreme law of the land, binding judges in every state notwithstanding contrary state law. And the first ten amendments, the Bill of Rights, together with later amendments such as the Fourteenth, place certain individual rights beyond ordinary lawmaking. Each state also has its own constitution governing its own government; a state constitution may protect rights more generously than the federal floor, though never less.

Judicial review and Marbury v. Madison

Who decides whether a statute crosses a constitutional boundary? In the American tradition the answer was fixed by Marbury v. Madison (1803). In the last days of the Adams administration, William Marbury was appointed a justice of the peace; his commission was signed and sealed but never delivered. The incoming Jefferson administration, through Secretary of State James Madison, refused to deliver it. Marbury sued directly in the Supreme Court, asking for a writ of mandamus - an order compelling an official to perform a duty - relying on a section of the Judiciary Act of 1789 that seemed to authorize such original suits.

Chief Justice John Marshall's opinion moved in three steps. First, Marbury had a legal right to the commission, which was complete when signed and sealed. Second, the law must supply a remedy for the violation of a legal right, even against a high officer of government. Third, and decisively, the Court could not grant the remedy, because the statutory grant of original jurisdiction went beyond the categories Article III allows for original suits in the Supreme Court. Faced with a statute that conflicted with the Constitution, the Court held the statute void.

The holding establishes judicial review: courts deciding cases must prefer the Constitution over an inconsistent statute, because the Constitution is law - supreme law - and it is the judiciary's task, in Marshall's phrase, "to say what the law is." Notice the craft as well as the principle: the Court asserted the power of review in the very act of declining to exercise power, giving the new doctrine a footing that the political branches could not easily defy. Many other democracies later adopted judicial review, some through specialized constitutional courts that hear such questions exclusively.

Statutes

A statute is a written law enacted by a legislature - a body such as a parliament or congress that represents the public. Statutes are the workhorses of a modern legal system, covering everything from crimes and taxes to contracts, employment, and the environment.

A typical statute begins as a proposed bill, is debated and amended, must be approved by the legislative body (and often signed by an executive), and then takes effect. Because statutes are written in general language, they must be interpreted and applied to specific facts, which is a major task of courts and agencies. When a statute is clear, it controls; courts apply it even if they might have chosen a different rule.

In the United States the path is exacting. A bill is introduced, studied in committee, and passed in identical form by both the House and the Senate - the requirement of bicameralism - and then presented to the President, who may sign it or veto it; Congress can override a veto by a two-thirds vote of each house. Enacted statutes are then organized by subject into codes, such as the United States Code, so researchers can find current law without reading session laws in order. States mirror this design with their own legislatures, governors, and codes.

Reading a statute

Interpretation deserves a first look now because every later module depends on it. Start always with the text, read in context: words take meaning from the sentence, the section, and the statute as a whole. Courts also presume the legislature did not intend absurd results, and they read specific provisions to control general ones. Where text remains open, judges consult the statute's evident purpose and structure; some also weigh legislative history such as committee reports, a practice that textualist judges treat with caution because only the enacted text passed both houses.

A classroom classic shows why interpretation is unavoidable. Imagine an ordinance: "No vehicles in the park." Cars are plainly covered. But bicycles? A child's battery-powered toy car? An ambulance answering an emergency call? A decommissioned army truck installed as a war memorial? The word "vehicle" alone will not answer; the interpreter must ask what mischief the rule targets - noise, danger, exhaust - and how far its language can honestly stretch. Statutory drafting can narrow such questions with definitions and exceptions, but no drafter forecloses them all. That is why statutes and courts are partners, not rivals.

How they relate

Constitutions and statutes work together in a hierarchy. The legislature may pass statutes, but only within the powers the constitution grants and subject to the rights the constitution protects. In many systems a court can exercise judicial review, examining whether a statute conforms to the constitution and refusing to enforce it if it does not. The relationship can be pictured as a pyramid: the constitution at the apex, statutes beneath it, and, as we will see, regulations and individual decisions below those. The higher the source, the more it governs the ones below.

Keep in mind that systems differ in the details, but the core idea is widely shared: a supreme constitution frames the whole system, and legislatures make statutes within it.

The hierarchy also creates a continuing conversation among institutions. If a court interprets a statute in a way the legislature dislikes, the legislature can simply amend the statute; the court's reading was law only because the statute was. But if a court holds a statute unconstitutional, the legislature cannot fix the problem by re-enacting it. Change must then come from a constitutional amendment or from the court itself overruling the precedent, both deliberately difficult. This is why constitutional rulings carry such weight, and why courts often try to resolve cases on statutory grounds first when they can.

Common misconceptions

First: "unconstitutional" does not mean unwise, unfair, or unpopular. It is a technical claim that a rule conflicts with a specific superior provision. A law can be foolish yet constitutional, or well-intentioned yet invalid. Arguments that skip the step of identifying the conflicting constitutional text are political arguments wearing legal dress.

Second: courts do not roam the statute books striking down bad laws. Judicial review operates only inside a live dispute between parties with a real stake; in the American federal system courts do not issue advisory opinions, though a few states and several other countries do permit them. Until someone with standing brings a proper case, even a plainly doubtful statute remains on the books.

Third: constitutional rights generally run against government, not against private actors. A private employer that restricts what workers say on the job may or may not violate an employment statute, but the constitutional free-expression guarantee is usually not the right tool, because no state action is involved. Knowing which source of law addresses which actor is half of issue-spotting.

Try it: the licensing statute

Apply the hierarchy. The national constitution guarantees freedom of expression. The legislature, citing an epidemic of false advertising, enacts a statute requiring anyone who publishes commentary about companies to obtain a government license, revocable for "irresponsible criticism." A blogger's license is revoked after she criticizes a large firm, and she sues. Work through the sources in order.

Start at the top. The statute is a real enactment by the recognized legislature, so it is presumptively valid positive law; the question is whether it conforms to the superior source. Licensing speech about matters of public concern, with revocation for criticism officials deem irresponsible, collides with the core of a free-expression guarantee: it conditions participation in public debate on government permission and invites viewpoint-based enforcement. A court exercising judicial review would ask whether the guarantee, properly interpreted, tolerates such a scheme; on these facts the statute would very likely be held invalid, and the revocation would fall with it.

Now vary the facts. Suppose instead the statute only required sellers to label paid endorsements as advertising, and the blogger objected. That narrower rule regulates commercial disclosure rather than licensing criticism, serves the fraud-prevention purpose directly, and leaves debate open; most systems would sustain it. The exercise shows the method: identify the sources in play, rank them, interpret each, and test the lower against the higher. That method, not any memorized result, is the lesson.

You now hold the top two layers of the pyramid. The next lesson adds the remaining two - judge-made common law and agency regulations - and shows how conflicts among all four are resolved.

Sources

1. Marbury v. Madison, 5 U.S. 137 (1803). Legal Information Institute, Cornell Law School. law.cornell.edu
2. Oyez. (n.d.). Marbury v. Madison. IIT Chicago-Kent College of Law. oyez.org
3. National Archives. (n.d.). The Constitution of the United States: A transcription. America's Founding Documents. archives.gov
4. United States Congress. (n.d.). The legislative process. Congress.gov ↗. congress.gov
5. Legal Information Institute. (n.d.). Judicial review. Cornell Law School. law.cornell.edu
6. Legal Information Institute. (n.d.). Statutory construction. Cornell Law School. law.cornell.edu
7. Eskridge, W. N. (1987). Dynamic statutory interpretation. University of Pennsylvania Law Review, 135(6), 1479. doi.org/10.2307/3312014

Two more sources complete the everyday picture: judge-made common law and agency-made regulations. Together with constitutions and statutes, they make up most of the law you will encounter.

Common law

Common law is law developed by courts through their decisions in individual cases, rather than by a legislature writing a code in advance. In common-law traditions (such as those of England, the United States, Canada, Australia, and others), when a court decides a case it not only resolves that dispute but may also announce a legal rule that guides future cases.

Over time, these decisions accumulate into bodies of doctrine - much of contract, tort, and property law began this way. Judges are expected to follow the reasoned decisions of earlier courts, a practice called precedent that we study in Module 4. Common law is flexible: it can adapt gradually as new situations arise. It is also subordinate to statutes - a legislature can change or override a common-law rule by passing a statute, and where a statute speaks, it prevails.

The name records its history. After the Norman Conquest of 1066, royal courts in England gradually built a body of decisions applied throughout the realm - law "common" to the whole kingdom, as opposed to scattered local customs. Alongside these courts grew a separate court of equity, the Chancery, which supplied remedies the common-law courts could not, such as injunctions and specific performance, when money damages would not do justice. Most jurisdictions have since merged the two systems into one, but the vocabulary survives: lawyers still classify remedies as legal (typically damages) or equitable (typically court orders), a distinction you will meet again in contracts and torts.

Not every system relies heavily on common law. In civil-law traditions (used across much of continental Europe, Latin America, and elsewhere), comprehensive written codes enacted by legislatures are the primary source, and judicial decisions carry less formal weight as binding precedent. Many countries blend features of both. This course teaches concepts common to both families but uses the common-law vocabulary of cases and precedent because it makes the reasoning skills especially visible.

How common law grows: a worked example

Watch the method in a famous sequence. Nineteenth-century courts generally held that a manufacturer who sold a defective product was answerable in negligence only to its immediate buyer - the party "in privity of contract" - so a consumer hurt by a product bought through a dealer usually could not sue the maker. Courts then began recognizing exceptions for things dangerous in themselves, such as poisons mislabeled as medicine. Each exception was a precedent, and the exceptions multiplied.

In MacPherson v. Buick Motor Co. (1916), a car's defective wooden wheel collapsed and injured the buyer, who had purchased from a dealer rather than from Buick. Judge Benjamin Cardozo, writing for New York's highest court, surveyed the accumulated exceptions and drew out their real principle: when a product is reasonably certain to endanger life or limb if negligently made, and the maker knows it will be used by persons beyond the immediate buyer, the maker owes a duty of care to those users. The privity limit fell, not by legislative repeal, but by judicial reasoning from precedent to principle - the common law's characteristic motion.

Erie and the limits of federal common law

A structural landmark shows how common law fits a federal system. Harry Tompkins was walking along the Erie Railroad's tracks in Pennsylvania when something projecting from a passing train - apparently an open door - struck him, costing him an arm. Under Pennsylvania court decisions, a person walking along (not across) the tracks was a trespasser to whom the railroad owed only a minimal duty. Tompkins therefore sued in federal court in New York, invoking diversity jurisdiction, because federal courts at the time followed Swift v. Tyson (1842) and applied their own "general law" in such cases, which was friendlier to him. He won a large verdict.

In Erie Railroad Co. v. Tompkins (1938), the Supreme Court overruled Swift. Justice Brandeis explained that the regime of general federal common law had produced two evils: unequal justice, since the outcome of an ordinary tort or contract case could depend on which courthouse the plaintiff chose, and systematic forum shopping to exploit that difference. The Court held that except where the Constitution or a federal statute governs, federal courts sitting in diversity must apply state law, including the decisions of the state's courts. In Brandeis's words, "There is no federal general common law."

The holding matters here for what it teaches about sources. Common law is not a free-floating body of wisdom that any court may restate; it is the positive law of a particular jurisdiction, made by that jurisdiction's courts. Federal common law survives only in limited enclaves involving distinctly federal interests. For the ordinary law of contracts, torts, and property in the United States, the states are the lawmakers, and their rules genuinely differ - which is why this course so often says "the rule varies by state."

Regulations

Modern legislatures cannot spell out every detail, so they often pass a statute that sets goals and then delegate authority to an administrative agency to fill in specifics. The detailed rules the agency issues are called regulations (or delegated or subordinate legislation). Regulations have the force of law, but only within the authority the statute grants: an agency cannot make a rule the statute did not empower it to make. Agencies also apply their rules to particular parties, for example by issuing licenses or penalties, subject to review by courts. Regulations let law keep up with technical and changing fields such as food safety, finance, and the environment.

The life cycle of a regulation is itself governed by law. The statute creating an agency, often called its enabling act, defines its mission and powers. In the United States, when a federal agency wants a significant new rule it typically must publish a proposal, take public comments, and respond to them before adopting a final rule - notice-and-comment rulemaking under the Administrative Procedure Act. Agencies also adjudicate individual disputes, inspect, license, and enforce. Each function mirrors, in miniature, a branch of government, which is why administrative law works hard to keep agencies inside their statutory lanes.

Courts police those lanes. A regulation beyond the granted authority is called ultra vires and will be set aside; so will rules adopted without required procedures or lacking reasoned explanation. A further American question has been how much weight courts should give an agency's own interpretation of an ambiguous statute. For four decades, Chevron v. Natural Resources Defense Council (1984) directed courts to accept reasonable agency readings; in Loper Bright Enterprises v. Raimondo (2024) the Supreme Court overruled that framework, holding that courts must exercise independent judgment about what a statute means. The episode is a live reminder that even the rules about rules can change.

Codes and mixed traditions

Civil-law systems organize private law differently. France's Civil Code of 1804 and Germany's Civil Code of 1900 aimed to state the law of persons, property, and obligations comprehensively, in systematic articles a citizen could in principle read. Judges in these systems decide cases by applying and construing code provisions; their decisions are studied and followed in practice, especially a settled line of decisions, but a single precedent does not formally bind later courts the way it can in common-law systems.

Several jurisdictions deliberately blend the families. Louisiana and Quebec pair civil codes for private law with common-law public institutions; Scotland and South Africa mix Roman-law inheritance with common-law method. And within every common-law country, statutes and regulations now occupy so much ground that daily practice everywhere involves reading enacted text. The families differ in emphasis and technique more than in results: most concrete disputes would come out similarly on either side of the divide.

The hierarchy of sources

When sources conflict, higher-ranked law controls. A common ordering, from highest to lowest, is:

1. Constitution - the supreme law; everything below must conform to it.
2. Statutes - enacted by the legislature within constitutional limits.
3. Regulations - issued by agencies under the authority of a statute.
4. Common law - judge-made rules that fill gaps and can be overridden by a valid statute.

The exact placement of common law relative to regulation varies with context, but two anchors are firm: the constitution is supreme, and a valid statute overrides conflicting common law. Knowing the ranking tells you which rule wins when two sources point in different directions.

Federal systems add a second axis: not just which type of source, but which government's law. Under arrangements like the American Supremacy Clause, valid national law displaces conflicting state or provincial law within the national government's powers, a doctrine called preemption. So a complete conflicts checklist asks two questions: is each rule valid within its own system's pyramid, and if rules from different governments still collide, which government has the superior claim on this subject? Details vary across federations, but the two-step structure is general.

Common misconceptions

First: "judges just make it up." Common-law judges are constrained by precedent, by the record and arguments before them, by the duty to give public reasons, and by appellate review. MacPherson shows creativity, but creativity disciplined by an existing line of authority; Cardozo claimed the exceptions had already outgrown the old rule.

Second: "regulations are not real law." A valid regulation binds exactly as a statute does, and violating one can cost a business its license or worse. The difference is pedigree and rank, not force. The practical lawyer often spends far more time in the regulations than in the statute they implement.

Third: "the common law is finished, since legislatures can always override it." Legislatures override selectively; the background rules of contract, tort, and property in common-law jurisdictions remain largely judge-made, and courts keep developing them for situations no statute addresses - data, drones, and gene technology reach courts before they reach codes.

Try it: which source wins?

Work this fact pattern. A statute directs the transport agency to regulate "motor vehicle safety equipment" and authorizes rules about brakes, lights, and tires. The agency, concerned about distraction, issues a regulation banning all eating while driving. Meanwhile, a century-old common-law rule in the jurisdiction holds drivers to ordinary reasonable care, and a newer statute says a driver is presumed negligent if involved in a crash while violating "any safety regulation." Dana, cited for eating a sandwich while driving and involved in a minor crash, challenges the citation. Analyze the sources.

Start with the regulation's pedigree. The enabling act covers safety equipment - brakes, lights, tires. A ban on eating regulates driver behavior, not equipment; the agency has likely acted ultra vires, outside its delegated authority, so the regulation is invalid however sensible it may be. Policy arguments for the ban belong in the legislature, which could enact it directly or broaden the delegation.

Now cascade the consequences. The negligence presumption is triggered only by violating a valid safety regulation; with the eating rule struck down, the presumption never engages. Dana's conduct is then measured by the common-law standard of reasonable care, under which eating while driving is simply evidence a fact-finder may weigh. Notice the order of operations: validate each rule against its superior source first, and only then apply the survivors to the facts. That two-step is the everyday use of the pyramid you have now completed.

Sources

1. MacPherson v. Buick Motor Co., 217 N.Y. 382 (1916). New York State Unified Court System, Official Reports archive. nycourts.gov
2. Erie Railroad Co. v. Tompkins, 304 U.S. 64 (1938). Legal Information Institute, Cornell Law School. law.cornell.edu
3. Swift v. Tyson, 41 U.S. 1 (1842). Legal Information Institute, Cornell Law School. law.cornell.edu
4. Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 (1984). Legal Information Institute, Cornell Law School. law.cornell.edu
5. Oyez. (n.d.). Loper Bright Enterprises v. Raimondo. IIT Chicago-Kent College of Law. (U.S. Supreme Court, decided 2024.) oyez.org
6. Administrative Procedure Act, 5 U.S.C. Section 553 (rule making). Legal Information Institute, Cornell Law School. law.cornell.edu
7. Lawson, G. (1994). The rise and rise of the administrative state. Harvard Law Review, 107(6), 1231. doi.org/10.2307/1341842

Courts are the institutions that apply law to disputes. Although details differ across countries, most court systems share a similar three-level structure and a basic division of labor between courts that find facts and courts that review legal questions.

Keep one organizing question in mind throughout: who decides what? Facts and law are different kinds of questions, answered by different actors under different rules, and almost everything about court structure follows from that division. A dispute over what happened - how fast the car was going, what the parties said - is a question of fact. A dispute over what the rule is or means - what "reasonable care" requires, whether a statute covers e-bikes - is a question of law. Trial courts specialize in the first; appellate courts specialize in the second.

Trial courts

A case usually begins in a trial court (sometimes called a court of first instance). This is where the facts are established: witnesses testify, documents and other evidence are presented, and a decision-maker determines what happened and applies the law to those facts.

The fact-finder may be a judge alone (a bench trial) or, for many cases in some systems, a jury of ordinary citizens who decide the facts while the judge decides questions of law and instructs the jury. The person who brings a civil case is the plaintiff; the one who must answer is the defendant. In a criminal case the government (the prosecution) brings the charge against the accused.

Trial work is mostly preparation, and each stage has a legal shape. A civil case opens with pleadings: the complaint stating the claim and the answer responding. Discovery follows, in which each side must disclose relevant documents and witnesses to the other - a feature designed to prevent trial by ambush and to promote settlement, since parties who see the same evidence can predict the verdict. Motions can end a case early: a court may dismiss a complaint that states no legal claim, or grant summary judgment when no genuine dispute of material fact exists and one side wins as a matter of law.

Only a small fraction of filed cases reach trial; most settle, and in criminal practice most charges resolve by plea. When trial does come, the rules of evidence filter what the fact-finder may hear, excluding, for example, much hearsay and unfairly prejudicial material. The judge manages this filter through rulings on objections; the fact-finder then weighs what survives. The verdict resolves the facts; the judgment applies the law to them and states who owes what.

The American court map

The United States runs two parallel systems. Each state maintains its own courts - trial courts of general jurisdiction, usually an intermediate appellate court, and a state supreme court - and these handle the vast majority of all litigation: family, contract, tort, property, and most criminal cases. The federal system is smaller and limited: 94 district courts conduct trials, 13 courts of appeals (12 regional circuits plus the Federal Circuit for specialized subjects) hear appeals, and the Supreme Court of the United States sits at the top with nine justices.

The Supreme Court controls its own docket. A losing party petitions for a writ of certiorari; the Court grants when at least four justices vote to hear the case, favoring questions on which the circuits disagree or issues of national importance. From thousands of petitions each year it typically agrees to decide fewer than a hundred. One more wrinkle matters: on questions of purely state law, each state's highest court has the final word; the U.S. Supreme Court reviews state decisions only insofar as they involve federal law. Module 3's jurisdiction lesson builds on this map.

Appellate courts

A party who loses may be able to appeal to an appellate court. Appeals are not do-overs of the trial. Appellate courts generally do not hear new witnesses or re-decide the facts; instead they review whether the trial court made a legal error - for example, misreading a statute or admitting evidence it should have excluded. They read the record and the parties' written briefs, hear argument, and then affirm (uphold), reverse (overturn), or remand (send back for further proceedings). Many systems have an intermediate appellate court and, above it, a supreme or high court that has the final say and whose decisions bind the courts below.

Mechanically, an appeal is an argument about the record. The appellant must ordinarily have preserved each claimed error by objecting at the time, so the trial judge had a chance to fix it; points raised for the first time on appeal are usually forfeited except for rare plain error. Appeals are typically heard by panels - three judges in the American courts of appeals - and produce written opinions: a majority opinion that states the court's holding, sometimes concurrences agreeing on the result for different reasons, and dissents. In important cases a full appellate court may rehear a panel decision en banc.

Not every error wins a reversal. Under the harmless-error principle, a conviction or judgment stands despite a mistake if the error did not affect the outcome or the party's substantial rights. This keeps appeals focused on errors that mattered rather than on perfection for its own sake.

Standards of review

The appellate court's deference depends on the kind of question, expressed in standards of review. Pure questions of law - what a statute means, what elements a claim requires - are reviewed de novo, with no deference to the trial judge, because uniform legal rules are the appellate court's core job. Findings of fact made by a trial judge are reviewed only for clear error, and a jury's verdict receives even greater respect: it stands if a reasonable jury could have reached it on the evidence. Discretionary case-management calls, such as many evidentiary rulings, are reviewed for abuse of discretion.

The logic is institutional competence. The trial judge and jury saw the witnesses; a transcript conveys words but not demeanor, so fact-findings get deference. Legal rules, by contrast, must be the same across many courtrooms, so the appellate court decides them fresh. When lawyers evaluate an appeal, the standard of review is often the whole ballgame: an argument styled "the jury believed the wrong witness" is nearly hopeless, while "the judge applied the wrong legal test" gets a full, fresh look.

Some questions are mixed, applying a legal standard to found facts: was this conduct "reasonable," was that confession "voluntary." Courts divide over how to review them. Some treat the application as fact-like and defer; others treat it as law-like and decide fresh; the choice often turns on whether the issue benefits more from uniform rules or from case-by-case judgment. Spotting a mixed question, and arguing for the standard of review that favors your client, is a staple of appellate advocacy.

Why the structure matters

Court level	Main job	Decides facts?
Trial court	Hold the trial; find facts; apply law	Yes
Appellate court	Review for legal error	No
Supreme / high court	Final review; set binding precedent	No

This layered design serves two goals. It gives each dispute a careful first hearing where evidence can be tested, and it provides a check on error and a way to keep the law uniform, because higher courts can correct mistakes and settle questions for everyone below them. Understanding which court does what tells you where facts are decided, where law is refined, and whose word is final.

The two appellate goals - error correction and law development - explain the division of labor at the top. Intermediate appellate courts exist mainly to correct errors in the mass of ordinary appeals. A supreme court that chooses its cases exists mainly to develop the law: to resolve conflicting interpretations, announce rules for recurring situations, and keep the system coherent. That is why supreme courts prefer cases presenting clean legal questions of broad importance over cases that merely look wrongly decided.

Inside the courtroom

A quick cast of characters completes the picture. The judge rules on law and procedure and, in bench trials, finds facts. The jury, where used, finds facts under the judge's instructions; jury service is one of the few occasions ordinary citizens exercise state power directly. Advocates present each side's case; the clerk maintains the file; the court reporter produces the verbatim record that makes appellate review possible. Witnesses supply evidence, and parties, though center stage, mostly watch professionals speak for them.

Comparative note: common-law trials are adversarial - the parties develop the evidence, and the judge referees. Many civil-law countries use a more inquisitorial process, in which the court itself takes a leading role in questioning witnesses and building the file. Each tradition has migrated toward the other in recent decades, and both aim at the same targets: accurate facts, correctly applied law, and a process the parties accept.

Common misconceptions

First: an appeal is not a second trial. No jury, no witnesses, no new evidence; only the record, the briefs, and legal argument. A disappointed litigant who wants to "tell the story again" has misunderstood what the appellate court is for.

Second: losing at trial does not guarantee review at the top. Litigants generally have a right to one appeal, but supreme courts with discretionary dockets decline the great majority of requests, and denial of review is not an endorsement of the decision below - it simply leaves it standing.

Third: a single trial judge's ruling does not make binding precedent. Trial decisions bind the parties, not future courts; precedent that binds comes from appellate holdings. And the dramatic courtroom objection has a mundane purpose: it preserves the issue for appeal and lets the judge correct the problem now, which is why lawyers object even when the jury seems annoyed.

Try it: sort the appeal

A jury finds a delivery company liable for a crash and awards damages. On appeal, the company argues: (a) the trial judge misinterpreted the statute defining "commercial driver"; (b) the jury should have believed the company's witness, who said the light was green; and (c) the judge wrongly admitted a photograph the company calls inflammatory. Sort each argument: what kind of question is it, what standard of review applies, and what are its prospects?

Worked analysis. Argument (a) is a pure question of law: statutory interpretation, reviewed de novo. If the reading genuinely affected the verdict, this is the company's best route to reversal, and possibly to remand for a new trial under the correct standard. Argument (b) attacks a credibility determination - the core of the jury's factual role. The verdict stands if any reasonable jury could credit the other witness, so this argument is almost certainly a loser. Argument (c) challenges an evidentiary ruling, reviewed for abuse of discretion; even if the court finds the admission questionable, the company must also show the error was not harmless.

Notice what the exercise reveals: the same appeal contains strong and weak arguments, and the difference tracks the structure of the system - law reviewed fresh, facts left where they were found, discretion respected within limits. That is the three-level design doing exactly what it was built to do.

Sources

1. Administrative Office of the U.S. Courts. (n.d.). Court role and structure. United States Courts. uscourts.gov
2. Administrative Office of the U.S. Courts. (n.d.). Comparing federal and state courts. United States Courts. uscourts.gov
3. Supreme Court of the United States. (n.d.). The court and its procedures. supremecourt.gov
4. Federal Judicial Center. (n.d.). Courts. History of the Federal Judiciary. fjc.gov
5. Legal Information Institute. (n.d.). Standard of review. Cornell Law School. law.cornell.edu
6. Federal Rules of Civil Procedure, Rule 52 (findings and conclusions by the court). Legal Information Institute, Cornell Law School. law.cornell.edu
7. Shavell, S. (1995). The appeals process as a means of error correction. The Journal of Legal Studies, 24(2), 379-426. doi.org ↗

Before any court can decide a case, it must have jurisdiction - the legal authority to hear that particular dispute and to bind the parties. A brilliant judgment from a court that lacked jurisdiction is void. Jurisdiction is therefore the threshold question in litigation, and lawyers often fight about it first. There are two main kinds.

Why fight about the forum at all? Because where a case is heard can shape how it ends. Different courts mean different procedures, different jury pools, different distances from witnesses, sometimes different applicable law, and always different costs for the parties. Plaintiffs choose the most favorable forum their lawyers can defend; defendants push back with jurisdictional objections. Understanding the rules of this opening contest is understanding a large fraction of real litigation strategy.

Subject-matter jurisdiction

Subject-matter jurisdiction is a court's authority over the type of case. Court systems are specialized: a small-claims court can hear only modest money disputes; a family court handles divorce and custody; a bankruptcy or tax court hears only its assigned subjects. If you file a bankruptcy petition in a traffic court, it will be dismissed for lack of subject-matter jurisdiction no matter how strong your case is. This kind of jurisdiction is set by law and generally cannot be created just because the parties agree to it.

The American federal courts illustrate the idea sharply, because their subject-matter jurisdiction is limited by design. The two great categories are federal-question jurisdiction, for claims arising under the Constitution, federal statutes, or treaties, and diversity jurisdiction, for suits between citizens of different states where the amount in controversy exceeds 75,000 dollars. Diversity must generally be complete - no plaintiff may share state citizenship with any defendant. Everything else belongs in state courts, which are courts of general jurisdiction and can hear nearly any claim.

Because subject-matter jurisdiction defines the court's very power, it cannot be waived, consented to, or forfeited. A federal court must dismiss a case the moment the defect appears, even if both parties want to stay, and the objection can be raised at any stage - on appeal, even by the party that chose the forum. Contrast this with personal jurisdiction, which protects the defendant and therefore can be waived, as we will see.

Personal jurisdiction

Personal jurisdiction is a court's authority over the parties, especially the defendant. Broadly, a court may exercise personal jurisdiction when the defendant has a sufficient connection to the place - for example, by living there, doing business there, or causing harm there - so that being sued in that location is fair. The underlying principle is fairness and notice: a person should not be hauled before a distant court with which they have no meaningful connection. Personal jurisdiction protects defendants from being sued just anywhere.

American doctrine began territorially: in the nineteenth century, a court's power ran to persons and property found within the state's borders, so service of process inside the state was the key. Modern commerce outgrew that picture, and International Shoe Co. v. Washington (1945) supplied the replacement. A Delaware shoe company based in Missouri employed a dozen salesmen in Washington but claimed it was not "present" there when the state sought unemployment-fund contributions. The Supreme Court upheld jurisdiction and announced the modern test: due process requires that the defendant have such minimum contacts with the forum that the suit "does not offend traditional notions of fair play and substantial justice."

Two branches grew from International Shoe. Specific jurisdiction reaches claims that arise out of or relate to the defendant's contacts with the forum: a company that purposefully sells into a state can be sued there over those very sales. General jurisdiction lets a court hear any claim against a defendant, wherever it arose, but only where the defendant is essentially at home - for an individual, domicile; for a corporation, ordinarily its state of incorporation and principal place of business, as the Court confirmed in Daimler AG v. Bauman (2014).

The practical consequence: a multinational retailer can be sued for anything in its one or two home states, and elsewhere only on claims connected to the forum. For plaintiffs' lawyers this map dictates where cases can realistically be filed; for businesses deciding whether to enter a new market, it prices the litigation exposure that comes with the customers.

Personal jurisdiction, unlike subject-matter jurisdiction, is the defendant's shield and can be surrendered. Defendants consent in advance through contract clauses, submit by appearing and defending without objection, or waive by failing to raise the defense promptly. And every exercise of jurisdiction requires proper notice: service of process delivering the complaint by an approved method reasonably calculated to inform the defendant. Jurisdiction without notice is as fatal as notice without jurisdiction.

Minimum contacts in practice

The touchstone of specific jurisdiction is purposeful availment: the defendant must have deliberately reached into the forum and taken the benefits of its market or protection of its laws. An online retailer that advertises to a state, ships hundreds of orders there, and services customers there has purposefully availed itself; if one of those products injures a local buyer, the local courts may hear the claim. The contacts are the defendant's own choices.

Contrast unilateral or accidental connections. If a customer buys a product in one state and later moves to another, carrying it along, the manufacturer did not choose the second state; the customer did. Courts are wary of resting jurisdiction on such third-party acts, and the precise reach of jurisdiction over goods that travel through the stream of commerce remains contested, with rules that vary across jurisdictions. For an introductory course, hold the core: ask what the defendant itself did to connect with the forum, and whether the claim grows out of those very connections.

A final variant deserves a sentence. Courts also exercise authority over property located within the forum, called jurisdiction in rem, to settle ownership of land, quiet title, or distribute estates. Even here the modern approach asks the same fairness question, since judgments about property inevitably affect the people who claim it; American law now runs even property-based jurisdiction through the minimum-contacts screen when it is used to reach the owner rather than the thing itself.

Jurisdiction in a federal system

Many countries divide authority between a national government and regional governments (states or provinces), and their court systems reflect this. There may be parallel sets of courts - national and regional - each with its own subject-matter jurisdiction. Some matters belong to national courts, some to regional courts, and some can be heard in either, a situation called concurrent jurisdiction.

Rules then decide where a case may or must go, and sometimes a case can be moved from one system to another. The details vary widely, but the core idea is simple: which court may hear a case depends on both the subject and the parties, and on how authority is allocated among levels of government.

The American version supplies concrete examples. Most federal-question and diversity cases may be filed in state court too, since concurrency is the default; a defendant sued in state court on a claim that could have begun in federal court can usually remove it to the federal district court. A few subjects are exclusively federal - bankruptcy and patent infringement among them - while enormous fields such as family and probate law remain, in practice, the province of state courts. A lawyer's forum analysis therefore runs: which systems have subject-matter jurisdiction, which locations have personal jurisdiction, and, among the survivors, which forum serves the client best?

Venue, a related idea

Do not confuse jurisdiction with venue, which is the specific geographic location, among courts that all have jurisdiction, where a case is properly heard (often where the events occurred or a party resides). Jurisdiction asks whether a court has authority at all; venue asks which of several authorized courts is the appropriate place. Both must be satisfied for a case to proceed in a given court.

Venue rules are also more forgiving in operation. A court with jurisdiction but poor venue may transfer the case to a better-located court in the same system rather than dismiss it, and courts may transfer for convenience of parties and witnesses even when venue is technically proper. Under the doctrine of forum non conveniens, a court may decline to hear a case that belongs, as a practical matter, in another system's courts entirely. And parties frequently settle the question in advance: contracts commonly contain forum-selection clauses choosing a specific place for any dispute, and courts generally enforce reasonable ones.

Common misconceptions

First: "you can sue anyone anywhere." Plaintiffs choose among lawful forums, but the defendant's due-process rights and the subject-matter rules cut down the menu sharply. The plaintiff's home court is often unavailable precisely because the defendant has no contacts there.

Second: "federal courts outrank state courts." They are parallel systems, not a ladder. A federal district court cannot reverse a state court; review of state decisions runs only through the state's own appellate courts, and the U.S. Supreme Court may review only the federal questions in them. On state-law questions, state supreme courts are supreme indeed.

Third: "the court that hears the case applies its own substantive law." Not necessarily. A forum applies its own procedure, but choice-of-law rules may direct it to another jurisdiction's substantive rules - a California court can find itself applying Oregon contract law. Jurisdiction and choice of law are separate questions, run through separate analyses.

Try it: the camp stove case

Rosa, a California resident, buys a camp stove from Summit Gear, an Oregon company, through Summit's website. Summit advertises heavily in California and ships thousands of orders there each year. The stove flares up and burns Rosa at a California campsite; her medical bills and losses exceed 90,000 dollars. Summit is incorporated in Oregon with its only offices there. Where can Rosa sue? Work jurisdiction, then venue.

Personal jurisdiction first. California: Summit purposefully availed itself of the California market through advertising and volume sales, and Rosa's claim arises from exactly those contacts, so specific jurisdiction in California is proper and fair. Oregon: Summit is at home there, so general jurisdiction reaches any claim. Subject matter next. State trial courts of general jurisdiction in either state can hear this product-liability claim. Federal court is also open under diversity jurisdiction: Rosa and Summit are citizens of different states and the amount in controversy exceeds the 75,000-dollar threshold, so she may file in federal district court, or Summit may remove a state-court suit.

Venue last. In the federal system, venue would lie in a district where a substantial part of the events occurred, such as the California district embracing the injury, or where the defendant resides in Oregon. Rosa will likely choose a California forum near her witnesses and doctors; Summit may try to move the case toward Oregon. Every move in that contest is an application of the concepts in this lesson - which is why jurisdiction, dry as it sounds, decides real cases before the merits are ever reached.

Sources

1. International Shoe Co. v. Washington, 326 U.S. 310 (1945). Legal Information Institute, Cornell Law School. law.cornell.edu
2. Oyez. (n.d.). Daimler AG v. Bauman, 571 U.S. 117 (2014). IIT Chicago-Kent College of Law. oyez.org
3. Legal Information Institute. (n.d.). Personal jurisdiction. Cornell Law School. law.cornell.edu
4. Legal Information Institute. (n.d.). Subject matter jurisdiction. Cornell Law School. law.cornell.edu
5. Federal question jurisdiction, 28 U.S.C. Section 1331. Legal Information Institute, Cornell Law School. law.cornell.edu
6. Diversity of citizenship and amount in controversy, 28 U.S.C. Section 1332. Legal Information Institute, Cornell Law School. law.cornell.edu
7. Venue generally, 28 U.S.C. Section 1391. Legal Information Institute, Cornell Law School. law.cornell.edu

One of the most important divisions in any legal system is between civil law and criminal law. (Here "civil law" means the law of private disputes, a different use of the phrase than the "civil-law system" contrasted with common law in Module 2.) The two differ in who brings the case, what is at stake, what must be proved, and what the court can order.

Hold on to the distinction's purpose as you learn its details. Civil law is the community's machinery for repairing private harms; criminal law is its machinery for condemning public wrongs. The same event can engage both machines, but each runs on its own fuel, follows its own rules, and produces its own kind of output. Mixing them up produces some of the most common misunderstandings about law in circulation.

Who sues, and why

Civil law governs disputes between private parties - individuals, businesses, or organizations - about their rights and duties toward one another. A civil case is brought by the injured party (the plaintiff) seeking a remedy, most often damages (money) or an injunction (a court order to do or stop doing something). The aim is usually to compensate the plaintiff or restore rights, not to punish. Contract disputes, personal-injury claims, and property disputes are civil.

Criminal law concerns conduct treated as an offense against society as a whole, not just against a private victim. A criminal case is brought by the state (through a prosecutor) against the accused, and its purpose is to hold the wrongdoer accountable through punishment - a fine, probation, or imprisonment - and to protect the public and deter future wrongdoing. Even when there is an individual victim, the case is styled as the government against the defendant, because crime is understood as a wrong against the public order.

You can read the difference in the case names. A civil caption pairs private parties: Garcia v. Chen, or a person against a company. A criminal caption pairs the sovereign with the accused: State v. Rivera, People v. Doe, United States v. Smith, with the government always the moving party. The caption tells you instantly who bears the burden, what standard applies, and what kind of judgment can issue.

Control of the case differs too. A civil plaintiff owns her claim: she decides whether to file, settle, or abandon it. In a criminal case the prosecutor exercises that control, deciding whether to charge, what to charge, and whether to accept a plea agreement. The victim is a witness, often an important one, and modern systems give victims rights to be heard and informed, but the victim is not a party and cannot compel or forbid a prosecution. That allocation reflects the theory: the offense is against the public, so a public official answerable to the public conducts the case.

Safeguards on the criminal side

Because the criminal machine can take liberty, its procedure carries heavier armor. The accused is presumed innocent, so the prosecution bears the entire burden of proof from start to finish. In the United States the Supreme Court held in In re Winship (1970) that due process requires proof beyond a reasonable doubt of every fact necessary to the crime charged; the case arose from a juvenile proceeding, and the Court insisted the standard applies whenever the state seeks to brand conduct criminal.

Further protections cluster around the criminal defendant: the right to counsel, and to appointed counsel for those who cannot pay; the right to a jury in serious cases; the privilege against self-incrimination, so the defendant cannot be forced to testify and silence may not be treated as proof of guilt; and protection against double jeopardy. Civil procedure has fairness rules too - notice, discovery, an impartial tribunal - but a civil defendant can be compelled to testify, and in many systems a fact-finder may draw adverse inferences from a civil party's refusal to answer. The asymmetry is deliberate: the graver the state's threatened response, the thicker the process.

The standard of proof

Because the stakes differ, so does the standard of proof - how convinced the fact-finder must be.

	Civil case	Criminal case
Who brings it	Private plaintiff	The state (prosecutor)
Purpose	Compensate or restore	Punish and protect the public
Typical result	Damages or injunction	Fine, probation, imprisonment
Standard of proof	Preponderance of the evidence (more likely than not)	Beyond a reasonable doubt

In most civil cases the plaintiff must prove the claim by a preponderance of the evidence, meaning it is more likely true than not - a bare tipping of the scales past 50 percent. In criminal cases the prosecution must prove guilt beyond a reasonable doubt, a much higher bar, because a wrongful conviction can cost a person their liberty or reputation. The heavier standard reflects the principle that it is worse to punish the innocent than to let some guilty go free.

Between the two sits an intermediate standard, clear and convincing evidence, used in many jurisdictions for matters graver than money but short of crime: civil fraud in some contexts, termination of parental rights, civil commitment. Which claims take which standard varies by jurisdiction, but the architecture is constant: as the consequences of error grow, the law demands more certainty before imposing them. The classic expression is Blackstone's maxim that it is better for ten guilty persons to escape than for one innocent person to suffer; the reasonable-doubt standard is that maxim built into arithmetic.

A famous illustration

The most widely known demonstration of the two tracks is the pair of cases against O. J. Simpson. In 1995 a criminal jury acquitted him of the murders of Nicole Brown Simpson and Ronald Goldman: the prosecution had not, in that jury's judgment, eliminated reasonable doubt. In 1997 a civil jury, hearing wrongful-death and related claims brought by the victims' families, found him liable and awarded damages ultimately totaling about 33.5 million dollars.

The outcomes are not a contradiction. The civil jury needed only to find liability more probable than not, heard a somewhat different evidentiary record - including testimony the defendant could be compelled to give in the civil forum - and answered a different question: not "shall this man be punished by the state" but "shall he compensate these families." One event, two machines, two standards, two verdicts, each internally coherent. Fix this example in memory; it resolves the apparent puzzle better than any abstract statement.

One act, two cases

The same conduct can trigger both kinds of case. Suppose a driver injures a pedestrian through reckless driving. The state may prosecute the driver criminally for the offense, seeking punishment, while the pedestrian separately sues the driver civilly for damages to cover medical bills. These are independent proceedings with different parties, purposes, and standards of proof, so the outcomes need not match. A defendant might be acquitted in the criminal case (guilt not proven beyond a reasonable doubt) yet still be held liable in the civil case (fault shown by a preponderance of the evidence). Keeping the two tracks distinct is essential to understanding how law responds to harm.

Double jeopardy does not change this. The guarantee forbids a second criminal prosecution for the same offense by the same government after acquittal or conviction; it says nothing about a civil suit over the same conduct, which seeks a different kind of judgment. Nor does a criminal conviction preclude the civil case - quite the opposite: a conviction can often be used in the later civil suit as strong or conclusive evidence of the underlying facts, since it was proved under the stricter standard. An acquittal, by contrast, proves only that doubt remained, and so it settles little in the civil forum.

Remedies compared

Civil remedies aim at the plaintiff's position: compensatory damages measure the loss and try to restore it in money, while injunctions and orders of specific performance direct conduct when money is inadequate. One civil remedy borrows criminal coloring: punitive damages, available in many jurisdictions for outrageous or malicious conduct, which punish and deter rather than compensate. Because of that function, courts require heightened culpability, some jurisdictions demand a higher standard of proof for them, and constitutional and statutory limits restrain their size; the details vary considerably by state.

Criminal judgments run through the defendant's person and record: imprisonment, probation and supervision, fines paid to the state, and a conviction that carries lasting collateral consequences for employment, licensing, and more. One criminal remedy borrows civil coloring: restitution, an order that the offender repay the victim's losses as part of the sentence. The two tracks thus trade tools at the margins while keeping their centers distinct: civil law prices harm, criminal law condemns it.

Common misconceptions

First: victims do not "press charges" in any controlling sense. A victim's cooperation matters enormously as a practical and evidentiary matter, but the charging decision belongs to the prosecutor, who may proceed without the victim's blessing or decline despite it.

Second: an acquittal is not a declaration of innocence. It is a finding that the state failed to carry its heavy burden. That is why "not guilty" and "innocent" are not synonyms, and why a civil jury may later find liability without any inconsistency.

Third: losing a civil case does not send anyone to prison. The judgment is an obligation to pay or to obey an order; only defiance of a court order can escalate, through contempt proceedings, toward coercive sanctions. And the criminal standard is "beyond a reasonable doubt," not "beyond all doubt": jurors may convict while acknowledging bare theoretical possibilities, so long as no doubt grounded in reason remains.

Try it: the bar fight

Alex, angry over a spilled drink, punches Blake in a crowded bar, breaking Blake's jaw. Work both tracks: name the proceedings, the parties, the standards, and the possible outcomes, then say how the two could diverge.

Criminal track: the state charges Alex with assault (in some codes, battery); the caption is State v. Alex. The prosecutor must prove each element - an intentional harmful striking, the resulting injury - beyond a reasonable doubt. Conviction could bring jail, probation, a fine, and restitution for Blake's medical bills. Blake is the key witness but not a party; the prosecutor could accept a plea to a lesser offense over Blake's objection.

Civil track: Blake sues Alex for the intentional tort of battery, seeking compensatory damages for medical costs, lost income, and pain, plus punitive damages if the jurisdiction allows them for such deliberate violence. Blake must prove the claim by a preponderance of the evidence, and Alex can be compelled to testify. Divergence is easy to imagine: a criminal jury unsure beyond reasonable doubt about who started the fight might acquit, while the civil jury, weighing probabilities on the same facts, finds for Blake. If instead Alex is convicted first, that conviction would powerfully support Blake's civil case. Two machines, one punch - and now you can trace each gear.

Sources

1. In re Winship, 397 U.S. 358 (1970). Legal Information Institute, Cornell Law School. law.cornell.edu
2. Packer, H. L. (1964). Two models of the criminal process. University of Pennsylvania Law Review, 113(1), 1. doi.org/10.2307/3310562
3. Legal Information Institute. (n.d.). Burden of proof. Cornell Law School. law.cornell.edu
4. Legal Information Institute. (n.d.). Preponderance of the evidence. Cornell Law School. law.cornell.edu
5. Legal Information Institute. (n.d.). Double jeopardy. Cornell Law School. law.cornell.edu
6. Legal Information Institute. (n.d.). Punitive damages. Cornell Law School. law.cornell.edu
7. Administrative Office of the U.S. Courts. (n.d.). Types of cases. United States Courts. uscourts.gov

Lawyers spend much of their time reading cases - the written opinions in which courts explain how they decided a dispute. Learning to read a case means learning to pull out its useful parts and to reproduce the reasoning yourself. The most widely taught framework for that reasoning is IRAC: Issue, Rule, Application, Conclusion.

Reading a case is not like reading a chapter that summarizes the law. An opinion is an argument written to justify one outcome in one dispute, and the general rule you need is embedded in particulars: names, dates, procedural moves, rhetorical asides. The skill is extraction - finding the question, the rule, and the reasoning inside the story - followed by reconstruction, where you run the same reasoning on new facts. This lesson teaches both halves.

The parts of an opinion

A typical opinion contains the facts (what happened), the procedural history (how the case moved through the courts), the legal issue (the question the court must answer), the holding (the court's answer and the rule it applies), the court's reasoning, and sometimes dicta (remarks not necessary to the decision). A student "briefs" a case by summarizing these parts in a few lines. The heart of the brief is the reasoning, and IRAC is the tool for laying it out.

The document also carries identifying machinery. The caption names the parties and their roles; the citation locates the opinion in the reports - Brown v. Board of Education, 347 U.S. 483 (1954), means volume 347 of the United States Reports, page 483, decided in 1954. When an appellate panel divides, the majority opinion states the law; a concurrence agrees with the result but writes separately, often on different reasoning; a dissent disagrees with the result. Concurrences and dissents bind no one, though a powerful dissent sometimes persuades a later court.

One warning for new readers: commercial publishers add headnotes and summaries above the opinion. These editorial aids are useful finding tools and are not law; more than one embarrassed advocate has cited a headnote the opinion does not support. Always verify against the court's own words.

Holding versus dictum

The most consequential sorting a reader performs is separating the holding from everything else. The holding is the rule actually necessary to resolve the issue presented on the facts before the court; it is what binds later courts under precedent. Dicta - statements in passing, hypotheticals, broad observations about what the law might be in other situations - can be insightful and persuasive, but no court is obliged to follow them.

Suppose a court holds a cafe liable for a customer's fall on a floor mopped without warning signs, and the opinion remarks that outdoor icy walkways "would present a different question." The rule about unmarked wet floors is holding; the remark about ice is dictum, because no icy walkway was before the court. A later advocate can invoke the ice comment only as persuasion. Learning to draw this line takes practice, and reasonable lawyers argue about where it falls - which is precisely why the skill is valuable.

A classic to practice on

Try the anatomy on a famous first-year case, Hawkins v. McGee (New Hampshire, 1929). The facts: a surgeon, seeking the work, repeatedly promised to make a young man's badly scarred hand "a hundred per cent perfect hand" by grafting skin from his chest. The operation left the hand worse - and growing hair. The patient sued, not for negligence, but for breach of the promise itself. The trial court let the jury award damages for pain and for the hand's worsening, and the surgeon appealed.

The issue: could the surgeon's words amount to an enforceable contractual guarantee, and if so, how are damages measured? The holding: a jury could find the repeated assurance was a warranty inviting reliance, and the damages are the difference between the value of the hand as promised and the value of the hand as it resulted.

The reasoning: contract law protects the expectation created by a bargain, so the injured party is put, in money, where performance would have put him. The pain of the operation was the agreed price of the promised result, so it is not separately recoverable. Facts, issue, holding, reasoning - every element of the anatomy is on display in a few pages, which is why generations of first-year students meet this case in their opening week. You will see its damages rule again in the contracts lesson.

The IRAC method

• Issue. State the precise legal question, ideally as a yes/no or narrowly framed problem. Good issues are specific: not "did the defendant do wrong?" but "does leaving a garden hose across a public sidewalk breach the duty of reasonable care?"
• Rule. State the governing legal rule from a statute or precedent. Here you set out the test the law provides, such as the elements that must be satisfied.
• Application. Apply the rule to the specific facts, arguing both sides. This is where analysis happens: you connect each part of the rule to the facts that satisfy or fail it. Most of your effort goes here.
• Conclusion. State the result that the application supports - who wins on this issue and why.

Each step rewards deliberate technique. Frame the issue with the legally significant facts inside it; the frame half-decides the case, since "may a shopkeeper eject a disruptive customer" and "may a shopkeeper use force against a customer" point toward different answers. State the rule as elements or a test, because elements create a checklist for the application. If the rule must be assembled from several cases, synthesize one statement rather than quoting three.

In the application, work element by element, and argue both sides where the facts allow: name the strongest opposing reading of each fact and answer it. This is not decoration; it is how you find weaknesses before an opponent or grader does. Then conclude - commit to the better answer rather than hedging, flagging any facts you would need to know to be confident. Some writers teach the same discipline under other acronyms, such as CREAC, which begins and ends with the conclusion; the underlying moves are identical.

A worked hypothetical

Consider this short problem. Facts: Dana runs a cafe. She mops the floor at noon but posts no warning sign and leaves no cones. Ten minutes later, Pat, a customer, slips on the wet tile and breaks a wrist. Pat sues Dana for negligence. Assume the rule: a person is negligent if (1) they owed a duty of reasonable care, (2) they breached that duty, (3) the breach caused harm, and (4) the plaintiff suffered actual damages. Now apply IRAC.

• Issue: Was Dana negligent in leaving a freshly mopped floor wet and unmarked, causing Pat's injury?
• Rule: Negligence requires duty, breach, causation, and damages, as stated above.
• Application: Duty - a cafe owner owes customers a duty to keep the premises reasonably safe, so duty is met. Breach - a reasonable owner who mops a floor would warn customers with a sign or cones; leaving a wet floor unmarked in a customer area falls below reasonable care, so there is a breach. Causation - Pat slipped precisely because the floor was wet and unmarked, so the breach caused the fall; had a sign been posted or the floor been dry, Pat likely would not have fallen. Damages - a broken wrist is a real physical injury with medical costs, satisfying damages.
• Conclusion: All four elements are met, so Dana is likely liable to Pat for negligence.

Notice how IRAC forces you to march through each element and tie it to the facts. That discipline is what turns a hunch ("the cafe seems at fault") into a legal argument. Once you can do this reliably, you can read a court's opinion and see the same structure at work in the judge's reasoning.

Change the facts, rerun the analysis

IRAC earns its keep when facts shift. Suppose Dana had placed a large yellow "wet floor" cone beside the mopped area. Duty is unchanged, but breach now likely fails: warning of a temporary hazard is what reasonable care demands, and a business need not guarantee dry floors at every moment. Suppose instead the cone was posted but Pat was reading a phone while walking. Breach by Dana may still exist if the warning was inadequate, but Pat's own carelessness enters: under comparative-fault rules, discussed in the torts lesson, Pat's recovery could be reduced by Pat's share of responsibility.

One more variation: Dana mopped, and Pat slipped thirty seconds later, before any reasonable owner could have posted a sign. Now causation and breach entangle - the hazard existed, but reasonable care takes at least a moment to implement. A thoughtful application concedes the point's difficulty and argues it, rather than pretending the element is obvious. Each variation changes one input and traces the output; that habit of controlled experiment is exactly how lawyers test a rule's edges.

Common pitfalls

Watch for five recurring failures. First, the conclusory application: "Dana breached because she was unreasonable" restates the conclusion without connecting facts to standard. Second, the one-sided application that never states the opponent's best argument. Third, the overbroad issue - "is Dana liable?" - which invites wandering. Fourth, treating dicta or a dissent as the rule. Fifth, ignoring procedural posture: an opinion affirming denial of summary judgment decides only that a reasonable jury could find for the plaintiff, not that the plaintiff should win, and briefs that miss this overstate holdings.

Try it: the delivery rider

Facts for your own run: a delivery rider, pedaling fast to meet a deadline, swerves onto a crowded sidewalk to pass a stopped bus and knocks down a pedestrian, breaking her ankle. Local rules prohibit riding on sidewalks. Apply IRAC to a negligence claim before reading on.

Worked analysis. Issue: was the rider negligent in riding onto a crowded sidewalk and striking the pedestrian? Rule: duty, breach, causation, damages; additionally, violating a safety rule designed to protect this class of victims can itself establish breach in many jurisdictions, a doctrine the torts lesson will name negligence per se.

Application: riders owe pedestrians reasonable care (duty); choosing a crowded sidewalk at speed, against a safety rule, falls well below that standard, and the deadline pressure explains but does not excuse it (breach); the pedestrian fell precisely because the rider swerved there (actual cause), and injuring a walker is the foreseeable risk that makes sidewalk riding wrongful (proximate cause); a broken ankle with medical bills is concrete damage. Conclusion: liability is very likely.

If your version argued both sides - perhaps testing whether an emergency, like avoiding a suddenly opened car door, might justify the swerve - you are already reasoning like a lawyer. The next lesson adds the system that makes such reasoning cumulative: precedent.

Sources

1. Levi, E. H. (1948). An introduction to legal reasoning. The University of Chicago Law Review, 15(3), 501. doi.org/10.2307/1597535
2. National Archives. (n.d.). Brown v. Board of Education (1954). Milestone Documents. archives.gov
3. Oyez. (n.d.). Brown v. Board of Education of Topeka (1), 347 U.S. 483 (1954). IIT Chicago-Kent College of Law. oyez.org
4. Legal Information Institute. (n.d.). Holding. Cornell Law School. law.cornell.edu
5. Legal Information Institute. (n.d.). Obiter dictum. Cornell Law School. law.cornell.edu
6. Legal Information Institute. (n.d.). Legal writing. Cornell Law School. law.cornell.edu
7. Hawkins v. McGee, Supreme Court of New Hampshire (1929). (Case discussed in the lesson; cited by name, court, and year only.) find source ↗

Common-law reasoning depends on precedent: the principle that courts should decide new cases consistently with how similar cases were decided before. The doctrine that gives precedent its force is called stare decisis, a Latin phrase meaning "to stand by things decided." Under stare decisis, a court is generally expected to follow the rule laid down in earlier, relevant decisions rather than deciding each case from scratch.

The previous lesson taught you to pull the holding out of a single opinion. This lesson supplies the system that makes those holdings add up. Precedent is what turns thousands of separate decisions into a body of law: each opinion becomes raw material for the next dispute, and judges build with what earlier judges left behind. Using the system well requires answers to four questions. Which decisions must a court follow? Which may it merely consult? What part of a decision does the binding? And how does law governed by precedent nonetheless manage to change? Those questions organize this lesson.

Why follow precedent?

Following precedent serves several values that a legal system prizes:

• Predictability. People and businesses can plan their affairs if they can anticipate how courts will rule.
• Equality. Like cases are treated alike, so outcomes do not turn on which judge a party happens to draw.
• Efficiency. Courts need not re-argue settled questions in every case.
• Legitimacy. Decisions rooted in an established body of law appear principled rather than arbitrary.

Consider what disappears if precedent loses its grip. A business cannot price a contract without knowing whether its standard terms will be enforced; a family cannot plan an estate without knowing whether its will formalities will hold. Justice Louis Brandeis captured the point in a famous observation: in most matters it is more important that the governing rule be settled than that it be settled exactly right, because people can adjust their affairs to a known rule but cannot adjust to uncertainty. Stability is not a mere convenience; it is part of what makes law usable at all.

Precedent also disciplines judges. A judge who must square today's decision with yesterday's cannot simply vote personal preference; the reasons given in past cases become constraints on present ones. And precedent is what makes legal advice possible. When a lawyer tells a client that a contract term will hold up or that a lawsuit is likely to fail, the prediction rests on the assumption that courts will follow the decided cases. Take that assumption away and legal advice becomes guesswork.

Vertical and horizontal stare decisis

Lawyers separate the doctrine into two directions. Vertical stare decisis is the duty of every court to follow the decisions of the courts above it in its own hierarchy. A trial court must apply the holdings of its intermediate appellate court and its supreme court; the intermediate court must apply the holdings of the supreme court. This duty is close to absolute. A lower court may criticize a controlling precedent, and may even predict that the high court will one day abandon it, but it must apply the precedent and leave any overruling to the court that created it.

Horizontal stare decisis describes a court's treatment of its own prior decisions. Highest courts generally follow their own precedents but retain the power to overrule them in compelling circumstances. Intermediate courts often add internal rules of their own: in the federal system, for example, a three-judge court of appeals panel is generally bound by earlier published decisions of its own circuit, which ordinarily only the full court sitting en banc, or the Supreme Court, can displace. The details vary across court systems, but the pattern is constant: a strong duty running downward, and a strong presumption running sideways.

Binding versus persuasive precedent

Not all precedent carries the same weight. A precedent is binding (mandatory) on a court only if it comes from a higher court in the same system whose decisions that court must follow. A precedent is merely persuasive if it comes from a court in another system, a lower court, or a coordinate court - such rulings may be considered for their reasoning but need not be followed. This is why the hierarchy of courts from Module 3 matters so much: a supreme court's holding binds all courts beneath it, while a decision from a court in a different jurisdiction can only persuade.

A crucial refinement: only the ratio decidendi - the essential rule on which the decision rested - is binding. Statements that were not necessary to the outcome, the dicta we met earlier, are at most persuasive. Extracting the ratio from an opinion, and separating it from dicta, is a core lawyering skill.

The universe of persuasive authority is large. Decisions from other states or countries, decisions of lower or coordinate courts, and dicta from any court can all be cited for their reasoning, alongside secondary sources such as treatises and the American Law Institute's Restatements. Persuasive weight is earned, not owed: it depends on the quality of the reasoning, the reputation of the court, and whether many jurisdictions have converged on the same answer. One structural point matters in a federal system: each state's highest court is the final authority on that state's own law, so even a federal court deciding a state-law question follows the state supreme court's rulings.

Finding the ratio: the level of generality

Identifying the ratio decidendi is harder than it sounds, because a holding can be stated at more than one level of generality and the opinion rarely fixes the level for you. Return to the cafe case from the IRAC lesson. Its rule might be narrow: a cafe that mops a floor must post a warning. It might be broader: a business must warn customers of temporary hazards it creates. Broader still: an occupier of premises owes visitors reasonable care. Each statement fits the outcome, yet they control very different ranges of future cases.

Advocates exploit this openness. The lawyer who wants the precedent to govern argues for the broad reading; the lawyer who wants to escape it argues for the narrow one. A useful discipline is the subtraction test: list the facts the court emphasized, and ask whether the outcome would have changed had a fact been absent. Facts that would have changed the outcome are material and belong inside the rule; facts that would not are candidates for exclusion. The level of generality is finally settled not by the original court but by later courts as they apply the case.

Distinguishing and overruling

Stare decisis is strong but not absolute; the system has two safety valves. First, a court may distinguish a precedent by showing that the present case differs in a legally significant way, so the earlier rule does not control. For example, a rule about slips on a wet indoor floor might be distinguished in a case about a natural accumulation of ice outdoors if the differences matter to the duty owed.

Second, in limited circumstances a sufficiently high court may overrule a precedent, declaring the old rule no longer good law - usually when the precedent has proved unworkable, badly reasoned, or out of step with later legal developments. Overruling is done cautiously, because stability is itself a value. Together, distinguishing and overruling let the common law stay consistent while still adapting over time.

Distinguishing has limits. The difference offered must be legally significant: it must matter to the reasons behind the rule, not merely make the cases look different. Indoor mopping and natural outdoor ice may genuinely differ, because an occupier controls one hazard and not the other. That the plaintiff wore a red coat is also a difference, but no rationale of premises liability turns on coat color, so the distinction fails. When courts distinguish a precedent aggressively and repeatedly, the case can be hollowed out - still nominally good law, but confined so tightly to its facts that it controls almost nothing.

Overruling is governed by announced considerations rather than whim. Courts weighing whether to overrule commonly ask whether the precedent's rule has proved unworkable in practice, whether later decisions have eroded its doctrinal foundations, whether the facts or understandings on which it rested have changed, and how severely reliance interests would be damaged. The poorer the precedent scores on the first three, and the lighter the reliance, the stronger the case for departure. That a precedent is wrong is usually treated as necessary but not sufficient; something more must justify the disruption.

The canonical overruling: Plessy and Brown

The most studied overruling in American law began with Plessy v. Ferguson (1896), in which the Supreme Court upheld a Louisiana statute requiring racially separate railway cars. The Court reasoned that state-mandated separation did not deny the equal protection of the laws guaranteed by the Fourteenth Amendment so long as the separate facilities were equal, and it dismissed the claim that enforced separation stamped one race as inferior. Justice John Marshall Harlan dissented alone, insisting that the Constitution tolerates no classes among citizens. "Separate but equal" became the constitutional charter for segregation.

Fifty-eight years later, in Brown v. Board of Education (1954), a unanimous Court led by Chief Justice Earl Warren held that racially separate public schools are inherently unequal and violate the Equal Protection Clause. The Court concluded that segregation in public education generates a sense of inferiority that the Plessy majority had waved away, and it refused to follow the separate but equal doctrine in that field. A line of later decisions applied Brown's principle to other public facilities, and Plessy's doctrine was dismantled.

Brown displays both faces of stare decisis at once. On one side stood more than half a century of precedent and the arrangements built on it; on the other stood a rule resting on a false factual premise and working a continuing injustice. The Court concluded that fidelity to Plessy would perpetuate the error rather than the law, and it overruled openly, in a reasoned opinion, at the level of the hierarchy entitled to do it. That is the doctrine's own account of when departure is legitimate: rare, explained, and made by the court with authority over the precedent.

Reliance and the strength of precedent

Reliance deserves separate attention, because it explains why stare decisis presses harder in some fields than others. Precedents governing property and contract sit near the top: titles have been recorded, loans secured, wills drafted, and businesses organized on the strength of settled rules, so a change punishes exactly the people who obeyed the law. Precedents about procedure or evidence generate thinner reliance, since few people arrange their lives around them, and courts weigh change in those areas more freely.

The source of the rule matters as well. In the federal system, precedents interpreting statutes receive especially strong stare decisis, because a legislature that dislikes the interpretation can amend the statute; its silence is often read as acceptance. Constitutional precedents receive comparatively less protection, because formal amendment is so difficult that the Supreme Court itself is usually the only realistic corrector of its own constitutional mistakes. Brown is the standard illustration. State courts describe their own practices in similar terms, though the details vary from state to state.

Common misconceptions

First misconception: stare decisis freezes the law. It does not. Distinguishing refines rules case by case, high courts occasionally overrule, and legislatures can change most court-made rules by statute at any time. The doctrine slows change and channels it through reasons; it does not forbid it.

Second misconception: any case a lawyer finds can be cited as binding. Pedigree comes first. A decision binds only the courts below the deciding court in the same hierarchy, and only its ratio binds; everything else is persuasive at most. Relatedly, age does not equal weakness: an old precedent repeatedly followed may be far sturdier than a recent one never tested.

Third misconception: one case settles a question. In practice a rule lives in a line of cases: the original decision plus the later ones that apply, narrow, or extend it. Competent lawyers read the line, not just the landmark, because the current shape of the rule is the product of all of them.

Try it: three authorities, one icy walkway

Facts: in the state of Arcadia, a tenant slips on natural ice on an apartment walkway the landlord never treated, and sues. The trial court has three authorities. Authority one: a 1990 Arcadia Supreme Court holding that a landlord who undertakes snow removal and performs it carelessly is liable for resulting injuries. Authority two: a decision of the neighboring state's supreme court holding landlords liable for all natural ice accumulations. Authority three: a remark in a 2010 Arcadia Supreme Court opinion about parking lots, observing that walkway ice "might warrant a broader duty." Classify each authority and predict the outcome before reading on.

Worked analysis. Authority one is binding: it comes from the highest court of the same hierarchy. But it is distinguishable, because its ratio covers landlords who undertake removal and do it badly, and this landlord undertook nothing; the material fact on which the rule turned is absent. Authority two is persuasive only. It comes from a different jurisdiction, so the trial court may study its reasoning but owes it no obedience.

Authority three is dictum. The 2010 case concerned parking lots, so the remark about walkways was not necessary to its decision; it binds no one, though it hints where the Arcadia Supreme Court might be headed. The trial court will therefore likely rule for the landlord under existing law, while noting that only the supreme court can adopt the broader duty. If the tenant appeals and the high court takes the case, the justices face the true stare decisis question: distinguish, follow, or extend. That is the machinery of precedent operating exactly as designed.

In sum, precedent converts individual judgments into a system: vertical stare decisis makes hierarchy meaningful, horizontal stare decisis keeps courts consistent, the ratio marks what binds, and distinguishing and overruling supply controlled flexibility. With the method of cases in hand, the course turns from how lawyers reason to what the major fields of law provide, beginning with the law of promises: contracts.

Sources

1. Schauer, F. (1987). Precedent. Stanford Law Review, 39(3), 571. doi.org/10.2307/1228760
2. Legal Information Institute. (n.d.). Stare decisis. Cornell Law School. law.cornell.edu
3. Legal Information Institute. (n.d.). Ratio decidendi. Cornell Law School. law.cornell.edu
4. Plessy v. Ferguson, 163 U.S. 537 (1896). Legal Information Institute, Cornell Law School. law.cornell.edu
5. Oyez. (n.d.). Brown v. Board of Education of Topeka (1), 347 U.S. 483 (1954). IIT Chicago-Kent College of Law. oyez.org
6. Planned Parenthood of Southeastern Pennsylvania v. Casey, 505 U.S. 833 (1992). (Cited for its discussion of reliance and the factors bearing on overruling.) Legal Information Institute, Cornell Law School. law.cornell.edu
7. Legal Information Institute. (n.d.). Precedent. Cornell Law School. law.cornell.edu

Contract law is the branch of private law that makes certain promises enforceable, so that people and businesses can rely on one another and plan ahead. Not every promise is a contract - a casual promise to meet a friend for coffee is not enforceable - but when the required elements are present, the law will stand behind the agreement.

Contracts are the legal infrastructure of cooperation. Every purchase, paycheck, lease, loan, insurance policy, and software license is a contract, which means most people enter several before lunch without noticing. The law's task is selective: it must enforce genuine bargains while filtering out jokes, casual social promises, and agreements extracted by force or fraud. The elements of formation are that filter. This lesson covers formation, the objective test for whether an agreement was really made, the writing requirements known as the statute of frauds, and what happens on breach, ending with a full IRAC problem.

Forming a contract

Most legal systems require several elements for a valid contract:

• Offer. One party proposes definite terms, showing a willingness to be bound if the other accepts.
• Acceptance. The other party agrees to those terms. Offer and acceptance together produce a "meeting of the minds," or mutual assent.
• Consideration. Each side gives something of value - a payment, a good, a service, or a promise. Consideration is what distinguishes an enforceable bargain from a one-sided gift promise. (Some systems reach a similar result through other doctrines, but the bargain idea is the common-law core.)
• Capacity and legality. The parties must have legal capacity to contract (for instance, not be minors or lack mental capacity), and the contract's purpose must be lawful. An agreement to do something illegal is not enforceable.

Take the elements one at a time. An offer must be definite enough that a court could tell what was promised, and it must show commitment rather than mere interest. "I might sell my car for around five thousand" opens a conversation; "I will sell you my car for five thousand dollars, payable Friday" invites a contract. In American law, advertisements are generally treated as invitations to deal rather than offers, since a store cannot sell one advertised item to every reader; but a sufficiently definite promise to the public can be an offer.

The classic illustration is the English case Carlill v. Carbolic Smoke Ball Co. (1893). A company advertised that it would pay 100 pounds to anyone who used its influenza remedy as directed and still caught the flu, adding that it had deposited 1,000 pounds in a bank to show its sincerity. Mrs. Carlill used the smoke ball, caught influenza, and sued. The court enforced the promise: the advertisement was an offer of a unilateral contract, accepted not by a reply but by performing the requested act, and the bank deposit showed the promise was meant seriously.

Acceptance is the offeree's agreement to the offer's terms. Under the traditional rule, the acceptance must mirror the offer; a reply that changes the terms is a counteroffer, which kills the original offer and proposes a new one. The offeror is master of the offer and may prescribe how acceptance must occur. An offer can also die before acceptance: the offeror may revoke it, the offeree may reject it, or it may simply lapse after a reasonable time. Silence, standing alone, is generally not acceptance.

Consideration asks whether each side's promise or performance was bargained for - sought by the other in exchange. Courts police the existence of a bargain, not its adequacy; a genuine exchange of a farm for a single peppercorn would qualify, and lopsided prices alone do not void contracts. What fails the test: past consideration (a promise to pay for a benefit already received), a promise to do what one is already legally bound to do, and illusory promises that commit to nothing. A safety valve called promissory estoppel lets courts enforce some promises without consideration when the promisee reasonably relied to their detriment; its scope varies by state.

Many contracts can be oral, but some kinds - often those involving land or large sums - must be in writing to be enforceable, under rules requiring written evidence for certain transactions.

The objective theory of assent: Lucy v. Zehmer

How does a court know the parties actually agreed? Not by reading minds. The rule is the objective theory of contract: assent is judged by a party's outward words and acts, as a reasonable person would understand them, not by secret intentions. The canonical case is Lucy v. Zehmer, decided by Virginia's highest court in 1954. One December evening in 1952, over drinks at the restaurant Zehmer ran, W. O. Lucy pressed A. H. Zehmer to sell his farm. Zehmer wrote on the back of a restaurant check that he and his wife agreed to sell the Ferguson farm to Lucy for 50,000 dollars, and both Zehmers signed.

Zehmer later insisted the writing was a joke between drinking companions, made to call what he took to be Lucy's bluff. The court enforced the contract anyway. The discussion had lasted some forty minutes, the terms were rewritten to include the wife and to address title, and nothing outward marked the transaction as jest. Because Lucy reasonably believed the deal was serious, Zehmer's undisclosed intention did not matter. And since land is unique, the remedy was specific performance: the Zehmers were ordered to convey the farm.

The objective theory is what makes mass contracting possible. Strangers can deal at a distance because each may rely on what the other says and does; a subjective test would let any regretful party escape by claiming a hidden reservation. The theory has boundaries: an offer a reasonable person would recognize as an obvious joke creates no power of acceptance, and assent procured by fraud, duress, or from someone lacking capacity can be undone. But within those limits, you are bound by what you reasonably appear to agree to - which is one reason lawyers tell clients to read before signing.

The statute of frauds

The writing requirement has a name: the statute of frauds, descended from a 1677 English act and carried, in some version, into the law of every American state. The idea is that certain promises are too consequential, and too easy to fabricate, to rest on memory alone. Commonly covered categories include contracts for the sale of land, agreements that cannot be performed within one year, promises to answer for another person's debt, and, under the Uniform Commercial Code, contracts for the sale of goods at or above a threshold price - 500 dollars in the classic version. Categories and details vary by state.

The statute demands a writing that records the essential terms and is signed by the party against whom enforcement is sought; a memo, a signed letter, or under modern law in most states an electronic record can suffice. Because a strict statute can defeat honest oral bargains, courts developed softening doctrines - part performance of a land contract, admissions in court, and reliance can take a case outside the rule, with the details again varying by jurisdiction. The practical lesson is simpler than the doctrine: for anything that matters, put the agreement in writing, not because law always requires it, but because proof beats memory.

Breach and remedies

A breach of contract occurs when a party fails to perform a duty the contract requires without a lawful excuse. When that happens, the law provides a remedy aimed, in principle, at putting the injured party in the position they would have occupied had the contract been performed. The usual remedy is damages - money to cover the loss the breach caused.

In special cases where money is inadequate (for example, a contract to sell a unique item such as a specific parcel of land), a court may order specific performance, compelling the breaching party actually to perform. The law generally expects the injured party to mitigate, that is, to take reasonable steps to limit their losses rather than let them pile up.

The standard measure is called expectation damages, and you have already met its most famous illustration. In Hawkins v. McGee, from the IRAC lesson, the surgeon promised a perfect hand and delivered a worse one; damages were the value of the hand as promised minus the value of the hand as it resulted, with the operation's pain excluded because pain was the price the patient agreed to pay. That is expectation in a sentence: give the injured party the value of the promised position, no less and no more. Contract damages compensate; they are not designed to punish, and punitive damages are generally unavailable for ordinary breach.

Three limits keep damages within reason. Foreseeability comes from Hadley v. Baxendale (1854): a mill's crankshaft broke, the carrier delayed delivering it to the engineers who were to use it as a pattern for a new one, and the mill stood idle longer than expected. The court denied the lost profits because the carrier had not been told the mill would sit stopped, and it limited damages to losses arising naturally from breach or within both parties' contemplation when they contracted. Certainty adds that speculative profits cannot be recovered, and mitigation, met above, subtracts losses the injured party could reasonably have avoided.

Expectation is not the only measure. A party who cannot prove expected profits may recover reliance damages, the out-of-pocket costs of performance, or restitution, the value of benefits conferred on the breacher. And specific performance has its own boundary: courts will not order a person to perform personal services, both because supervision is impractical and because compelled labor is offensive to liberty. A singer who abandons a concert owes money, not a command performance.

A quick example

Suppose Maria agrees in a signed writing to sell Sam 100 chairs for a set price, delivery next month, and Sam pays a deposit. There is an offer, acceptance, and consideration (chairs for money), the parties have capacity, and the purpose is lawful - a valid contract. If Maria then refuses to deliver without excuse, she has breached. Sam can recover damages, typically the extra cost of buying comparable chairs elsewhere, and he is expected to mitigate by making a reasonable substitute purchase rather than simply demanding an open-ended sum. This example shows the life cycle of a contract: formation, performance or breach, and remedy.

Common misconceptions

First misconception: a contract must be a formal signed document. In fact most contracts are oral or arise from conduct; handing over money for coffee forms one without a word. Writing is legally required only for statute-of-frauds categories, though it is prudent everywhere.

Second misconception: breaking a contract is illegal, or even criminal. Breach is neither a crime nor a punishable wrong; it is a failure that triggers a duty to compensate. The law's aim is to protect the value of the bargain, not to condemn the breacher, which is why punitive damages are generally unavailable.

Third misconception: an unread contract cannot bind you. Under the objective theory, signing manifests assent whether or not you read the terms, and courts routinely enforce unread agreements. The doctrine has limits - fraud, lack of capacity, unconscionable terms, and consumer-protection statutes that vary by jurisdiction - but "I did not read it" is not, by itself, a defense.

Try it: the wedding photographer

Facts: Elena signs a written agreement with Marco, a photographer, for wedding photography on June 10 at a price of 2,000 dollars, and pays a 500 dollar deposit. On June 1, Marco cancels to take a better-paying corporate event on the same date. The only comparable photographer still available charges 2,600 dollars, and Elena hires her. Elena sues Marco. Run the full IRAC before reading the analysis.

• Issue: Is Marco liable for breach of contract, and if so, what can Elena recover?
• Rule: A contract requires offer, acceptance, consideration, capacity, and legality. Breach is nonperformance without lawful excuse. Expectation damages put the injured party where performance would have, limited by foreseeability, certainty, and mitigation; specific performance is unavailable for personal services.
• Application: Formation is clean: a signed writing with definite terms shows offer and acceptance, and services for money is consideration. Marco's cancellation is a breach; a more profitable opportunity is not a lawful excuse. Elena mitigated by hiring the comparable substitute. Her expectation loss is the 600 dollar difference between the substitute's price and the contract price, plus return of her 500 dollar deposit, since she should pay net 2,000 for the photography she was promised. The substitute cost was plainly foreseeable and certain.
• Conclusion: Marco is liable for about 1,100 dollars - the deposit back plus the 600 dollar cover difference. No court will order him to shoot the wedding, and no punitive award applies.

Notice what the analysis excluded. Elena's anger at Marco does not increase damages, and emotional distress is generally not compensable in contract, with narrow exceptions that vary by jurisdiction. The measure is the value of the promise, coolly computed - which is exactly what makes contract remedies predictable enough for commerce to rely on.

In sum, contract law enforces bargains evidenced by offer, acceptance, and consideration, judges assent objectively, requires writing for a few high-stakes categories, and on breach substitutes the money value of performance. Its duties are chosen. The next lesson turns to obligations nobody chooses: the duties tort law imposes on us all.

Sources

1. Fuller, L. L., & Perdue, W. R. (1936). The reliance interest in contract damages: 1. The Yale Law Journal, 46(1), 52. doi.org/10.2307/791632
2. Legal Information Institute. (n.d.). Contract. Cornell Law School. law.cornell.edu
3. Legal Information Institute. (n.d.). Consideration. Cornell Law School. law.cornell.edu
4. Legal Information Institute. (n.d.). Statute of frauds. Cornell Law School. law.cornell.edu
5. Legal Information Institute. (n.d.). Expectation damages. Cornell Law School. law.cornell.edu
6. Uniform Commercial Code, Article 2 (Sales). Legal Information Institute, Cornell Law School. law.cornell.edu
7. American Law Institute. (n.d.). Publications. (Publisher of the Restatement of the Law, Contracts, the influential secondary restatement of contract doctrine described in this lesson.) ali.org
8. Carlill v. Carbolic Smoke Ball Co., Court of Appeal of England and Wales (1893); Hadley v. Baxendale, Court of Exchequer (1854); Lucy v. Zehmer, Supreme Court of Appeals of Virginia (1954). (Cases discussed in the lesson; cited by name, court, and year only.) find source ↗

A tort is a civil wrong - other than a breach of contract - for which the law provides a remedy, usually damages. Tort law lets a person harmed by another's wrongful conduct recover compensation. It differs from criminal law, which punishes offenses against society, and from contract law, which enforces voluntary agreements; a tort duty is imposed by law on everyone, whether or not the parties ever agreed to anything.

Tort law is, at bottom, the law of injuries and who must pay for them. A surgeon slips, a driver glances at a phone, a factory's chemical leaks, a false rumor spreads: in each case someone is worse off, and the legal question is whether the loss stays where it fell or shifts to the person who caused it. The answer turns on categories. This lesson works through negligence and its elements, the causation puzzle made famous by a package of fireworks, the shortcut called res ipsa loquitur, the intentional torts, strict liability, and the defenses, ending with a full IRAC problem.

Negligence

The most important tort is negligence, which we met in the IRAC lesson. Negligence is not about intending harm; it is about failing to exercise the care a reasonable person would use, and thereby injuring someone. Its four familiar elements are:

1. Duty - the defendant owed the plaintiff a duty of reasonable care.
2. Breach - the defendant failed to meet that standard of care.
3. Causation - the breach in fact caused the harm, and the harm was a foreseeable result.
4. Damages - the plaintiff suffered actual loss or injury.

The touchstone is the reasonable person: an objective standard asking what a reasonably careful person would have done in the situation. Falling below that standard, and causing harm as a result, is negligence.

The standard's objectivity was settled early, in the English case Vaughan v. Menlove (1837). The defendant stacked a hay rick near his neighbor's boundary and was warned repeatedly that it might ignite; he answered that he would chance it. The rick caught fire and burned the neighbor's cottages. His defense was that he had honestly used his own best judgment. The court rejected it: the measure is the caution a person of ordinary prudence would observe, not the defendant's personal capacities. Good faith is not the test; reasonable conduct is.

Objectivity has reasons. An external standard gives everyone equal security, since your safety does not depend on how careless your neighbor happens to be, and it spares juries from probing sincerity. The law does calibrate it in places: professionals are held to the knowledge and skill of their profession, children are usually judged against a reasonable child of similar age unless engaged in an adult activity like driving, and physical limitations such as blindness are built into the standard. Mental shortcomings, by contrast, generally are not - the ordinary rule in American courts, though formulations vary.

Causation: two questions, and Palsgraf

Causation splits into two distinct questions. Actual cause asks whether the harm would have happened but for the breach: if the pedestrian would have been struck even had the driver been sober, the drinking did not actually cause the injury. Proximate cause asks a normative question: is the harm close enough to the risk that made the conduct careless that liability is fair? A breach can set off a chain of consequences no one would have predicted, and proximate cause is where the law decides how far down the chain responsibility runs.

The most famous proximate cause case is Palsgraf v. Long Island Railroad Co. (New York, 1928). As the opinion recounts the facts, a man ran to catch a moving train while carrying a small package wrapped in newspaper. Two railroad employees, one aboard the car and one on the platform, pushed and steadied him as he jumped on, and the package fell to the rails. It held fireworks, which exploded. The shock, the court said, knocked down a set of scales at the other end of the platform, injuring Helen Palsgraf, who stood waiting for a different train.

Chief Judge Benjamin Cardozo, writing for a divided court, held the railroad not liable to Mrs. Palsgraf. Negligence, he reasoned, is relational: a defendant owes care only against risks that are foreseeable, and to persons within the zone of danger those risks create. Nothing about a plain newspaper-wrapped package suggested peril to a bystander far down the platform, so no duty to her was breached. Judge William Andrews dissented: everyone owes the world care, and the real question is proximate cause, a practical judgment about how far consequences should carry liability.

Both opinions remain teaching tools because modern courts use both moves. Some frame foreseeability as a limit on duty, others as the core of proximate cause, but either way the unexpected victim and the wildly unexpected consequence usually fall outside liability. One refinement is settled: the eggshell plaintiff rule. If the type of harm is foreseeable, the defendant is liable for its full extent, even if a hidden frailty made this victim's injuries surprisingly severe. You take your victim as you find them.

Res ipsa loquitur

Proving breach usually requires evidence of what the defendant did wrong, but sometimes the accident is the evidence. In Byrne v. Boadle (England, 1863), a barrel of flour fell from the upper window of a flour dealer's warehouse and struck a pedestrian walking below. No witness could say how it came loose. The court held the plaintiff could recover anyway: barrels do not leap from warehouses when proper care is taken, so the event itself bespoke negligence. The doctrine's Latin name, res ipsa loquitur, means the thing speaks for itself.

As courts apply it, the doctrine typically asks whether the accident is of a kind that ordinarily does not occur without negligence, whether the instrumentality was under the defendant's control, and whether the plaintiff's own conduct is ruled out as the cause. When those conditions hold, the jury may infer breach without direct proof, though the defendant can rebut the inference. Its procedural effect - a permissible inference in most states, a stronger presumption in a few - varies by jurisdiction. Res ipsa matters most where the defendant alone knows what happened, as in surgical mishaps involving unconscious patients.

Intentional torts and strict liability

Not all torts are about carelessness. Intentional torts involve deliberate conduct that invades a protected interest - examples include battery (intentional harmful or offensive contact), assault (intentionally causing fear of imminent harmful contact), trespass (intentional intrusion on another's land), and defamation (a false statement that harms reputation). Here the wrong lies in the intent to do the act, not merely in a lapse of care.

Intent has a precise meaning worth pausing on: the defendant must intend the contact or invasion, or know it is substantially certain to result - intending injury is not required. The point is captured by Garratt v. Dailey (Washington, 1955), in which a five-year-old pulled a chair away as his aunt was sitting down and she fell and fractured her hip. The court sent the case back to determine whether the boy knew with substantial certainty that she would try to sit where the chair had been. A prank with no wish to harm can still be a battery.

A third category is strict liability, where a defendant can be liable for certain harms even without intent or negligence, simply because the activity is treated as one whose risks the actor should bear. Classic examples include keeping dangerous wild animals or carrying on abnormally hazardous activities, and, in many systems, liability for defective products. Strict liability shifts the focus from the defendant's state of mind to the nature of the activity and who is best placed to bear its risks.

Products liability deserves a special note because it touches nearly everyone. In most American states, a commercial seller of a product with a manufacturing defect, an unreasonably dangerous design, or inadequate warnings can be held liable to injured users without proof of negligence. The doctrine grew in the twentieth century on a simple logic: manufacturers create the risk, profit from it, can spread losses through pricing and insurance, and are best positioned to make products safer. Details differ across states, but the strict-liability core is widely shared.

Defenses and purposes

Tort defendants can raise defenses. For negligence, a common one is comparative fault, under which a plaintiff's own carelessness reduces their recovery in proportion to their share of the blame; another is assumption of risk, where a plaintiff knowingly accepted a danger. Tort law as a whole serves two broad aims: to compensate the injured and to deter unreasonably risky conduct by making people internalize the costs of the harm they cause. Understanding these categories - negligence, intentional torts, and strict liability - lets you sort most civil wrongs and see what each requires.

The treatment of plaintiff fault has a history worth knowing. The older rule, contributory negligence, barred recovery entirely if the plaintiff was careless at all: one percent of fault meant zero compensation. Its harshness drove reform, and during the twentieth century nearly all states replaced it with comparative fault. Some use a pure version, in which even a mostly at-fault plaintiff recovers a reduced share; others use modified versions that cut off recovery once the plaintiff's fault reaches half. A small number of jurisdictions still apply the old complete bar. Which regime governs depends entirely on the state, so this is a question lawyers check early.

Damages have their own architecture. Compensatory damages cover economic losses such as medical bills and lost wages, and noneconomic harm such as pain. Because actual damage is an element of negligence, there is no negligence claim for a harmless near miss; intentional torts differ, since even a dignitary violation without physical injury can support at least nominal damages. Punitive damages are reserved for outrageous conduct and are limited by statute and constitutional doctrine, with caps varying by state. In practice, liability insurance pays most awards, and the great majority of tort claims settle before trial.

Common misconceptions

First misconception: every injury has a defendant. Tort law is not an insurance scheme; unless the elements of some tort are met, a loss stays with the person who suffered it. Genuine accidents without fault, outside the strict-liability pockets, produce no liability at all.

Second misconception: intending no harm means no intentional tort. As Garratt v. Dailey shows, intending the contact, or knowing it is substantially certain, suffices; motive and malice go to damages, not to whether the tort occurred.

Third misconception: a person cleared of a crime cannot be sued. Criminal and tort proceedings are separate tracks with different parties and different burdens of proof, so an acquittal - a failure of proof beyond a reasonable doubt - does not prevent a civil jury from finding liability on a preponderance of the evidence for the same conduct.

Try it: the falling flowerpot

Facts: Priya keeps heavy flowerpots on her third-floor balcony railing. One pot wobbled in wind before and once fell into the empty courtyard; she put it back without securing it. Today a gust knocks it off, and it strikes Theo, who is walking on the public sidewalk below while reading his phone. He suffers a concussion. Theo sues Priya for negligence. Run the IRAC yourself before reading on.

• Issue: Is Priya liable to Theo for negligence, and does his phone use change the result?
• Rule: Negligence requires duty, breach, actual and proximate causation, and damages; under comparative fault, a plaintiff's own negligence reduces recovery, and in a few jurisdictions still bars it.
• Application: Priya owes reasonable care to passersby below, a foreseeable class (duty). Leaving an unsecured heavy pot on a railing after it had already fallen once is exactly what a reasonable person would not do (breach). But for the unsecured pot, Theo walks on unharmed, and a pot striking a pedestrian is the precise risk that made the conduct careless, so both causation questions are satisfied. A concussion is real damage. Theo's phone reading invites a comparative-fault argument, but walking while distracted did not put him anywhere a careful walker could not be, so his share of fault should be small.
• Conclusion: Priya is liable; Theo's recovery may be modestly reduced, with the exact effect depending on the state's comparative-fault rule.

Notice the shape of the analysis: the prior fall did heavy work on breach, because notice of a hazard makes ignoring it unreasonable; and Palsgraf did quiet work on duty, because a sidewalk pedestrian is squarely within the zone of danger an unsecured balcony pot creates. Change Theo to a burglar on Priya's balcony and the duty question transforms.

In sum, tort law sorts injuries into three bins - carelessness, intention, and activity-based strict liability - and asks in each whether the defendant's conduct, the causal chain, and the plaintiff's own behavior justify shifting the loss. Next the course turns from harms to holdings: the law of property.

Sources

1. Palsgraf v. Long Island R.R. Co., 248 N.Y. 339 (1928). New York State Unified Court System, Official Reports archive. nycourts.gov
2. Prosser, W. L. (1953). Palsgraf revisited. Michigan Law Review, 52(1), 1. doi.org/10.2307/1285361
3. Legal Information Institute. (n.d.). Negligence. Cornell Law School. law.cornell.edu
4. Legal Information Institute. (n.d.). Reasonable person. Cornell Law School. law.cornell.edu
5. Legal Information Institute. (n.d.). Res ipsa loquitur. Cornell Law School. law.cornell.edu
6. Legal Information Institute. (n.d.). Strict liability. Cornell Law School. law.cornell.edu
7. Vaughan v. Menlove, Court of Common Pleas (1837); Byrne v. Boadle, Court of Exchequer (1863); Garratt v. Dailey, Supreme Court of Washington (1955). (Cases discussed in the lesson; cited by name, court, and year only.) find source ↗

Property law governs the rights people have in things - what they may own, use, exclude others from, and transfer. Property is fundamental to economic life: it lets people invest in and trade resources with confidence that their rights will be protected.

Though we speak of owning things, property is really a set of legal relationships among people about things: my ownership of a bicycle is a claim against everyone else, not a bond with the bicycle. That perspective explains this lesson's itinerary. We sort property into its two great categories, ask how ownership begins with a famous fox, unpack the bundle of rights, slice ownership across time with estates, survey the landlord-tenant relationship, meet the doctrine by which a trespasser can become an owner, and map the smaller rights called easements, ending with an IRAC problem about a fence in the wrong place.

Real and personal property

A basic division separates two kinds of property:

• Real property is land and things permanently attached to it, such as buildings and, generally, fixtures. Because land is durable and location-specific, real property has its own detailed rules for sale, recording of ownership, and resolving boundary disputes.
• Personal property is everything else that can be owned - movable things (a car, a laptop, furniture) called tangible personal property, and intangible rights such as a bank balance, shares, or intellectual property like a copyright.

The line matters because the rules differ on each side. Transfers of land require formal deeds and are recorded in public registries; personal property usually changes hands by simple delivery or sale. The boundary cases are fixtures: personal property so attached to land that the law treats it as part of the realty. A furnace bolted into a house sells with the house; a freestanding bookshelf does not. Courts look to how firmly the item is attached and what the parties appear to have intended, and the details vary by jurisdiction, which is why real estate contracts often list what stays.

First possession: Pierson v. Post

Every chain of ownership starts somewhere, and the law's oldest answer is possession. The canonical case is Pierson v. Post (New York, 1805). Lodowick Post, hunting with horse and hounds, was pursuing a fox across a wild, unpossessed stretch of beach. Jesse Pierson, though he knew Post was in chase, killed the fox himself and carried it off. Post sued, arguing that his pursuit gave him a right to the animal. The question for the court was precise: what act converts a wild animal, which belongs to no one, into property?

The majority held that mere pursuit gives no property right. Occupancy requires capture, or at least mortally wounding the animal and continuing the chase; anything less, the court reasoned, would breed endless quarrels, while a clear capture rule gives everyone a certain signal of ownership. Judge Livingston dissented: he would have asked the custom of hunters, which rewarded the pursuer and scorned the interloper. The disagreement is the enduring lesson - clear, administrable rules versus flexible standards attuned to community practice - a choice that recurs throughout property law.

The first-possession principle radiates outward. A finder of lost goods acquires rights against everyone except the true owner: in the old English case Armory v. Delamirie (1722), a chimney sweep's boy who found a jewel could recover its value from the goldsmith whose apprentice kept it, because prior possession beat everyone but the rightful owner. Possession, in short, is a real but relative title.

The bundle of rights

Lawyers often describe ownership as a bundle of rights rather than a single all-or-nothing thing. The bundle can include the right to use the thing, the right to exclude others from it, the right to transfer it (by sale or gift), and the right to the income it produces. These sticks in the bundle can be separated.

A landlord who leases an apartment, for example, keeps ownership but hands the tenant the right to possess and use the space for a time; an owner can grant someone an easement (a limited right to use land, such as a right of way) while keeping the rest. Seeing property as a divisible bundle explains how renting, lending, licensing, and shared ownership all work.

Estates in land: slicing ownership across time

The common law also slices the bundle by time, through interests called estates. The largest is the fee simple: ownership of potentially infinite duration, inheritable and freely transferable - what people ordinarily mean when they say they own a home. A life estate is ownership measured by a life: the life tenant holds the property until death, after which it passes to whoever holds the remainder, or returns to the grantor. A will might leave a house to a surviving spouse for life, remainder to the children; both interests are real, present legal rights, even though only one includes possession today.

Two plain-English notes complete the picture. First, a life tenant may use the property but not commit waste - gutting the house or letting it rot injures the remainder holders, and they can sue to stop it. Second, leases are estates too: a tenant for a year holds a genuine, if temporary, property interest, not a mere permission. Slicing ownership across time is simply more bundle-splitting, and it is how families pass homes down, how developers structure long-term ground leases, and how future interests can be bought and sold before possession ever arrives.

Landlord and tenant

The lease is the most common property relationship most people will ever have, and it is a hybrid: part conveyance, because the landlord transfers the possession stick for a term, and part contract, because each side takes on promises, above all the tenant's promise of rent. Modern law adds protections that the bare conveyance never carried. In most states, residential leases include an implied warranty of habitability: the premises must be fit for human habitation - heat, water, structural safety - and the tenant cannot bargain the warranty away. Remedies for breach, such as withholding rent or repairing and deducting, vary considerably by state.

Possession is also protected on the way out. In most jurisdictions a landlord cannot retake an apartment by changing the locks or removing doors; eviction requires notice and a court proceeding, and self-help removal is widely prohibited. Security deposits are regulated too, with deadlines and itemization rules that differ from state to state. The pattern across all of this is the bundle again: the tenant's temporary sticks receive real legal armor, and the landlord's reversion waits behind them.

Adverse possession

Property law contains a doctrine that surprises nearly everyone: a trespasser who occupies land long enough can become its owner. Adverse possession generally requires possession that is (1) actual - really using the land, (2) open and notorious - visible enough that an attentive owner would notice, (3) exclusive - not shared with the owner or the public, (4) hostile - without the owner's permission, and (5) continuous for the statutory period. The period and details vary widely by state: some require about five years plus payment of property taxes, others demand ten, fifteen, or twenty or more.

The doctrine's logic is not a celebration of trespass. Limitation periods quiet stale claims and let titles match long-settled facts on the ground; the doctrine protects people who have built lives around boundaries everyone treated as real, and it prods owners to pay attention to their land. Note the leverage of the hostility element: possession with permission never ripens into ownership, so a neighborly written license stops the clock entirely. In modern practice the doctrine appears less in dramatic land grabs than in quiet boundary disputes - fences, driveways, and garden strips a few feet over the line.

Easements and servitudes

Not every land right is possession. An easement gives its holder a limited right to use another's land: a driveway across a neighbor's lot, a utility's right to run lines. Easements appurtenant benefit a neighboring parcel and pass automatically with it; easements in gross benefit a person or company, like the utility. They arise by express grant, by implication from prior use, by necessity when a parcel would otherwise be landlocked, or by prescription - long, open, adverse use, adverse possession's cousin for use rather than ownership. Recorded covenants, like the homeowners association rules from the first lesson, similarly bind later buyers who take with notice.

Acquiring and transferring property

Property can be acquired and moved in several standard ways:

• Purchase. The most common route - a sale transfers ownership in exchange for payment, often documented (for land, by a deed).
• Gift. A voluntary transfer without payment, generally requiring an intent to give and delivery.
• Inheritance. Property passes on death by a valid will or, if there is none, under default rules of intestate succession.
• Creation or discovery. Making a new thing, or in some cases finding and taking control of something unowned, can establish ownership.

Because ownership can be complex and contested, legal systems maintain ways to prove and protect it, such as public registries for land titles and rules for resolving competing claims. The core idea to carry forward is that property is a set of legally protected relationships between a person and a thing - and, really, between people about a thing - captured well by the image of a bundle of rights that can be held whole or divided.

Common misconceptions

First misconception: ownership is absolute. Even a fee simple is hedged by public limits - zoning, environmental rules, nuisance law, taxes, and eminent domain, under which government may take property for public use upon payment of just compensation, as the United States Constitution requires. Ownership is the bundle minus the public's reserved sticks.

Second misconception: possession is nine-tenths of the law. Possession is powerful - it starts title in Pierson, arms the finder in Armory, and can ripen into ownership by adverse possession - but against a demonstrable true owner within the limitation period, title wins. The saying describes leverage, not law.

Third misconception: a handshake can sell land. The statute of frauds from the contracts lesson requires land sale agreements to be in writing, and an actual conveyance requires a delivered deed, ordinarily recorded. Oral land deals fail in nearly every circumstance, which is precisely why the writing rule exists.

Try it: the shifted fence

Facts: a previous owner of Nora's lot built a fence three feet over the boundary onto Omar's land. Nora bought in 2010, and ever since has gardened the strip, maintained the fence, and built a small shed there. Omar never gave anyone permission and never checked his line. In 2024 a survey reveals the encroachment, and Omar demands the strip back. The statutory period in this state is ten years. Analyze Nora's adverse possession claim with IRAC before reading on.

• Issue: Has Nora acquired title to the three-foot strip by adverse possession?
• Rule: Adverse possession requires actual, open and notorious, exclusive, and hostile possession, continuous for the statutory period - here ten years - with elements and period varying by state.
• Application: Gardening, fence maintenance, and a shed are actual use of the kind an owner would make. A standing fence and shed are visible to any attentive owner, so the possession is open and notorious, and Nora alone used the strip, satisfying exclusivity. No permission was ever given, so the possession is hostile; in most states her state of mind is irrelevant, though a minority require a good-faith belief or claim of right, which could matter. Her own fourteen years exceed the ten-year period, so she need not even add her predecessor's time, though tacking successive possessions is generally allowed.
• Conclusion: Nora likely owns the strip; Omar's demand comes years too late.

Now flip one fact: suppose Omar had written in 2012, "You are welcome to use the strip for now." Permission destroys hostility, the clock never runs, and Omar keeps his land. The doctrine rewards attentive owners, and a single letter is cheaper than a lawsuit.

In sum, property law assigns beginnings (first possession), structure (the bundle and estates), relationships (landlord and tenant, easements), and endings and corrections (transfer, inheritance, adverse possession). With private rights mapped, the course turns to the state's most severe instrument: criminal law.

Sources

1. Rose, C. M. (1985). Possession as the origin of property. The University of Chicago Law Review, 52(1), 73. doi.org/10.2307/1599571
2. Merrill, T. W., & Smith, H. E. (2001). What happened to property in law and economics? The Yale Law Journal, 111(2), 357. doi.org/10.2307/797592
3. Legal Information Institute. (n.d.). Property. Cornell Law School. law.cornell.edu
4. Legal Information Institute. (n.d.). Fee simple. Cornell Law School. law.cornell.edu
5. Legal Information Institute. (n.d.). Adverse possession. Cornell Law School. law.cornell.edu
6. Legal Information Institute. (n.d.). Easement. Cornell Law School. law.cornell.edu
7. Pierson v. Post, Supreme Court of Judicature of New York (1805); Armory v. Delamirie, Court of King's Bench (1722). (Cases discussed in the lesson; cited by name, court, and year only.) find source ↗

Criminal law defines conduct that is treated as an offense against society and sets the conditions for punishing it. Because a conviction can take away liberty, criminal law is hedged with strict requirements: the state must prove its case beyond a reasonable doubt, and it must usually establish two distinct elements for each offense.

Three features set criminal law apart from everything studied so far. The plaintiff is the state itself, prosecuting in the community's name whether or not the victim wishes. The outcome is punishment and public condemnation, not mere compensation. And because the stakes include liberty, the defendant receives protections no civil litigant enjoys. This lesson covers the two building blocks of every offense, the mental-state ladder, how homicide law grades killings, the constitutional safeguards that surround prosecution, and the main defenses, closing with an IRAC problem about a fire that killed a man nobody meant to kill.

Act and mental state

Most crimes require both a wrongful act and a wrongful state of mind, traditionally captured by two Latin terms:

• Actus reus - the "guilty act," meaning the prohibited conduct itself (or, sometimes, a wrongful failure to act when there was a legal duty to act). There must be a voluntary act; punishing someone for mere thoughts, or for a purely involuntary movement, is generally not permitted.
• Mens rea - the "guilty mind," meaning the mental state that accompanied the act. The law grades culpability by the defendant's mental state, so the same physical act can be a serious crime, a lesser crime, or no crime at all depending on what the person intended or knew.

As a rule, both must be present at the same time for liability. A small set of minor offenses are strict liability crimes that require no mens rea (some traffic or regulatory offenses), but serious crimes require a culpable mental state.

Each half of the pair has fine print. The voluntary act requirement excludes convulsions, reflexes, and movements during unconsciousness; a driver who strikes someone during an unforeseeable seizure has not acted at all in the criminal sense. Omissions are punishable only when law imposed a duty to act - by statute, by a status relationship such as parent and child, by contract such as a lifeguard's, or because the defendant created the peril. A stranger who watches a swimmer drown commits no crime in most states, however harshly we may judge the choice; the criminal law generally commands restraint, not rescue.

Levels of intent

Legal systems commonly rank mental states from most to least culpable. A widely used ordering runs roughly:

1. Purposely / intentionally - the person acted with the conscious aim of causing the result.
2. Knowingly - the person was practically certain the result would occur, even if it was not their goal.
3. Recklessly - the person consciously disregarded a substantial and unjustifiable risk.
4. Negligently - the person failed to perceive a substantial risk that a reasonable person would have perceived.

The higher the mental state, the more serious the offense tends to be. This is why, for instance, killing someone on purpose is treated far more severely than causing a death through carelessness.

This four-step ladder comes from the Model Penal Code, a model statute published by the American Law Institute in 1962 to replace a tangle of common-law terms - malice, wantonness, general and specific intent - with definitions juries can actually apply. Many states adopted its structure in whole or in part; others, and federal law, keep older vocabulary, so the labels in any real case depend on the jurisdiction's code. In plain English the four levels are: I wanted it to happen; I knew it would happen; I saw the risk and ran it anyway; I never noticed a risk any careful person would have seen.

Statutes attach these mental states to specific elements. "Knowingly possessing stolen property" requires knowing the goods were stolen, not merely knowing you possessed them. That is why an honest mistake of fact can defeat liability: taking another's umbrella genuinely believing it yours negates the intent theft requires. A mistake about what the law forbids, by contrast, is generally no excuse.

Homicide: grading by mental state

Homicide law is the clearest demonstration of the ladder at work, because the harm - a human death - is constant while the offense varies with the mind. In broad American terms, murder is unlawful killing with the gravest mental states: purpose to kill, knowledge that death is practically certain, intent to inflict grievous harm, or such extreme indifference to human life that courts call it a depraved heart. Many states divide murder into degrees, reserving the first for premeditated killings. Voluntary manslaughter is an intentional killing mitigated by adequate provocation in the heat of passion, before a reasonable cooling time. Involuntary manslaughter is killing through recklessness or gross negligence.

One more doctrine deserves mention because it departs from the ladder entirely. Under the felony murder rule, retained in many states, a death caused during certain dangerous felonies - arson, robbery, burglary, kidnapping - can be charged as murder even without any intent to kill. The rule is heavily criticized and heavily limited, and its scope varies sharply from state to state; a few have abandoned it. It is worth knowing both as law and as a reminder that grading reflects contested policy choices, not arithmetic.

The burden of proof and the defendant's rights

Every criminal case begins with the presumption of innocence, and the prosecution must prove every element of the offense beyond a reasonable doubt - a requirement the Supreme Court recognized as a constitutional command of due process in In re Winship (1970). The standard embodies a deliberate asymmetry famously voiced by the English jurist William Blackstone: better that ten guilty persons escape than that one innocent person suffer. Two consequences follow. An acquittal is a finding that guilt was not proven, not a certificate of innocence; and the same conduct can still support a civil suit under the lower preponderance standard, as the torts lesson noted.

The right to counsel makes the other safeguards usable. In Gideon v. Wainwright (1963), Clarence Earl Gideon was charged in Florida with breaking into a pool hall, could not afford a lawyer, was refused an appointed one, defended himself, and was convicted. His handwritten petition reached the Supreme Court, which held unanimously that the Sixth Amendment right to counsel is fundamental and applies to the states through the Fourteenth Amendment, so states must provide counsel to felony defendants who cannot pay. Retried with a lawyer, Gideon was acquitted - the case's own demonstration of why counsel matters.

Three years later, Miranda v. Arizona (1966) addressed the police station. Ernesto Miranda confessed after two hours of custodial questioning without ever being told of his rights. The Court held that before custodial interrogation, police must warn a suspect: you have the right to remain silent; anything you say can be used against you; you have the right to an attorney; and one will be appointed if you cannot afford it. Statements obtained without the warnings and a valid waiver cannot be used in the prosecution's main case. The rule protects the Fifth Amendment privilege against self-incrimination inside an inherently pressuring setting, and later cases have refined its edges.

Defenses and the aims of punishment

A defendant may raise defenses that negate an element or provide a justification or excuse. Common examples include self-defense (using reasonable force to protect oneself from an imminent threat), duress (being coerced by a serious threat), mistake in some circumstances, and insanity (a recognized mental condition that prevents the required understanding), the precise contours of which vary by system.

Self-defense repays a closer look. Its core is a reasonable belief that force is immediately necessary against unlawful force, plus proportionality: deadly force is permitted only against threats of death or serious bodily harm. Beyond that core, states genuinely split. Some impose a duty to retreat before using deadly force when retreat is safely possible, at least outside the home; many others have enacted stand-your-ground rules removing that duty. Nearly everywhere, a person who starts the fight loses the defense unless they clearly withdraw. Which rule applies is a question of state law, full stop.

Insanity is an excuse, not a justification: it concedes the act was wrong but denies the actor could be blamed. The oldest test comes from the M'Naghten case (England, 1843), which asks whether a disease of the mind left the defendant unable to know the nature of the act or that it was wrong. Some American jurisdictions use this test, others the Model Penal Code's broader substantial-capacity formula, and a few states have abolished the standalone defense. Contrary to its screen reputation, the defense is rarely raised and rarely succeeds, and a successful verdict typically leads to commitment, not the courthouse door.

Underlying the whole enterprise are the traditional purposes of punishment: retribution (deserved punishment for wrongdoing), deterrence (discouraging future crime by the offender and others), incapacitation (preventing further harm, for example by imprisonment), and rehabilitation (helping the offender become law-abiding). Different societies and eras emphasize these aims differently, which shapes how they define crimes and set penalties.

Common misconceptions

First misconception: police must read Miranda warnings at every arrest, or the case is dismissed. The warnings are required only for custodial interrogation; officers may arrest silently, and volunteered statements remain admissible. A violation suppresses the improper statement; it does not erase the prosecution.

Second misconception: an acquittal proves innocence. It proves only that the state failed to establish guilt beyond a reasonable doubt - a deliberately demanding target. That is also why acquitted defendants can still face civil judgments for the same conduct.

Third misconception: murder requires wanting the victim dead. Extreme-indifference murder and the felony murder rule both convict without purpose to kill. Conversely, motive - the reason behind the crime - is not an element at all; the state proves mental state, not why.

Try it: the warehouse fire

Facts: Devon, furious at a former employer, sets fire to the company warehouse at 2 a.m. to destroy its inventory, having watched the building and concluded it was empty at night. A security guard recently hired to sleep on site dies of smoke inhalation. Devon is charged with murder and insists he never meant to hurt anyone. Work the IRAC before reading on.

• Issue: What grade of criminal homicide, if any, fits a death Devon caused but did not intend?
• Rule: Murder requires purpose, knowledge, or extreme indifference to human life; involuntary manslaughter requires recklessness or gross negligence; and in many states the felony murder rule makes a killing during a dangerous felony such as arson murder regardless of intent, though its scope varies by state.
• Application: Purpose is absent, and knowledge fails too: believing the building empty, Devon was not practically certain anyone would die. The live questions are lower on the ladder. Torching a commercial building at night consciously disregards a substantial risk that someone - a guard, a cleaner, a firefighter - is or will be inside, which supports at least recklessness; a jury might find the disregard so callous as to show extreme indifference. And if the state retains felony murder with arson as a listed felony, the death during the arson supplies murder liability without any inquiry into intent.
• Conclusion: In a felony-murder state, murder; elsewhere, likely involuntary manslaughter or depraved-heart murder, depending on how the jury reads his disregard of the risk.

Two footnotes complete the analysis. Whatever grade the prosecution chooses, it must prove the mental state beyond a reasonable doubt, and Devon, if indigent, will have appointed counsel to hold it to that burden - Winship and Gideon operating in an ordinary case.

In sum, criminal law pairs a voluntary act with a graded mental state, surrounds prosecution with the reasonable-doubt standard, counsel, and warnings, and recognizes defenses that justify or excuse. One question remains for the course: what the law demands of the professionals who run this machinery. The final lesson turns to the duties of lawyers.

Sources

1. Singer, R. (2000). The Model Penal Code and three two (possibly only one) ways courts avoid mens rea. Buffalo Criminal Law Review, 4(1), 139-216. doi.org ↗
2. In re Winship, 397 U.S. 358 (1970). Legal Information Institute, Cornell Law School. law.cornell.edu
3. Gideon v. Wainwright, 372 U.S. 335 (1963). Legal Information Institute, Cornell Law School. law.cornell.edu
4. Miranda v. Arizona, 384 U.S. 436 (1966). Legal Information Institute, Cornell Law School. law.cornell.edu
5. Legal Information Institute. (n.d.). Mens rea. Cornell Law School. law.cornell.edu
6. Legal Information Institute. (n.d.). Actus reus. Cornell Law School. law.cornell.edu
7. Legal Information Institute. (n.d.). Insanity defense. Cornell Law School. (The lesson's discussion of the M'Naghten test, an 1843 English decision, is cited by name and year only.) law.cornell.edu

Legal ethics is the body of rules and principles governing the professional conduct of lawyers. Because lawyers hold special powers - access to the courts, confidential information, and influence over people's rights - the law and the profession impose duties designed to keep that power trustworthy. These duties fall into two broad groups: duties to the client and duties to the court and the public. Much of the challenge of legal ethics lies in balancing the two.

The rules are law, not etiquette. In the United States, each state's highest court licenses lawyers and disciplines them, and most states pattern their rules on the American Bar Association's Model Rules of Professional Conduct, first adopted in 1983 and enacted locally with real variations. Violations are investigated by disciplinary authorities and can end in reprimand, suspension, or disbarment, separate from any malpractice liability to the client. This lesson maps the client-facing duties, the system-facing duties, the recurring puzzles of who the client is and how fees work, and the collisions that make the subject genuinely hard.

Why the rules exist: a lesson from Watergate

The profession has policed itself for centuries, but the modern shape of American legal ethics education traces to a scandal. The Watergate affair of the early 1970s, which ended in President Nixon's resignation, implicated a remarkable number of lawyers - White House counsel, campaign attorneys, even attorneys general - men trained in law who helped conceal a burglary and obstruct its investigation. The bar's response was institutional. In 1974 the American Bar Association began requiring accredited law schools to teach professional responsibility, and a dedicated national ethics examination, the Multistate Professional Responsibility Examination, followed and is now required for admission in nearly every state.

The lesson drawn was specific: technical brilliance does not supply professional limits. Lawyers close to power had treated their skills as tools for whatever the client wanted done. The rules studied in this lesson institutionalize the opposite premise - that a lawyer's judgment about what may not be done is part of the job itself.

Duties to the client

• Competence. A lawyer must have and apply the knowledge, skill, and preparation reasonably needed to handle a matter, and must keep their skills current.
• Diligence. A lawyer must act with reasonable promptness and dedication in pursuing the client's lawful objectives.
• Confidentiality. A lawyer must protect information relating to the representation. This duty encourages clients to be candid, which is essential to good advice. A closely related evidentiary rule, attorney-client privilege, shields confidential lawyer-client communications from being compelled as evidence, with narrow exceptions.
• Loyalty and avoiding conflicts of interest. A lawyer must act in the client's interest and must not represent competing interests without meeting strict conditions. A conflict of interest arises when the lawyer's duties to one client, to a former client, or the lawyer's own interests would materially limit the representation.
• Communication. A lawyer must keep the client reasonably informed and explain matters enough for the client to make informed decisions.

Confidentiality and privilege are constantly confused, so draw the line carefully. The duty of confidentiality is a professional rule: it covers all information relating to the representation, whatever its source, and it applies everywhere - at a dinner party as much as in court - continuing after the matter ends and, in most jurisdictions, after the client's death. The privilege is a rule of evidence: it covers only confidential communications between lawyer and client made to obtain or give legal advice, and it operates only against compelled disclosure in official proceedings.

A comparison makes it concrete. A lawyer who learns from a newspaper that her client's factory failed an inspection holds that fact in confidence, but it is not privileged, because it was not a lawyer-client communication. The client's late-night email asking for advice about the inspection is both confidential and privileged. Everything privileged is confidential; only a narrow slice of what is confidential is privileged. Both protections have exceptions: the privilege never covers communications made to further a crime or fraud, and confidentiality yields in defined situations - such as preventing reasonably certain death or serious bodily harm - with the scope varying by state.

Conflicts of interest come in three families. Concurrent conflicts involve two current clients: a lawyer generally may not act against a present client even in an unrelated matter. Former-client conflicts forbid switching sides in the same or a substantially related matter, since the old client's confidences would become weapons. Personal conflicts involve the lawyer's own stake - business deals with clients, a fee tied to a particular outcome of advice, relationships across the aisle. One lawyer's conflict is ordinarily imputed to the whole firm, though screening cures some cases. Many conflicts can be waived by informed consent confirmed in writing; some are so corrosive that no consent suffices.

Who is the client?

The question sounds trivial and decides cases. A lawyer who represents an organization represents the entity itself, not its officers or shareholders - a distinction that turns sharp when an executive's interests split from the company's, as in an internal investigation. Corporate counsel often begin employee interviews by explaining that they act for the company and that the company, not the employee, controls the privilege. Joint clients present the mirror problem: co-founders or spouses represented together share one representation, so as between them the lawyer generally keeps no material secrets, and if their interests split, the lawyer may have to withdraw from representing either.

Payment does not settle the question. A parent who pays for an adult child's defense, or an insurer that pays defense counsel for its insured, is not thereby the client; the lawyer's loyalty, confidentiality, and obedience on the objectives of representation run to the represented person. The rules require that the paying stranger not direct the lawyer's judgment. Keeping the flow of money separate from the flow of loyalty is one of the profession's quiet disciplines.

Fees and client money

Fees must be reasonable, judged by factors such as the time and difficulty involved, customary local rates, the results obtained, and the lawyer's experience. The common structures are plain: hourly billing charges for time; flat fees price a defined task; and the contingent fee pays the lawyer a percentage of any recovery, and nothing otherwise. Contingent fees are the door to court for people who cannot pay by the hour, which is why they dominate personal injury practice; they generally must be in writing, and they are prohibited in criminal cases and, in most states, in most family-law matters.

Client money is sacred ground. Funds held for a client - settlements received, advance deposits against fees - must sit in a separate trust account, never mixed with the lawyer's own. Commingling accounts, or borrowing from client funds even briefly and with intent to repay, is among the surest routes to disbarment in every state. Many states also offer fee arbitration programs so that disputes over bills need not become lawsuits between lawyer and client.

Duties to the court and the public

A lawyer is not merely a hired advocate; as an officer of the court, a lawyer also owes duties to the legal system itself.

• Candor toward the court. A lawyer must not knowingly make false statements to a court or offer evidence the lawyer knows to be false, and must not mislead the court about the law.
• Fairness to others. A lawyer must not unlawfully obstruct evidence, harass, or use clearly improper tactics against opposing parties and witnesses.
• Not assisting crime or fraud. A lawyer may advise a client about the law but must not counsel or help a client to commit a crime or fraud.

Candor reaches further than not lying. A lawyer who learns that something material she told the court was false must correct it. She must disclose controlling legal authority directly adverse to her position if the other side misses it - a striking rule: you may have to hand the judge the precedent that hurts you. And if a client or witness has testified falsely, the lawyer must take remedial measures, which under the Model Rules can escalate, if nothing else works, to disclosure to the tribunal; jurisdictions differ on the details. Frivolous claims are forbidden throughout: every position needs a basis in law or a good-faith argument for changing it.

Within those limits, advocacy is meant to be wholehearted. The older codes called it zealous advocacy, and the idea survives in the modern duty of diligence: the lawyer resolves doubts in the client's favor, presses every legitimate argument, and leaves judgment of guilt or fault to the tribunal, because the adversary system assigns that role to courts, not counsel. This is why defending a plainly guilty client is honorable work within the bounds: the defense lawyer's job, as the previous lesson showed, is to hold the state to its proof. The bounds - no perjury, no fabricated evidence, no harassment - are what make the zeal tolerable.

When duties collide

These obligations can pull in different directions, which is what makes legal ethics genuinely hard. Consider the duty of confidentiality versus the duty of candor to the court. Suppose a client tells a lawyer, in confidence, information suggesting a document the client wants to submit is forged. The lawyer's loyalty and confidentiality point one way; the duty not to offer evidence known to be false, and not to assist a fraud, point the other.

Ethics rules resolve such conflicts through carefully drawn lines - here, the lawyer generally may not present evidence they know to be false, and the duty of candor to the court and the prohibition on assisting fraud limit what confidentiality can shield. The lesson is that a lawyer's role is bounded: zealous advocacy operates within limits set by honesty toward the court and the law. That boundary is what allows an adversarial system, in which each side argues hard for its client, to serve justice rather than undermine it.

Common misconceptions

First misconception: a lawyer must keep every secret forever, no matter the cost. Confidentiality is broad but not absolute: exceptions permit, and in some states require, disclosure to prevent serious harms, and the privilege never protects communications used to plan a crime or fraud. The exact lines vary by state.

Second misconception: defending a guilty client is unethical. The opposite is closer to true. The system tests guilt through adversarial proof, and it works only if someone makes the state carry its burden; a defense lawyer who refuses to judge the client is performing the role, not betraying it.

Third misconception: having a lawyer present makes a conversation privileged. Privilege requires a confidential communication seeking legal advice. Business advice is not covered, inviting an outsider into the meeting usually destroys confidentiality, and copying a lawyer on an email does not shield the email. Privilege is earned by the communication's purpose, not by a lawyer's presence.

Try it: the two-founder startup

Facts: Amara represents Ben and Chloe jointly in forming their startup and drafting the founders' agreement, which forbids either founder from selling shares to a competitor. A year later Ben privately tells Amara that he is secretly negotiating exactly such a sale and asks her to keep it from Chloe while she updates the company's contracts. Work through Amara's duties before reading on.

Worked analysis. Start with the client question: Amara represents Ben and Chloe jointly, so within the matter she generally may not keep material information from one for the other's benefit; Ben's request asks her to do precisely that. Confidentiality does not resolve it, because the duty runs to both joint clients, and Ben's plan is material to Chloe's interests in the founders' agreement Amara herself drafted. Next, loyalty: the founders' interests now conflict directly, and no single lawyer can advance both, so the representation is materially limited.

Now the boundaries. Continuing to paper the company's contracts while concealing the planned breach edges toward assisting a fraud, which no duty to Ben can license. Amara should tell Ben she cannot keep this from Chloe within a joint representation, urge disclosure, and if he refuses, withdraw from the joint matter rather than serve one client by betraying the other; whether and what she may then reveal depends on her state's confidentiality exceptions. The scenario compresses the lesson: identify the client, respect the structure of the representation, and refuse the role of instrument in a wrong.

In sum, a lawyer owes clients competence, diligence, confidentiality, loyalty, and candor about their own matters, owes courts honesty and fair dealing, and owes the public a refusal to lend skill to fraud. These duties are the course's fitting close: the rule of law from the first lesson survives only if the people who operate the system can be trusted with it.

Sources

1. American Bar Association. (n.d.). Model Rules of Professional Conduct. (Cited without a link: americanbar.org ↗ refuses automated requests, so no verified direct URL could be embedded. See especially Rule 1.6 on confidentiality, Rule 1.7 on concurrent conflicts, Rule 1.13 on the organization as client, Rule 1.15 on safekeeping property, and Rule 3.3 on candor toward the tribunal.)
2. Moliterno, J. E. (2013). The deepest embarrassment: Watergate. In The American Legal Profession in Crisis (pp. 96-107). doi.org/10.1093/acprof:oso/9780199917631.003.0005
3. Upjohn Co. v. United States, 449 U.S. 383 (1981). (Cited for the corporate attorney-client privilege and the entity-as-client problem.) Legal Information Institute, Cornell Law School. law.cornell.edu
4. Legal Information Institute. (n.d.). Legal ethics. Cornell Law School. law.cornell.edu
5. Legal Information Institute. (n.d.). Attorney-client privilege. Cornell Law School. law.cornell.edu
6. Legal Information Institute. (n.d.). Conflict of interest. Cornell Law School. law.cornell.edu
7. Legal Information Institute. (n.d.). Fiduciary duty. Cornell Law School. law.cornell.edu