The fundamental problem with this reasoning in the case of a precedent is that it suffers from a kind of circularity. It is true that legal systems that follow previous practice give rise to hope that previous decisions will be followed in the future. However, it is important to remember that only legitimate expectations should be taken into account in decision-making, and not just any expectation that someone forms. The mere fact that a decision has been taken in the past does not in itself give any reason to believe that it will be followed in the future and certainly does not give rise to the claim that it will be followed. The fact that a company orders stationery from a supplier at the beginning of the year may raise hope and perhaps even an expectation that it will do so again later in the year, but the supplier has no «legitimate» expectation that it will do so, and the company does not set a precedent for itself. If, on the other hand, there is an institutional practice of following past decisions, the confidence of those who are the subject of future decisions may raise legitimate expectations, but the institution is free at any time to announce that it will no longer treat past decisions as binding and will instead decide on each case on the merits. Similarly, in some institutional contexts, it is common for previous decisions to be followed so that a decision is made on the condition that it «does not set a precedent» for the future. It should be noted that modern common law advocates a particularly strong version of stare decisis, which requires subsequent courts to follow previous decisions, even if these cases have been misresolved under existing law. Ordinary lawyers often assume that a doctrine of the decision of the gaze necessarily requires that subsequent courts be bound by such erroneous decisions. This follows from the next stream of thought.

If subsequent courts were not obliged to follow bad decisions, then they would only be «bound» by previous and correct judgments. But an earlier correct decision simply concludes that the law was already supported when it was enacted. Thus, ordering the courts to follow cases that were not flawed would simply mean asking them to do what they are legally required to do (i.e., to apply the law), which would make the doctrine of precedent superfluous. The flaw in this argument lies in the assumption that in each case there must be only one legally correct result, while other results are false. This neglects the possibility of cases where the merits of the dispute are legally uncertain, so that there is more than one possible outcome that would not be bad. To say that a case is «legally indefinite» implies a number of situations, such as the merits of opposing arguments, which are legally the same or in which conflicting considerations cannot be rationally classified in relation to each other. In such cases, the decision amends the law without making any mistakes. The common law could therefore have restricted its doctrine of stare decisis by holding that subsequent courts were not bound by earlier decisions that had been incorrectly decided. [2] Instead, it developed a different practice – that of «annulment», in which some courts were given limited power to withdraw their binding status from previous decisions because they had been incorrectly decided. Thus, the common law version of the doctrine of precedents does not necessarily result from the fact that precedents have practical authority.

Nevertheless, the idea of having to follow even bad decisions is a common feature of the decision-making of many institutions and will be at the heart of this entry. Similar reasoning helps to make the outcome of cases more predictable by giving weight to existing legal decisions and doctrines. However, this only happens in a certain context, where, despite the fact that policymakers do not share a unified normative vision, there is a high degree of agreement on the existence and meaning of certain values. A certain degree of agreement is necessary for decision-makers to consider a case as analogous, as it is based on what they consider to be the correct justification for the previous decision. It also means that, although they do not agree in different cases on the solution found in different cases, it is unlikely that the disagreement will be deep, but it will be reasonable. (For an argument that this requirement may be overstated, see Sunstein 1993, 769-73). However, the strongest justification for analog thinking lies in the value of reproducibility. This is often expressed in terms of the importance of «consistency» in law (MacCormick 1978, 153, 187-8; Sunstein, 1993, pp. 778-9; see also Raz 1979, 204-6 on «Partial Reform»). Arguments in favour of coherence generally emphasize its instrumental value. [22] This is related to the reproducibility of legal decisions.

There are two important features of legal decision-making. The first is the fragmentary nature of legal documents. The second is the plurality of decision-making bodies. Legal documents – precedents, laws, conventions, principles – are fragmentary in two ways: (a) they are the work of many different hands at different times and with different points of view, and (b) different areas of law owe at some times and at times more than others. As a result, case law tends to show little overall consistency, whereas it may possess thick local coherence. The pluralism of decision-makers is also twofold: (a) there are many people who make decisions with the same body of material, and (b) these people do not share a unified evaluation perspective. Given the fragmentation of legal material and the plurality of decision-makers, there is considerable room for disagreement when decision-makers are faced with new issues. It is often said that this creates a clear contrast with laws, in which a canonical formulation of the established legal norm is provided.

Given the flexibility available to subsequent courts to determine the relationship between the previous decision, it is misleading to believe that decisions establish binding rules for subsequent courts.