Link Search Menu Expand Document

3.4.4 Legal Powers

By Laurence Diver

On this page

  1. Working definition
  2. Examples of how ‘legal powers’ is used
  3. The meaning of ‘legal powers’ in terms of MoE, affordance, LPbD
    1. Mode of Existence
    2. Affordance
    3. Legal Protection by Design

Working definition

  1. A legal power refers to the ability of a legal subject to achieve an intended legal effect
  2. A legal power is attributed by positive law
  3. One can have a legal power to attribute legal powers
  4. All legal powers are both constituted and constrained by positive law
  5. The written or unwritten Constitution of a state attributes legal powers to legislate, administrate and adjudicate, thus calling them into existence (the constitution ‘constitutes’ these powers), and qualifying them (the constitution also ‘regulates’ these powers, e.g. by attributing them to countervailing powers)
  6. In the case of public law legal powers are constrained by the legality principle.
  7. In the case of private law legal powers are constrained by the reasonableness principle (or equity in common law jurisdictions).
  8. The attribution of legal power plays out in all domains of law:
    1. Private law, for instance, attributes to the owner of a legal good the legal power to transfer related property rights, provided specific conditions have been fulfilled
    2. Criminal law, for instance, attributes to the court the legal power to impose specified (maximum) punishments, provided the conditions of a specific criminal offence have been fulfilled
    3. Administrative law, for instance, attributes to legal subjects the legal power to object to decisions made by public administration, provided specific conditions apply
    4. International law, for instance, attributes to states the legal power to conclude treaties, subject to the constraints imposed by the sources of international law

A legal power gives a legal subject the ability to perform juridical acts that have the legal effect of changing the legal status of itself and/or other legal subjects.

Legal powers highlight the fact that law is not just about regulating what legal subjects can do, as is often assumed, but also about empowering them to create new states of enforceable legal effect.

Public legal powers enable authorities to, for example, legislate (parliament) or adjudicate (courts and tribunals). Parliament has the legal power, for example, to impose a universal speed limit on public roads within its jurisdiction.

Private legal powers enable legal subjects to, for example, conclude contracts of sale, employment, or marriage. They also enable the creation of new legal subjects such as companies, charities, or indeed children (that is, the legal subjectivity of the child that is born).

Legal powers often accrue as the result of holding subjective rights. For example, one has the legal power to sell one’s car (power of disposition) because one owns it (the subjective right of ownership entails the power of disposition).

Legal powers are generally dependent on the legal subject having the capacity necessary to exercise them. For example, a minor or someone with serious cognitive impairments will be unable to conclude a marriage contract, due to lack of majority age or the ability to give fully-informed consent, respectively (in some jurisdictions both conditions will apply to the minor).

Mode of Existence

Where the legal subject is the fundamental actor within the legal Welt, legal powers are the fundamental means of acting within its institutional domain, in ways that are compatible both with its existing granular processes and with its implied philosophy. This again is based on institutional facts built around speech acts that follow conventional procedures specified by positive law. Legal powers are thus distinct from other forms of power (e.g. political, physical, technological) in that they create new institutional facts that are recognised by the legal normative order.

The kinds of legal norm that define the nature of a legal power, and how validly to exercise it, are sometimes referred to as ‘secondary rules’. This is in contrast to the ‘primary rules’ that specify what it is we should or should not do under some legal duty.1 To demonstrate the difference, consider that the legislature exercises its legal power, specified under a constitutional secondary rule, to create a speed limit that will apply on public roads. Similarly, the legal subject exercises her power, specified by a secondary rule found in the law of contract, to purchase a car from the showroom. As we saw earlier, this contract contains primary rules that require both her to pay the purchase price and the showroom to supply the car,2 while the speed limit is a primary rule that requires her to drive her new car at less than a certain speed (subject to exemptions).

Legal subjects are free to exercise a legal power provided they have the necessary capacity, and the other conditions specified in the secondary rule(s) defining that power are met.

Those conditions will have a common general structure:

  1. What [legal subject(s)], having
  2. either what active capacity or what competence based on some position or appointment [e.g. a public office],
  3. in what required circumstances, and
  4. in the absence of what vitiating circumstances or factors,
  5. by what if any special procedures or formalities, and
  6. by what act
  7. in respect of what if any other [legal subject(s)]
  8. having what general capacity
  9. can validly bring about a certain legal change?3

(although it should be noted that these elements are unlikely to be found in one place in the positive law.)

Each of these elements plays a part in the valid performative of the speech act that seeks to exercise the power. In this way, the myriad and varied acts that legal powers make possible can nevertheless be fit into a recognisable structure, which in aggregate facilitates the stable (but flexible) set of mutual expectations underpinning societal stability. Once again, this allows us to abstract from the complexity of real-world circumstances those factors which have relevance with respect to the institutional definition of the legal power.

Given the institutional nature of legal powers, failure to meet the necessary requirements results in either a legal nullity (something void, with no legal effect), or a legal effect that is voidable, that is to say an effect that is liable, or vulnerable, to ‘cancellation’ by a court.4

We can see here the temporal aspect of the institutional mode of existence of legal norms: those who are party to the purported exercise of the legal power may well continue to act as if it were valid, even if legally speaking it is void. But once it is realised that something is wrong, the ability of the parties to contest the institutional fact that arose from that exercise comes into play. For example, the showroom supplies the customer with a car that was significantly defective in a way that was not immediately apparent. After discovery of that fact, the purchaser can argue that the car was not of ‘merchantable quality’ under consumer protection law, and thus that the contract ought to be ‘cancelled’.

Conferral of a power cannot be by sovereign fiat but must be done on a legal basis, subject to the checks and balances of the rule of law

The above is an example from the private law of contract, but this is also a common experience in administrative law, with respect to the exercise of legal powers by public authorities. Many administrative law cases, at least in the common law world, hinge on disputes as to whether or not a public authority has acted ultra vires (that is, outside the powers given it by positive law). In such cases the purported legal effect of the act of a public authority can also be ‘cancelled’ by the court.

All legal powers must derive from secondary rules that are legitimated under the legality principle. This means the conferral of a power cannot be by sovereign fiat but must be made on a legal basis, subject to the checks and balances of the rule of law. Again, any purported attempt to subvert this is made challengeable by dint of the mode of existence of legal powers: the institutional nature of the secondary rule and the normative order within which it is purported to be exercised.5 As with subjective rights, this has important implications for the legitimacy of computational systems that ‘confer powers’, in terms of who confers them, what ‘confer’ means in that context, and what sorts of ‘power’ can possibly be so conferred.


As with subjective rights, we can think of legal powers in terms of two levels of affordance — first, those provided by the medium and technologies that underpin law-as-we-know-it, and secondly those that determine what is possible within the legal-institutional dimension, or Welt, that is thereby made possible.

At that second level, institutional law affords us the empowerment discussed above — it allows the legal subject to become a “private legislator” in ways that can reflect contingent needs and interests,6 all of which is given the normative backing of legal enforceability, subject to contest, that facilitates “legitimate mutual expectations”.7

Given the institutional nature of legal powers, they can never be exercises of brute force or bare sovereign power, because the mode of existence of text-driven law resists this by its very nature. Legal powers are relationships of affordance between legal subjects and the broader normative order, again configured pragmatically and co-constitutively by what is actually needed to make such a system work in practice. In this way the parts entail the whole, and vice-versa: a system that seeks to enable the exercise of powers, enforceable when necessary but always subject to checks and balances, cannot be a system that facilitates brute force. It follows that the converse is also true.

Legal powers are not isolated permissions, but necessarily come with normative ‘baggage’ that is what qualifies them as ‘legal’.

Returning to the notion of permissions discussed above, we can appreciate then that the valid exercise of a legal power, much like the assertion of a right, is not merely a case of answering the ‘single-dimensional’ question of ‘do I have permission X so that I can perform action Y’. Like the rest of the law, the exercise of the power takes place within a much more complex and multi-dimensional framework that includes not just the set of requirements listed above (whether implicitly or explicitly specified), but also the broader context of legal process that facilitates contest of that exercise after-the-fact.

In this sense, then, legal powers are not isolated capabilities, where the legal system affords a singular legal act, but rather they have attached to them a whole raft of normative ‘baggage’ that is what qualifies that singular act as legal. What makes a power ‘legal’, then, as opposed to something else, is that which implies the system of legal effect within which it has purchase. This is the institutional Welt made possible by the affordances of text.

The above implies that powers cannot be conferred arbitrarily, but must have a legitimate source in the form of a secondary legal norm. Where it is not possible to identify such a source, it is doubtful whether the resulting power can be said to be legal, and so we run the risk that that power is arbitrarily defined, lying outside the normative framework of legal protection that can curb any excesses it might otherwise enable.

Legal Protection by Design is concerned with constraining design so that it retains a connection to the legal-institutional Welt.

Even where the conferred power is innocuous, that is to say its effects are not particularly harmful,8 if the medium that makes it possible is used to facilitate the putative exercise of traditionally legal powers, the effects might be significant in terms of their effacement of the architecture of protection provided by the current mode of existence of law.

Where the conferral of powers is done by the (private) designer of a system, their role in determining the nature of the power might not accord with the constraints placed on the legislature by the legality principle in its creation of secondary legal norms.9 This ‘sovereign designer’ creates a playing field that is separate from the legal ‘game’,10 with commitments that might bear no relation to legal protection whatsoever, and powers (as affordances within the geography of the system) granted to the ‘user’ that are not institutional by nature, with all the attendant normative structure this mode of existence entails. Legal Protection by Design is concerned with constraining any such system so that it retains this connection to the institutional Welt, ensuring that a technological instantiation of a power reflects the institutional affordances of its textual counterpart.


  1. H.L.A. Hart, The Concept of Law (Clarendon Press 1994), pp. 27–42. 

  2. As we saw earlier, these duties combined with complementary rights to receive ownership of the car and to receive the purchase price, respectively. 

  3. N. MacCormick, Institutions of Law: An Essay in Legal Theory (Oxford University Press 2007), pp. 156-57. 

  4. The precise legal term for this kind of ‘cancellation’ varies between jurisdictions and legal domains, as do the exact consequences in terms of legal effect, but for present purposes the notion of cancellation is sufficient. 

  5. A dramatic example is the attempt in 2019 by the UK Government to ‘prorogue’ (temporarily suspend) Parliament, which was found by the Supreme Court to lie far beyond the limits of its legal power to do so. See Cherry and others (Respondents) v Advocate General for Scotland (Appellant) (Scotland) [2019] UKSC 41. 

  6. Hart (n 1), p. 41. 

  7. Hildebrandt, Law for Computer Scientists and Other Folk (n 11), p. 45. 

  8. An obvious question that arises is ‘who would assess this?’, for the answer to which one need only point to legality and the rule of law as the animating forces that lie at the centre of law-as-we-know-it. 

  9. This is the central problem considered in L.E. Diver, Digisprudence: Code as Law Rebooted (Edinburgh University Press, forthcoming). 

  10. H.Y. Kang and S. Kendall, ‘Legal Materiality’ in M. Del Mar, B. Mayler and S. Stern (eds), Oxford Handbook of Law and Humanities (Oxford University Press 2019); C. Vismann and M. Krajewski, ‘Computer Juridisms’ (2007) 29 Grey Room 90. 

This page was last updated on 16 March 2023.