Debate: Al Martinich vs. John Deigh on law. Part 1: Martinich.

This is the first in a series of debates about entries in The Oxford Handbook of Hobbes, ed. A.P. Martinich and Kinch Hoekstra (OUP 2016 – see here).

This first debate features a critique by one of the editors, Al Martinich, of John Deigh’s chapter ‘Political Obligation’. Deigh responds in Part 2.


The Laws of Nature are the Laws of God in Leviathan

A.P. Martinich

University of Texas at Austin

The most difficult issue to resolve with respect to the correct interpretation of Hobbes’s view about the laws of nature in Leviathan is their relation, if any, to the command of God. I think that Hobbes’s considered view is that the laws of nature are genuine laws and impose obligations because they are commanded by God. Let’s consider what Hobbes says about laws in general. All laws consist of two aspects (Hobbes 1651: 137). The first and more salient component is the content of the law. Hobbes usually expresses the content of the laws of nature in the indicative mood, such as ‘You lay down your right to all things’ and ‘You keep your covenants’. Expressing laws in the indicative mood may seem odd because Hobbes thinks that laws are commands and commands are often expressed as imperatives; and Hobbes himself sometimes formulates laws as imperatives, such as, ‘Make peace’. However, formulating laws in the indicative mood is not odd; many contemporary jurisdiction formulate laws in just this way. Here’s a randomly selected law:

(a) A person commits an offense if he unlawfully appropriates property with intent to deprive the owner of property.
(b) Appropriation of property is unlawful if:

(1) it is without the owner’s effective consent;
(2) the property is stolen and the actor appropriates the property knowing it was stolen by another; or
(3) property in the custody of any law enforcement agency was explicitly represented by any law enforcement agent to the actor as being stolen and the actor appropriates the property believing it was stolen by another (Texas Penal Code: Section 31.03).

Given the propensity to think of laws as commands, I think it is amazing that Hobbes so rarely slips into the imperative mood.

One place where he does use the imperative is in his explanation of the distinction between commands and counsels. He gives, ‘Do this’, as an example of a sentence that may be a command or counsel depending upon who said it and with what intention. If a person with authority said it to a subordinate and expects the subordinate not to have any reason for obeying other than that the authority wants it done, then ‘Do this’ is a command. If the person does not have authority over the addressee and intends to direct the addressee to perform an action beneficial to the latter, then a primary desires is counsel.[1] My guess is that Hobbes chose the underdetermined sentence, ‘Do this’, because he did not want his readers to form a judgment about whether the sample utterance was a command or counsel on the basis of its content. If Hobbes had chosen, ‘Do not take the belongings of other people without their permission’, his readers may have thought, ‘This is a command’, and if he had used, ‘Get an education’, his readers may have thought ‘This is counsel’. Yet, for Hobbes, each of these latter examples could be command or counsel. A more unfortunate aspect of his selection of ‘Do this’—and the same would hold if he had chosen either of the other two examples I suggested—is that he does not clearly enough communicate the idea that these utterances can be made explicit commands or counsels by prefixing them with a performative phrase, ‘I command’ or ‘I counsel’.

The upshot of this discussion is that a command, like other illocutionary acts, has two elements, a propositional content and a force, and that often the force is fully expressed only when an illocutionary act verb in performative form prefixes the content of the law. However, sometimes the words expressing the force should not be expressed. This is especially true when the laws of nature are to be proved using only a logic that applies to sentences in the indicative mood, as was the case in the seventeenth century. There was no logic of imperatives or performative utterances at that time. So, while ‘You lay down your right to all things’ and ‘You keep your covenants’ can be proved straightforwardly, ‘Make peace’ cannot. In order to make it provable, it has to be paraphrased as, ‘You make peace’. Here’s a proof briefly stated: (1) You act to satisfy your desire to preserve yourself. (2) If you do not make peace, then you do not act to satisfy your desire to preserve yourself. Therefore, (3) You make peace.

Well, precisely what expression is supposed to express the force of the laws of nature? I think there are only two plausible possibilities: Either reason commands them or God commands them. In favor of the former is the fact that Hobbes sometimes says that reason commands. However, Hobbes should not think that reason has any authority over human beings; and it is easy to use synecdoche without realizing that the statement is not strictly true. Consider the slogan ‘Guns kill people’. Since both John Deigh and I believe that Hobbes’s explicit view of reason is only instrumental and suggests means to given ends, I will not discuss the possibility that reason commands the laws of nature further. That leaves the disjunct that God commands the laws of nature.

There are positive reasons for holding that God commands the laws of nature. Hobbes says they are. They are the ‘the undoubted word of God’ because God speaks to human beings through reason. Hobbes knows that human psychology makes people prone to act on dangerous desires. So fear of death and hope for a comfortable life may not be sufficient to preserve them. Fear of the punishments of God, an invisible power, helps people obey them. The legal (forensic) character of the laws of nature make people more inclined to obey. Many scholars think that Hobbes’s definition of religion as fear of invisible powers is part of Hobbes’s surreptitious campaign against theism. That is highly implausible. Christians understood fear of God to be of the essence of their religion. The beginning of wisdom is fear of the Lord. After Adam and Eve sin, they are afraid of God. When God wants to make a covenant with the Israelites, they are afraid of him, and plead with Moses to intercede for them. Martin Fotherby, bishop of Salisbury, wrote:

Religion itselfe is a kind of fear. Religion is nothing else, but the fear of some God: as the Creator himself hath expressly defined it. … Religion is an affection, which begetteth a care of worshipping a certain superior Nature, which is commonly called God (Fotherby 1622: 122).

We may find religions of fear repulsive, but the seventeenth-century English did not. So it would be absurd for Hobbes to try to get people to understand that there is something wrong with Christianity specifically and religion generally by ‘writing between the lines’.

A vivid example of divine fear is the Engagement Controversy. Thousands of the English balked at taking the Engagement for fear of violating the third law of nature, ‘You keep your covenants’. In the state of nature the fear of God inclines people to obey the laws of nature because God is ‘the revenger of their perfidy’ (70). God, the avenger, is doing his job only if he is taking revenge for violation of his laws. This is only to say that fear of God’s power helps motivate people to obey the laws of nature, not that it is sufficient to get everyone always to follow them.

I believe that John Deigh accepts at least the spirit of what I have said about the laws of nature with respect to how they can be proved. What he does not accept is that for Hobbes the laws of nature are commands of God (Deigh 2016: 304-7). He, like most nonlaw theorists, quotes Hobbes’s comment at the end of chapter 15, ‘These dictates of reason men use to call by name of laws, but improperly; for they are but conclusions or theorems concerning what conduceth to the conservation and defence of themselves’ (Hobbes 1651: 80). Deigh maintains that since Hobbes had been proving laws of nature for two chapters, he must mean that the laws of nature are identical with the dictates of reason. But if that were true why didn’t Hobbes say that? Why doesn’t he use the phrase, ‘the laws of nature’, again if his position is that the laws of nature as such are not laws? That would eliminate any idea that those laws were laws. One explanation is that he did not want to be saddled with the paradoxical sentence, ‘The laws of nature are not laws’. Counting against this explanation is the fact that Hobbes embraced paradox (see Parkin, 2016). A different explanation, one that has some force with me is that, as great a stylist as he was, Hobbes sometimes phrased his points sloppily. But there is a better explanation. He shifts to a different locution, ‘dictates of reason’, precisely because he wants to indicate what part of a law of nature is provable, namely, the propositional or action-guiding part, and what part makes it a command. Propositions, when proved by reason, and commanded by God, are laws. Until the end of chapter 15, he could not discuss these propositions as commanded because he could not prove sentences of the form, ‘I, N.N., command that p.’ As theorems, the laws of nature have to be formulated without a performative prefix. In order to convey that the laws of nature are laws only insofar as they are commanded, as he intends to do in this context, Hobbes adopts the phrase, ‘dictates of reason’. So now, having proved what he wanted to prove, he takes the next step and reports what makes them laws: ‘if we consider the same theorems as delivered in the word of God that by right commandeth all things, then they are properly called laws’ (Hobbes 1651: 80). This statement is not an ‘inconvenient aside’ (Deigh 2016: 309).

One might ask, ‘Why God?’ I have already explained why it could not be reason. And there does not seem to be any alternative. Holding that God commands the laws of reason is, given the background assumptions of seventeenth-century intellectuals, inference to the best explanation. When Anglo-American philosophers describe their methods, they often say that they are argument and analysis. If they think about Kierkegaard and Nietzsche, they may think of insight as another method. But one of the most powerful and pervasive methods is inference to the best explanation. God’s command is a better explanation than anything else Hobbes had available.

Nonlaw interpreters think that when Hobbes says, ‘men use to call [the theorems] by the name of laws, but improperly’, he is including himself in the ‘men’. But if he were including himself, he is using an odd locution. When a person refers to a group indefinitely—that is, without a quantifier—and says something critical of it, the implication is that they do not include themselves in the group unless they go out of their way to include themselves. When I say, ‘Professors do not help students enough with their writing’, I’m not talking about myself. I mean professors for the most part, not universally. So if Hobbes wanted to include himself among the people who use the name of laws improperly, he should have said so: ‘Men, including me, use to …’. It would have been very odd for Hobbes to admit that he uses one of his technical terms improperly. If ‘laws of nature’ was used by him improperly, one may wonder why he did not preface his proofs of those ‘laws’ by saying something like this: ‘men use to call [the theorems I am about to prove] by the name of laws, but improperly. I will use the appropriate term “dictates of reason”, and because they are deducible from the definition of “law of nature”, I will also refer to them as theorems.’

The apparent oddity of the passage, ‘men use to call’ is removed if we take it in the most straightforward way, given that he says (1) that God is a sovereign by nature; (2) that reason is one of the ways by which ‘God declareth his laws’ to human beings; and (3) that the laws of nature are found out by reason (Hobbes 1651: 186, 187, 64). It is no good to object that (1) and (2) are expressed long after, in chapter 31, for that chapter belongs as much to Part Two, ‘Of Commonwealth’, as any of the earlier chapters; and its position at the end of Part Two is appropriate since it aids the transition to Part Three, ‘Of a Christian Commonwealth’. Few Hobbes interpreters accept (1) and (2). But that is not pertinent. The pertinent point is that Hobbes’s formulation of his theory includes the laws of nature as laws.

In philosophy, of the making of many objections, there is no end. Deigh recently devised a new one that I consider ingenious. It consists of three components. The first is Hobbes’s insistence that definitions are ‘the starting points of the reasoning, which is to say, the adding and subtracting of words, that yield the theorems of science’ (Deigh 2016: 309). The second is that the nonlaw interpretation ‘conforms … more closely’ to Hobbes’s method of science (Deigh 2016: 306). The third is that, taking Hobbes’s definition of ‘law of nature’ strictly, one will see that it entails that laws of nature are not laws.

I accept the first component but not the second or third. Concerning the second, one problem is that Hobbes’s practice does not always conform to his official method. He does not invariably deduce conclusions from definitions although terms that have been defined may occur in the premises. It is fortunate that Hobbes would not invariably follow his explicit method because much of his philosophy would then disappear. Also, whether he is conscious of his lapse or not, he does not follow his explicit method with respect to his assertion that God commands the laws of nature. Near the beginning of chapter 31, Hobbes says, ‘There wants onely, for the entire Knowledge of Civill duty, to know what are those Lawes of God’ (Hobbes 1651: 186). Hobbes knows that a political philosopher in the seventeenth century has to deal with the issue of the relation of God’s laws to civil laws. Heck, we need to do that in the United States in the twenty-first century!

Also, I think that Deigh places too much weight on ‘assessing Leviathan as a work of philosophy’ when it comes to “evaluating the competing interpretations” (Deigh 2016: 305n29). For any given doctrine, almost any philosopher may have made a mistake, even a demonstrable one. Hobbes thought no one could be ‘so stupid as both to mistake in geometry and also to persist in it when another detects his error to him’ (Hobbes 1651: 21). Yet he was. I also do not agree with Deigh’s claim that law interpreters are committed to the proposition that ‘the source of the obligation to which those [sovereign-making] contracts give rise is the third law of nature’ and not ‘the contracts on which a commonwealth is founded’ (Deigh 2016: 305). I don’t think law interpreters are so committed. The laws of nature and sovereign-making covenants are two different kinds of things. So nothing stands in the way of holding that obligation comes from different sources. The laws of nature bind simply because God is a natural sovereign with irresistible power who promulgates them through reason. Covenants bind when people mutually lay down some rights by their voluntary actions. To say that there are two sources for some object with the name N is not to say that N is ambiguous. For example, one source of wealth is inheritance and another is industriousness; one source of obligation is the laws of nature and another is covenanting. If ‘obligation’ were ambiguous, then the sentence, ‘The laws of nature are a source of obligation and so is covenanting’ would be semantically defective. But it is not.

Deigh’s remark that Hobbes uses, not a categorical statement, but a hypothetical one to explain when the dictates of reason are laws (‘if we consider the same theorems as delivered in the word of God that by right commandeth them, then are they properly called laws’ (Deigh 2016: 306; Hobbes 1651: 80) is also not probative. Even aside from the fact that the logical form of scientific statements are hypothetical, I want to point out that all sorts of true philosophical propositions can be expressed as hypotheticals, such as ‘if we consider the proposition expressed by, “Study mathematics”, as advice, then it is counsel.’

I now turn to the second part of Deigh’s argument that the laws of nature are not laws because ‘law’ in ‘law of nature’ does not mean law. He says,

when a term consisting of two names joined together is defined, the definition preempts the use of the definition of either name whenever that name occurs as a component of the term. Likewise, it preempts the use of either name with some undefined meaning whenever the name occurs as a component of the term. The preemption is necessary to prevent the term from having two distinct extensions and thus being ambiguous (Deigh 2015: 310).

Deigh’s description of Hobbes’s view of definitions has to be too strong, for it has the consequence that ‘right’ in ‘right of nature’ does not mean right; ‘commonwealth’ in ‘commonwealth by acquisition’ does not mean commonwealth; and ‘laws’ in ‘civil laws’ would not mean laws because each of these technical terms are multi-name scientific ones (cf. Hobbes 1651: 102, 136). What is true is the weaker proposition that component ‘words’ of a multi-name technical term may not have the same meaning that it has outside of that technical term (hereafter: ‘ordinary meaning’). With this proposition, Deigh could go on, as he does, to assert that it is plausible that ‘law’ in ‘law of nature’ does not have its ordinary meaning just as ‘liberty’ in ‘civil liberty’ does not have its ordinary meaning. Deigh says:

 In chapter 21, for instance, he defines ‘liberty of subjects’ after first defining ‘liberty’. His definition of ‘liberty’ is the absence of external impediments to motion. His definition of ‘liberty of subjects’, is the absence of laws forbidding action. Hobbes, however, also observes that laws in themselves are not external impediments to action and hence that one speaks absurdly in holding that a law deprives one of liberty. ‘Liberty’, therefore, when it occurs in ‘liberty of subjects’, does not have the meaning with which Hobbes defines it. As Hobbes puts the point, in such occurrences it is not used with it proper sense (Deigh 2016: 310).

In other words, ‘law of nature’ is a multi-name, technical term and should be understood as a unity as if it were a fused expression, ‘law-of-nature’ (like ‘morning glory’, which is sometimes written ‘morning-glory’), or as if it were spelled without intervening spaces (‘lawofnature’, like ‘butterfly’). Deigh alludes to Hobbes’s statement that only natural liberty is ‘properly called liberty’ (Hobbes 1651: 108). So, Deigh infers, for Hobbes, civil liberty is not liberty.

In chapter 21, the division of liberty into natural and civil strongly suggests that each is a kind of liberty because each signifies the absence of a relevant kind of ‘Opposition’ or ‘external impediment’ (Hobbes 1651: 107). That ‘liberty’ in ‘civil liberty’ is a genuine kind of liberty is suggested by Hobbes’s claim that the ‘Liberty of Subjects’ extends to ‘all kinds of actions, by the laws praetermitted’ (Hobbes 1651: 109). In the Latin Leviathan, Hobbes says ‘a citizen is said to have liberty for those actions about which nothing is laid down in the laws, and these things alone’ (Hobbes 1668: 105: ‘Libertatem habere Civis dicitur illas actiones de quibus in Legibus nihil definitur, & ad eas solas’). If civil liberty were not a kind of liberty, then Hobbes’s saying that a citizen has ‘liberty’ would be analogous to saying that a morning-glory is a glory or a butterfly is a fly. (It is also no good to argue that a citizen ‘is said’ to be free but is not really free, just as, calling a tail a leg does not mean that dogs have five legs because Hobbes often uses ‘is said’ when the way things are said is the way they are.)

The civil laws themselves are ‘Artificiall Chains’, which subjects “have fastned [sic] at one end, to the lips of that Man, or Assembly, to whom they have given the Soveraigne Power; and at the other end to their own Ears’ (Hobbes 1651: 108-9; cf. Hobbes 1668: 105). Of course, if one switches from the category of civil liberty to natural liberty (‘corporall Liberty’), then subjects have ‘freedom from chains, and prison’ and it is absurd for them to ‘clamor as they doe, for the Liberty they so manifestly enjoy’ (Hobbes 1651: 109). Hobbes then returns to the category of civil liberty and points out that it is equally absurd for subjects to complain about their artificial chains and ‘a Sword in the hands’ of the Sovereign, for without those things, ‘all other men would be masters of their lives’ (Hobbes 1651: 109). In other words, they would be slaves, unfree persons.

I don’t think Hobbes would accept the objection that artificial chains cannot be impediments because they are artificial, for if he did, then sovereigns, who are artificial persons, would not be persons; and it is crucial to Hobbes’s theory of authorization to have artificial persons who can represent their subjects. Again it would be absurd for him to divide persons into natural and artificial, as he does (Hobbes 1651: 80).

How then should Hobbes’s claim that only ‘natural liberty … is properly called liberty’ and his claim shortly later that ‘we take liberty in the proper sense, for corporal liberty’ be taken (Hobbes 1651: 108, 109)? (In the Latin Leviathan, the second occurrence of ‘proper’ is not translated.) ‘Proper’ may mean applicable to only one thing. But it does not always have that narrow meaning. It can also mean more appropriate or strictly applicable to one thing than any other, as when someone says, ‘The proper part of the U.S. Constitution excludes its Preamble.’ My friend Jonathan has told me that my fedora is a proper hat unlike my ‘gimme hat’. (Caps are hats, as a person wearing a cap would quickly discover in a courtroom with a sign saying ‘No hats’.) From ‘Natural liberty is proper liberty’ one is not justified in inferring ‘Civil liberty is improper liberty’, only ‘Civil liberty is nonproper liberty’. Nonproper liberty could be ‘liberty’ in a derivative or broader sense. Deigh should not put too much weight on the word ‘proper’ because Hobbes did say that when the theorems of chapters 14 and 15 are delivered in the word of God, they are ‘properly called Lawes’ (Hobbes 1651: 80). Also, seventeenth-century usage allowed for x-es that were not ‘proper’: a kind of human liberty was not ‘proper’ (Owen 1643: 125); and human kings were kings but only God was a ‘proper king’ (Rutherford 1644: 39).

There are three positive reasons for holding that natural laws are laws. First, Hobbes divides laws into two kinds in chapter 26, one of which is natural law (Hobbes 1651: 48). If natural laws were not laws, then it would be absurd for him to make this division, similar to dividing horses into two kinds, equine and saw. The second reason is the way Hobbes explicates his definition of ‘law of nature’. He begins by saying that law is inconsistent with right, where right is liberty, Rights are contrasted with laws: ‘Law, and Right, differ as much, as Obligation, and Liberty; which in one and the same matter are inconsistent’ (Hobbes 1651:  64). If a law of nature were not a law, then it would be pointless or misleading for Hobbes to use the concept of law and its contrary ‘liberty’ to explain his definition. Hobbes is not using the words ‘law’ in ‘law of nature’ metaphorically or as part of a fused expression.

Third, Hobbes indicates in Leviathan, chapter 25, ‘Of Counsel’ that a complete law consists of two parts. One part consists of a proposition that is action-guiding, such as ‘you lay down your right to all things’ or ‘you do not drive faster than 20 mph in a school zone’. The other part is the force with which that proposition is expressed; and the force is determined by the speaker expressing the proposition or with what purpose. When the proposition is issued by an authority as binding, it has the force of a command; and it is a law. When the proposition is expressed, say, by a friend who recommends a course of action for the addressee’s own good, it is counsel or advice.[2] (Hobbes does not consider the fact that authorities are sometimes friends and use their authority for the addressee’s own good.) It is easy to be mistaken about the force that a proposition has. Hobbes uses ‘Do this’ as his example. One defect of this example is that the words are so vague or noncommittal, ‘do’ [what?] ‘this’ [what?]. My guess is that Hobbes feared that a sentence such as ‘You do not kill’ would be interpreted as a command and one such as ‘You study hard’ would be interpreted as counsel. Another deficiency of ‘Do this’ is that it is in the imperative mood, as I explained above. So, Hobbes might have used an indicative sentence to illustrate the confusion of commands and counsels. ‘You are off the streets by midnight’ would have served. Although most philosophers say that sentences with the main verb in the indicative mood are assertions, there are many reasons to think that view is wrong (Martinich 1984:  72-6). In any case, if Hobbes knew speech-act theory, he might have said that the relevant sentence is indeterminate with respect to its force since none is expressed and it can be used either to give advice if the speaker was advising a person how to me alert at 8:00 AM or issue a command if the speaker was commanding the addressee about a curfew.

As I said, the propositional part of a law of nature can be proved by reason from the definition of ‘law of nature’. But who is the authority or sovereign who commands these laws? Hobbes says it is God, who is sovereign in virtue of his irresistible power. And God promulgates the content of the laws through reason, which is ‘the undoubted word of God’ (Hobbes 1651:  195). Since no one in seventeenth-century England doubted that God commanded the laws of nature, Hobbes did not need to produce a proof.[3] Richard Cumberland complained that Hobbes did not prove that God commanded them and attempted to do so in the 400 substantial pages of De Legibus Naturae (1672). Though he labored mightily, I see nothing in that book that would count as a demonstration. Later, John Locke tried and failed to prove that the laws of nature were commanded by God; and he knew that he had failed. But he did not give up his belief that they were.

When Hobbes is talking only about the propositional part of a law, the part that can be proved by reason, he says that the laws of nature are not laws and sometimes signals this narrow reference by using the term ‘dictate of reason’. Why should he use a different term unless he wanted to mark a difference? And when he says that the laws of nature are laws, he often makes clear who the authority is who makes the propositions laws, God.[4]



Deigh, John. 2016. “Political Obligation,” in Martinich and Hoekstra, pp. 624-40.

Martinich, A. P., and Kinch Hoekstra, eds. 2016. Oxford Handbook of Hobbes. New York: Oxford University Press.

Martinich, A. P. 1984. Communication and Reference. Berlin: Walter de Gruyter.

Owen, John. 1643. Theomachia Autoexousiastike; or A Display of Arminianisme. London.

Parkin, Jon. 2016. “Hobbes and Paradox,” in Martinich and Hoekstra, pp. 624-42.

Rutherford, Samuel. 1644. Lex, Rex; The Law and the Prince. London.

Searle, John. 1969. Speech Acts. Cambridge: Cambridge University Press.



[1] Hobbes does not point out that, given his characterization of command and counsel, something can be both: Suppose there is a good God, a sovereign by nature, who wants to nudge human beings, and so directs the laws of nature to human beings, which he expects them to understand through reason as his mode of promulgation. The omission does not surprise me. Hobbes was a fallible philosopher, as are we all.

[2] This two-fold structure is familiar from speech act theory. See Searle 1969 and Martinich 1984.

[3] My guess is that, given Hobbes’s epistemological views, a person could never have more than empirical evidence for a belief that God commanded the laws of nature.

[4] I want to thank Leslie Martinich and Jake Galgon for their helpful comments.