Chapter 3

The Making of Defence Policy

It has been suggested above that a key task for a government involved in defence transformation is the construction of a sensible defence policy. In the past, this policy may have been dictated by others, or necessitated by a security environment which has now changed completely. In this chapter, therefore, the organisation and structure of defence policymaking, and the larger context into which it must fit will be considered. The respective roles of civilians and the military in this process are discussed in the next chapter.

The need for a hierarchy

All governments pursue policies of various kinds, and there will be many connections and overlaps between policies in different areas, some designed and some not. Defence policy is, in the end, a component rather than an objective itself, and so is best thought of as a member of a hierarchy of policies which will run along the following line:

Two caveats must immediately be raised:

Firstly, this dependence on the foreign policy hierarchy is not exclusive. In some countries, defence policy can be partly a subset of domestic policy — in the internal security area, for example — and overall policies on finance and the budget, industry and manpower will obviously have implications for the way in which defence policy is made.



Secondly, this hierarchy does not only work in one direction. The assets belonging to the military, as well as training and experience, can all influence — and even determine — aspects of foreign and security policy by enabling possible courses of action or not. The size of the military, the budget, the procurement programme and so forth, will all be important considerations in the formulation of other types of policy.

Nonetheless, this hierarchy is helpful as a reminder of the order in which things must happen. If defence policy, especially if made by the military, comes first and is allowed to dominate, then foreign and security policy will be nothing more, in effect, than policies designed, as far as possible, to allow the military to do what it wants with the least damaging consequences. This is pretty much what happened in Germany in the years before 1914, and in Japan before 1941, with results which do not need spelling out in detail. Defence policy is situated towards the bottom of the hierarchy, because it is essentially about execution: it is one of the practical ways in which foreign policy is given effect. As long as the ends (foreign policy) determine the means (defence policy), then the relationship is a healthy one.

A few words about Clausewitz

The primacy of political aims in war is the result of a doctrine generally — and rightly — associated with the great Prussian military thinker Carl von Clausewitz (1780-1831). Since Clausewitz is very often misunderstood, it may be worth a brief section on what he meant. Clausewitz is popularly supposed to have said that war "is the continuation of politics by other means", and this has impressed some people as shameful and others as inaccurate. Clausewitz made a number of statements of this general type, but his discussion in Book I of On war — the only one to be fully revised — should be taken as definitive. While it is true that part 24 of Book I is headed ‘War is merely the continuation of policy by other means’, he adds that:

"war is not merely an act of policy but a true political instrument, a continuation of political intercourse, carried on with other means ... The political object is the goal, war is the means of reaching it, and means can never be considered in isolation from their purpose."12

Two things obscure what this means. The first is language, with English being unusual in having separate words for policy — deciding and implementing a plan — and politics — the activity carried on by politicians to get power and win votes. Here, a ‘political instrument’ might be better expressed ‘an instrument of the policy of the state’. What Clausewitz is not saying is that war is just a continuation of normal political life: today a vote in parliament, tomorrow a press conference, on Wednesday, Russia is invaded. Secondly, Clausewitz speaks from the pre-Nuremberg era, when states could and did contemplate aggressive wars to further their foreign policy goals. It is better, perhaps, to say something like ‘the use of military force’ to avoid confusion today.

Put like this, all that Clausewitz is really saying is that it is pointless to make use of military forces without a wider object in view. The means (military forces) must never be allowed to dictate the ends. The corollary, of course, is that military planning and operations must always be carried out in a wider political context, and that defence policy must always serve the overall policy objectives of the government. Although this seems logical enough, it is surprising how frequently it has been ignored. Usually, the reason is that the military, whose approach to war is necessarily technocratic, has too powerful a voice in what is to be done. It is not enough to declare that there is a larger aim in view before conducting military operations, and it is unlikely that many operations have been launched with no idea about what they were intended to achieve. But, there are two important criteria which have to be met before there is any chance of success:

The measures must be appropriate, ie there must be some logical connection between the military action and the political objective.



There must be an understanding of how the one is to affect the other, and some means of measuring progress towards it.

How is this to be done? It takes place at the operational level. This lies between the strategy (mainly the concern of politicians) and the tactics (mainly the preserve of the military), and is the hardest to get right, whether in a major war, or in a peacekeeping operation. Failure usually occurs because civilians and military members do not really understand enough of each other’s business to have a sensible conversation, and the two sides become separated from each other: no-one is looking at where military operations and policy objectives actually overlap.

An example of successful operational level practice would be the management of the Vietnam war by the government in Hanoi. There was a clear political goal (unification of the country under Hanoi’s control). This could not be accomplished until the Americans went home, so that had to be achieved first. This implied a long, patient, but low-level military campaign, designed to convince the Americans that they could not win, fought in such a way that it negated, as far as possible, American superiority in numbers and weaponry. It also dictated a policy of the use of force to achieve what were mostly political, rather than military ends, and tactics were developed accordingly. By contrast, the Americans were stuck with ill-defined objectives, a disconnection between political and military policies, and a tactical doctrine which stressed the defeat of the enemy through superior firepower. Once the Americans had left, the emphasis shifted to a conventional war against Saigon.

Defence policy and security policy

At this point, a few words on the relationship between defence policy and security policy may be helpful. As part of its foreign policy, a government will have a series of relations with states, or groups of states, which have a defence and military component of some kind. The desire to join a regional security structure of some kind will be partly political (to heighten profile, or to build a closer relationship), and partly defence-based (to make use of other nations’ experience or to achieve concrete security gains, for example). Likewise, the management of this membership, once it has been gained, will involve a great deal of work which is partly political and partly military.

Such subjects as these are the preserve of security policy. Other examples would include arms control, weapons non-proliferation, peacekeeping, treaties and deployments.

There are also a whole series of subjects, such as joint exercises or the procurement of foreign equipment, where there is a foreign policy dimension that has to be respected. Security policy is best characterised in practice as the area of government policy where both the foreign and defence ministries have a major interest and responsibility.

Take, for example, the first deployment of a new class of ship in an expanding navy. It is necessary to decide which countries the ship will visit, for how long, and what it will do.

Obviously, there are some essentially naval objectives in such a visit, such as gaining operational experience, practicing deployments away from home, and learning how to cope with unfamiliar ports.

There are wider defence objectives as well: getting to know other navies in the region, promoting defence contacts and raising the profile of one’s own military forces. There may be valuable public relations and recruiting benefits.

Equally, there will be major foreign policy concerns. Some nations may welcome a visit, others would not. Ambassadors in several countries will be lobbying. The ship’s visits might coincide with visits by the president or a trade show.

None of these single sets of arguments is in itself decisive, and the eventual itinerary will be an amalgam of all of them.

Security policy is perhaps the most complex and difficult area of policymaking, and the hardest to undertake successfully. It requires foreign and defence ministries to co-operate closely, and to be reasonably knowledgeable about what the other does. It requires an outcome which actually makes sense in its own terms, and is not simply the result of splitting the difference. Countries where this kind of co-ordination is a problem (such as those with weak defence ministries) frequently find themselves committed to a position which makes little sense. By contrast, countries where the hierarchy set out above is not respected, and the military has too strong a voice, frequently saddle themselves with policy objectives which cannot be implemented; in that deadly phrase employed by diplomats, ‘not negotiable’. This leads to the concept of the ‘policy community’.

The policy community

Institutions are bound to disagree with one another from time to time. Any policy which makes sense, and any position which is robust and commands respect, in the absence of a strategic genius, will be the product of more than one hand, in a way which resolves these disagreements. But, this is not to say either that the resulting policy must be a simple compromise, or that it must represent a victory by one institution over another. Nor is it necessarily the result of some complex bargaining process among interest groups.

There are, broadly, two ways in which these disagreements can be resolved, depending on the administrative traditions of the state. In a hierarchical bureaucracy of specialists, with powerful leadership from the top (the kind of bureaucracy which Max Weber assumed was normal), policies will tend not to be discussed outside the immediate hierarchy until they have been agreed to by those at the top. Indeed, in a system where power flows downwards and initiative comes from the top, senior officials often become personally committed to ideas and positions early on. By contrast, in a less hierarchical and more informal bureaucracy, ideas can be road-tested at a lower level first, and the egos of powerful individuals need not be endangered. It is a question, really, of where the inevitable arguments are and when they are resolved. Human nature being what it is, the earlier these issues can be resolved, the easier the process of decision will be. Although elements of the second type of bureaucracy are recommended in practice, that is only because some systems, in practice, seem to work more effectively than others. It is recognised that these characteristics are deeply ingrained in societies: the German system and the Australian system, although each will no doubt change, are unlikely ever to resemble each other closely.

Yet, even in the most apparently hierarchical system — such as many Asian ones — a great deal of informal, lower-level consensus-building actually takes place, which speeds the final decision along. A development of this practice is the policy community, which is defined as all those with knowledge of the problem and an interest in solving it. The second element is crucial, since it enables people to step outside of their narrow roles and look for solutions without their egos being engaged. There are, of course, people who have a right to be consulted, or who demand that they are. But, they are not the people who are spoken to privately, and asked: ‘What do you think of this as an idea?’ or, ‘One possible solution might be ...’. To repeat, the further up in a hierarchy positions go before being modified, the sharper will the differences with other hierarchies be. The answer, therefore, is to develop the widest consensus at the lowest level, which is a point returned to in the next chapter.

In practice, policy communities are indispensable to the solution of practical problems, and tend to exist in even the most rigid of structures, whether or not they are openly acknowledged. The less common they are, the more individual ministries or other hierarchies will have a private discussion, uninformed about what other people are thinking, and will take an inflexible position, often without realising that it is very different from the position of others. If political figures are involved, the resulting struggle can become very personal; even if not, the status and dignity of hierarchies can become involved. Decisions made under this system are often bad ones, because compromise has become difficult, and, once the decision is made, there will be clear winners and losers. If this kind of system works at all, it is usually because a strong figure intervenes — often a president — but, of course, this individual is most likely to choose from among several options, rather than to try to find a consensus which serves everybody’s interests.

The issue of the proper organisation of departments and ministries interested in defence and security problems is a complex one. It is not really a question of dominance, but a question of management: How are the available resources put together in the most productive way? In turn, of course, this question cannot be divorced from the question of the organisation of the state itself, of which this will be an example.

There are two basic organisational models of government co-ordination, and two procedural ones. In each case, the distinction is between a ‘top-down’ and a ‘bottom-up’ approach. The two organisational variants are:

a strong central organisation, working for the head of state or government, which can try to initiate policy and involve itself in a degree of detail; and



a system in which power is distributed mainly to departments, with only a small central institution, which probably only has a co-ordinating role.

As always, these are tendencies rather than absolute types. Most European systems feature a strong central staff of some kind, although its ability to initiate policy, as opposed to just blocking things it does not like, can be limited. While this system will generally produce a definite answer, this is often only when the president’s foreign affairs advisor finally gets this far down in his or her in-tray. The penalty is usually lost time, and an outcome which favours one ministry (usually the foreign ministry) over another. But, the absence of a strong centre creates its own problems, since arguments can often go on for ever. The Japanese system, for example, always takes a long time, because of the need to find consensus between departments, and this reflects the generally distributed nature of power in Japanese society as a whole.

The two procedural variants are:

a system which makes a clear division between policy, made by the minister and his or her advisors, and execution, undertaken by officials; and



a system where officials themselves help with the formulation of policy, as well as its implementation.

The first system makes co-ordination between departments rather more difficult, because, until the minister or his or her advisors has spoken, officials can never be sure what line they should propose to others. These things are obviously easier when officials themselves originate policy, since departments can work together on something which they jointly present to ministers.

These variants will greatly affect the mechanics of the way in which policy is made. An extreme case is perhaps that of the United States, whose policymaking system is large and complex, and which finds consensus on many issues difficult. Another complication is that the higher levels of American administration will be sprinkled with political appointments, many of whom are mainly interested in making a reputation for themselves so that they can further their careers when they leave government. The usual American response is to set up a policy co-ordinating committee to produce a compromise position which, like most compromises, is so complex and fragile, that US representatives then have almost no margin for manoeuvring.

Given the inherent complexity of defence and security issues, the large number of interests involved, the degree of control exerted by allies and neighbours, and the difficulty of the co-ordination process, it is reasonable to ask whether there are any steps which can be taken to make the process run more smoothly. There is no doubt that sharp demarcation lines and institutional rivalries cause problems. In many administrative systems, the responsibilities of government departments and even offices within them, are set down in writing, and have the force of law. In this kind of system, it is difficult to share power, and a department — usually the foreign ministry — will, in theory, have charge of security questions. But, in practice, the help of the defence ministry is going to be needed if sensible policies are to be adopted. One of the weaknesses of a system like this is that important defence decisions, because of their political element, are made by people in the foreign ministry, the cabinet or the president’s office. The military will then be given a fait accompli, arrived at for political reasons, and asked to make it work.

If there is a solution to this kind of problem, it is to exploit the maximum flexibility which the system will allow. It is helpful if those who work in one department have at least some experience of what other departments do, and some understanding of what their priorities are. It is not necessary to be transformed into a mini-diplomat, or a mini-officer. But, an awareness of how an official’s opposite number thinks prevents a dialogue of the hearing-impaired, and increases the trust that, when someone says something which sounds reasonable, but which cannot be verified from own experience, they are probably right. While it is customary for foreign ministries to speak on behalf of the delegation at international meetings, it would be odd indeed if they did not have colleagues from other departments available next to them.

At a more institutional level, it is a good idea, if the system will permit it, for positions on any issue which may involve co-operation between departments to find initial agreement from the bottom up, and preferably among people who already know one another. The alternative, which is separate analyses, followed by negotiations between two or more ministries, tends to result in a victory for the more powerful ministry (or minister), rather than the best idea. Sensible human beings ought to be able to put together a position which makes sense from a variety of angles.

This process is itself assisted if there are a fair number of civilian defence officials. This point will be returned to later, but here it should be stressed that experience suggests that relationships between diplomats and military persons, without intermediaries of some kind, are often difficult. There are officers who are politically aware, and there are diplomats who know something about defence, but probably not enough of each, and seldom in the right place at the right time. Diplomats are members of an international ‘free masonry’ dedicated to ensure smooth relations between states, who sometimes believe that agreements and good relations between states are really ends in themselves. This, combined with the special language and baroque arrangements under which negotiations take place, make life very difficult for the average officer whose previous job may have been commanding an infantry battalion, where things were somewhat simpler. It is the task of the civilian bureaucrat to understand both sides, and to help in the process of forming consensus. The bureaucrat will be used to working with both sides, and will tend to deal with the diplomat more easily. Their background and training, as well as the type of work they do, will be more similar, and they can, by taking on the political side of the defence ministry’s involvement, leave the military to concentrate on their own areas of expertise.

So far the concentration has been on the idea of a policy community within government. There are many states where this is the norm, and where there is limited interest outside government in defence and security policy. These are often settled states, whose militaries are widely regarded as legitimate. But, there are many other states where the situation is quite different.

For example, there are some states where the constitution provides for the separation of powers, which may give parliament an active role, even if only a negative one (the United States is an extreme example). More generally, electoral systems with proportional representation, which tends to produce coalitions, will generally move the balance of power to parliament, because of the need to secure agreement from the parties and keep the coalition together. There are also states whose recent history of civil-military relations has been poor, or where the military has played too strong a role in politics. In many cases, it will be correct to bring civil society into the policymaking process deliberately as a way of restoring confidence in the military. Finally, there will be states where the military has a mainly internal focus, or where there are major political issues — such as conscription — which are of interest to the general public.

It is doubtful whether there are any general rules about the best way to involve civil society in defence issues. Perhaps the only fundamental one is the need to avoid confusing functions. It is generally right for government to originate proposals, because it has the mandate (and the resources) to do so. If parliament turns itself into a second centre of decisionmaking, the result is frequently chaos and paralysis, since the two can cancel each other out (as in the United States), and nothing gets done. Parliament here represents the interests of the voter, the citizen and the taxpayer, and fulfils its task best when it requires the government, or its officials, to appear before it and give a reasoned explanation of what has been done in its name. This will be discussed further in a later chapter.