The perils of overspecifying "parliamentarism"
I consider "Why Not Parliamentarism?" a "working book". In the same way many people publish working papers so that they can get more feedback before they publish the final version of their text, I wanted to publish what I had already written and get as much feedback as possible and then publish a modified version later on. Perhaps the part of my book which is getting the most criticism is the definition I use for parliamentarism: executive subordination to the legislature. This subordination exists in a continuum. However, because countreis tend to divide themselves as those with a lot of executive subordination to the legislature and those with little subordination, the categories of "presidentialism" and "parliamentarism" are still very descriptive. Although I am very satisfied with how it describes the idea we make of parliamentarism and the differences in observed behavior of governments and other types of organizations, many scholars are wholly opposed to it. For an influential part of political science, parliamentarism is defined by the presence of the confidence requirement. Przeworski, for example, writes:
"While in parliamentary systems the government is elected by the legislature and in presidential systems by popular vote, the crux of the distinction is whether the executive can be removed by the legislature. Systems in which the government cannot be removed by the legislature are presidential; those in which it can be are parliamentary (or, where the president is popularly elected, mixed)."
But this definition is too restrictive, and clinging to it hurts not only our understanding of parliamentarism, but our progress prospects. That is, this debate matters a lot for our actions, it is not a mere intellectual exercise.
Definitions come in different forms: they can be legal, based on use, conceptual, or operational. Think of the word "democracy". In international law, it does not have a precise definition, but we do have a UNGA resolution which highlights some common features, including
"respect for all human rights and fundamental freedoms, inter alia, freedom of association and of peaceful assembly, freedom of expression and opinion, freedom of thought, conscience, religion or belief, the right to be recognized everywhere as a person before the law and the right to take part in the conduct of public affairs, directly or through freely chosen representatives, to vote in a pluralistic system of political parties and organizations and to be elected at genuine, periodic, free and fair elections by universal and equal suffrage and by secret ballot guaranteeing the free expression of the will of the people, as well as respect for the rule of law, the separation of powers, the independence of the judiciary, transparency and accountability in public administration and decision-making and free, independent and pluralistic media"
For a usage definition of democracy we can look into a dictionary and find the very simple " government by the people ".
Academics will concentrate themselves, however, on conceptual (or explicative) and operational definitions. The conceptual definition of democracy is cause for much debate. As Wikipedia tells us, " No consensus exists on how to define democracy – indeed, one study found that at least 2,234 descriptions of democracy exist in the English language - but legal equality, political freedom and rule of law have been identified as important characteristics." Overspecifying democracy is a known risk. Most people recognize that democracy is not merely a matter of calling one's country democraticn, nor does it equate holding elections. Lastly, there are operational definitions of democracy. Scientists want to study the phenomenon of democracy and, in the absence of a perfect definition, they use operational definitions which catpure important aspects of said phenomenon. In the case of democracy, there are several operational definitions: ranks above X in Polity IV, for example.
Coming back to parliamentarism. What reason do we have to define it exclusively as government subject to parliamentary confidence? It is not legal, for sure, no such binding definition exists. It is not from use either. If we look at the dictionary definition of parliamentarism (or parliamentary goverment) we variations of the Merriam Webster definition: "a system of government having the real executive power vested in a cabinet composed of members of the legislature who are individually and collectively responsible to the legislature." We see that this definition does include the confidence of the legislature, but it also includes aspects which most scholars will reject define parliamentarism: that the power be vested in a cabinet (many parliamentary countries have prime ministerial government instead of a cabinet government), and that the cabinet is composed of members of the legislature (many parliamentary countries do not have members of parliament as ministers). Most likely this overspecification is due to the characteristics of British government when the concept of parliamentarism was first being devised. So the argument can't be "check the dictionary".
Is it a definitive conceptual (or explicative) definition? According to Carnap, “[t]he task of explication consists in transforming a given more or less inexact concept into an exact one […]. We call the given concept (or the term used for it) the explicandum, and the exact concept proposed to take the place of the first (or the term proposed for it) the explicatum.” The idea is to use a concept already in use by people intuitively and try to tear out what are the commonalities it has, and also what makes it different from other things. The Internet Encyclopedia of Philosophy brings the following example:
A widely accepted instance of a simple and largely unproblematic explication is the 2006 definition of ‘planet’ by the International Astronomical Union (IAU). The discussion about that term was triggered by a number of discoveries of objects in orbit around the sun that are similar to the nine bodies that had until then been recognized as planets. Since there was no binding definition of ‘planet’ at that point, insecurity arose about whether to call certain objects planets. The IAU member assembly established a definition according to which a planet is “a celestial body that (a) is in orbit around the Sun, (b) has sufficient mass for its self-gravity to overcome rigid body forces so that it assumes a hydrostatic equilibrium (nearly round) shape, and (c) has cleared the neighbourhood around its orbit” (IAU 2006). This disqualified Pluto as a planet, whereas the other eight planets kept their status; and to a large degree the new understanding of the term ‘planet’ incorporated key aspects of the earlier use patterns, while at the same time being much clearer
Does the confidence requirement satisfactorily capture what makes parliamentary countries distinct? No. Let us think of the "perils of presidentialism" as identified by Juan Linz: winner-takes-all, dual legitimacy, fixed mandates, personalism. While the confidence requirement will present a huge blow to the winner-takes-all nature of presidentialism, by itself it may have many issues containing the other problems. I have already written about how Weimar Germany, a "parliamentary" country by this standard, suffered from the symptoms of presidentialism. A powerful president claimed legitimacy for himself (backed by influential thinkers such as Carl Schmitt). That president had a fixed mandate which allowed him to confront parliament without fear of losing his position. The figure of the president and the perceived incapacity of the parliament bred a kind of politics very much centered in personal figures. But we also see these problems in the failures of constitutional monarchies which were "parliamentary" by the confidence requirement. The fascist rise in Italy had an important participation of King Emmanuel III. Dual legitimacy, fixed "mandate", personalism. A similar scenario developed in Greece.
The confidence requirement also leaves out countries with constitutions which do avoid many of the perils. Take Switzerland. Its government cannot be removed by a simple vote of non-confidence, so it is not considered parliamentary according to that definition. But the government is appointed by the Federal Assembly. This allows them to avoid the dual legitimacy problem quite well. The legitimacy of the government completely derives from parliament, not direct elections. Because there is no prime minister or president of Switzerland, they also avoid personalism and winner-takes-all. This means Switzerland functions very similarly to a pure parliamentary government.
Going further, is it not natural to think of parliamentarism as "rule by parliaments"? If so, it would allow us to include other important phenomena as part of one big phenomenon. Take, for example, Gary Cox's thesis that a main explanation for the credibility of sovereign debt is the parliamentary monopoly on authorizing the sales of sovereign promises. This credibility allowed England to borrow much more easily than other countries, which contributed to its development. Is this not an element of the establishment of parliamentarism in the country? I think it is. But it is not related to the confidence requirement. To stay with the English example, was the English Civil War of the 17th century not a conflict about parliamentarism? The first formal motion of non-confidence in the British parliament was in 1742. Was there no parliamentarism before then?
Having argued that executive subordination to parliament is a better conceptual definition, we can examine the confidence requirement as an operational definition. For that purpose, it is a generally good one. Unlike the more general concept of executive subordination, it can be verified from constitutions, it has a clear answer, and allows for quantitative studies. Indeed, many of the studies I cite use this very definition. It is far from perfect, however, as seen from the examples above. A more robust understanding of the phenomenon will rely on several different operational definitions so that we get a clearer picture. One such definition may well be the one I use in the book, adapted from Tyson Roberts. Roberts calls countries "parliament-based" if the executive is chosen by the legislature. I call them with the more familiar name: parliamentary. That would make Switzerland parliamentary, but not Weimar Germany nor pre-fascist Italy or monarchical Greece. This increases our understanding of the parliamentary phenomenon. I should stress: I'm not proposing we use one or the other, I'm proposing we use different operational definitions to get a clearer picture. Choice of the executive by the legislature is very far from capturing all aspects of parliamentarism, of course.
Several scholars will argue that I'm completely leaving out the concept of semi-presidentialism. But the introduction of the semi-presidentialism does not do away with the problem of a continuum of parliamentary prominence over the executive, it only includes more categories. After we categorize countries in presidential, semi-presidential, and parliamentary, we learn that semi-presidential countries also differ significantly, some are called "premier-presidential" and others "president-parliamentary" (and we still have no name for the Swiss system). But these even more detailed categories will not exhaust the differences either. Is it not simpler to think of a continuum?
Why it matters
Although the discussion above may seem purely academic with no practical consequence, the opposite is the case. The way we understand parliamentarism is very consequential for what actions we think are worth pursuing for strengthening it. If parliamentarism is reduced to having the confidence requirement, I fear it may induce complacency on the promotion of parliamentarism both for presidential and for parliamentary countries. For presidential countries, because such a huge step may be perceived as completely out of reach. This is a regular claim in the literature. But if we think of a continuum, we can take small steps towards that goal and implement it gradually (as was the case in various countries which are parliamentary today). For parliamentary countries, it may induce a neglect of a presidentialization of politics (a presidential backsliding, if you will) which gradually deteriorates the quality of governance and may eventually risk the stability of the whole system. So this is a call for initiative and vigilance. You can always improve your governance in a presidential country by more or less gradual actions which limit presidential powers. You should always prevent creeping presidential/personalized powers in your society as well. If you fight for every inch of democracy, then you should fight for every inch of parliamentarism as well.