Why are some countries’ constitutions easy to understand, while others are extremely difficult to interpret? James Melton outlines results from a cross-national study of the interpretability of constitutions. He notes that European constitutions are more difficult to interpret, on average, than constitutions from any other region of the world. While the negative consequences of low interpretability can be mitigated by other factors, making constitutions easier to understand should be a priority for future constitutional drafters.
According to research recently published in the British Journal of Political Science by Zachary Elkins, Tom Ginsburg, Kalev Leetaru, and myself (available here – ungated until the 18th of June), Europe is home to some of the world’s most incomprehensible constitutions. The article offers the first empirical assessment of constitutional interpretability, which is defined as the ability of a constitution to produce inter-subjective agreement about its meaning. The measure of constitutional interpretability developed in the article uses data from the Comparative Constitutions Project (CCP). For each constitutional text, two (or more) coders have answered a series of questions on the structure and content of that text using the CCP’s survey instrument. Disagreement between coders’ responses indicates a problem of interpretation. Constitutional interpretability is assumed to be correlated with the number of interpretation problems that arise in the CCP’s data. Constitutions with low interpretability are expected to have more problems of interpretation than constitutions with high interpretability.
Figure 1: Predicted Interpretability of 25 European Constitutions
Notes – Interpretability is calculated using equation 4 in Melton, Elkins, Ginsburg, and Leetaru (2013). The measure is adjusted to remove the variance associated with the interpretation procedure (i.e. the coding procedure and coders employed by the CCP) and rescaled to range from zero to one, where zero represents the lowest observed level of interpretability and one the highest observed level of interpretability. The constitutions represented in the figure are those in force in each country circa 2006.
Based on this measure, the average European constitution has a lower level of interpretability than the average constitution from any other region. Figure 1 illustrates the predicted interpretability of a sample of 25 European constitutions. Each dot reflects the level of interpretability of the constitution in force in that country circa 2006. The vertical reference line illustrates the mean level of interpretability across all 397 constitutions for which an interpretability score can be calculated. Of the 25 constitutions represented in figure 1, only three – Montenegro’s constitution of 2006, Serbia’s constitution of 2006 and Spain’s constitution of 1978 (as amended in 1992) – have a higher level of interpretability than the sample mean.
Why are Europe’s constitutions so difficult to understand?
Problems of interpretation arise from three sources: the context in which the constitution is written, the text of the constitution itself, and the procedure used to interpret the text. Although most of the variance in interpretability is attributable to the procedures used to interpret the text (i.e. the coding procedure and coders employed by the CCP), this does not explain the low interpretability scores received by European constitutions. Even after removing the variance associated with the interpretation procedure, Europe’s constitutions are still among the least interpretable.
Another potential reason for Europe’s low levels of constitutional interpretability is the context in which those constitutions were written. European constitutions tend to be older than constitutions written elsewhere because European countries are older. The average constitution in most regions was promulgated in the late 1960s or early 1970s, but in Europe, the average constitution was written in the 1920s. In fact, about half of the constitutions in our sample written prior to the 1960s are from Europe, and a number of those early constitutions are still in force today. Since language, culture and legal norms change over time, modern readers might have trouble understanding texts written nearly a century ago. The study does find an association between the age of the constitution and interpretability, but the effect is quite small and unlikely to explain the consistently low levels of interpretability found in Europe. In fact, there is little evidence that any of the contextual factors assessed in the article have much of an effect on interpretability.
The third, and final, source of variance in interpretability is the text of the constitution itself. Several textual factors are associated with low levels of interpretability. Perhaps the most important of these is constitutional specificity. Constitutions that cover a lot of topics tend to have lower levels of interpretability, while constitutions that cover topics in greater detail tend to have higher levels of interpretability. In addition, constitutions that have a higher percentage of one time words tend to be easier to interpret. These textual attributes are probably the best explanation of Europe’s low levels of interpretability because European constitutions tend to be shorter, more detailed and have fewer one time words than constitutions written in other regions of the world.
The effect of the aforementioned textual attributes suggests that a constitution’s age might have an indirect effect on interpretability. Constitutions written recently tend to be longer and much more detailed than their counterparts written a century ago. The former more closely resemble legal statutes than classic examples of constitutions written in the United States or in 19th century Europe. They tend to have lots of short provisions that clearly and fully articulate the institutional structure and, as a result, are easier to interpret.
What are the implications of low interpretability?
There is significant debate about the virtue of interpretability. Arguments supporting vague or unclear laws are not uncommon. However, advocates of plain language note several benefits associated with interpretability. For instance, a necessary condition for compliance with the law is that subjects understand its meaning. Without clear laws, one is unsure how to behave and, thus, the relationship between the enforcers and the subjects can become arbitrary. Interpretability is also related to theories of constitutional self-enforcement. Self-enforcement requires that subjects of the constitution are willing to take coordinated action to punish those who violate its tenets. Such coordination is only possible if there is agreement on what constitutes a violation and how violations are to be punished, which is facilitated by an easily interpretable constitution.
Ultimately, interpretability facilitates adherence to the rule of law. Uninterpretable constitutions do not make such adherence impossible, but neither do they support its development. There might be few consequences of having an unclear constitution in Western Europe, where lack of interpretability is counterbalanced by durable constitutions and elites that are committed to the rule of law. Lack of interpretability is likely more problematic in Eastern Europe, where the foundation for the rule of law is still in its infancy. I am not suggesting that constitutions in Eastern Europe (or anywhere) should be rewritten to make them more interpretable, but I do think that interpretability is something for which future constitutional drafters, and possibly the Venice Commission, should strive.
Please read our comments policy before commenting.
Note: This article gives the views of the author, and not the position of EUROPP – European Politics and Policy, nor of the London School of Economics.
Shortened URL for this post: http://bit.ly/15jo3vQ
James Melton – University College London
James Melton is Lecturer in Comparative Politics in the Department of Political Science at University College London.
Since the Italian Constitution has a very high interpretability for the Italians but you say that it has a low interpretability, I have to say that your method of assessment has a very low interpretability.
I agree with Giorgio about the Italian Constitution: although it has been written decades ago, it still holds a strong power; citizens and policy makers care about it, the text is so good that it’s easy to understand and pleasant to read (just to mention a fact: Roberto Benigni made an entire show on our Constitution, showing the beauty of it).
I believe that Mr Melton’s research points out a result that is achieved with different criteria but at a general level that doesn’t really takes into consideration all of the different specifics that every Nation has, therefore I would suggest to start analyzing the situation from these results but then, as a second step, a debate with national constitutional expert should be launched in order to develop a deeper understanding and confrontation with each european reality.
Bosnia? Dayton Const. (:
It seems an interesting work. I personally have some doubts about the method and the interpretation of the results.
As for the method, constitutions (as any other legal norm) are not working in a vacuum. Therefore, interpretation is necessary and is done in accordance with the legal and cultural context. The fact that a constitution is interpreted and its interpretation changes over time is what keeps it alive and adaptable to the new social needs. (I see in the last paragraph of your post you consider that point).
Many constitutions listed as “low interpretability” are actually quite clear (in most of their parts) for national people and lawyers (e.g. France and Italy). Plus, modern constitutions (after the WWII) generally have different parts, some more general (written in plain language) some other more technical (written for lawyers). Even more some part have been added later or have been changed. I am curious if this has been taken into consideration when constitutions were analysed.
As for the interpretation of the data, I am not sure about the statement “interpretability facilitates adherence to the rule of law”. Some pretty clear pieces of law have been crafted under totalitarian regimes… Finally, it seems that the interpretability of the constitution does not matter much in the end, as the ultimate outcome depends on external factors, e.g. “elites committed to the rule of law”.
Anyway, thank you for your post that stimulates debate!
As Giuliano already wrote: “I am not sure about the statement “interpretability facilitates adherence to the rule of law”” To test this proposition the author(s) would indeed need to demonstrate that there is a systematic disregard for rule of law principles. For example, is there a systematic deferrence by judges towards legislative or executive in countries w/ low interpretability scores. Or use survey data, crime statistics et cetera. And what about countries without written constitution or judicial review?
What is also missing: A lot of difficulties in interpretability might stem from frequent constitutional amendments (e.g. Germany, see Astrid Lorenz’ work on this). Yet constitutional amendments may indicate two things: adherence to the rule of law (otherwise why would decision-makers bother to change constitution) and/or constitutional provisions have been previously perceived as ambigious by some party and thus clarification was needed. Note also that ambiguity is nothing objective, but usually depends on how interested/affected/litigating parties try to frame constitutional provisions in a favorable light for themselves (for an example on the EU of this logic, see Jupille 2004)
Less is more! Plain language and short sentences make a text clear, and a clear constitution is essential to the rule of law. But generally, I would also recommend a short constitution. More people are likely to read it and comment on it. The drafters, too, will find it more manageable.
I’d be interested to see where Kosovo would fall on your chart. The 2008 Kosovar constitution has almost 20,000 words, but has stirred intense debate about interpretation. Dictated by international politics, the text covers too many topics, lacks some important details, and has very few one-time words, while the language is often imprecise and ambiguous.
Finally, the constitution is supposed to be the fundamental law, not an administrative code. And some issues are not just about words. You wish to explicitly prohibit “discrimination at restaurants and hotels,” fine. But what about bars or swinger clubs?