Keywords

1.1 A Book Addressing the Lack of Pan-Mediterranean Demographic Studies

The Mediterranean region has been much studied by human and social sciences. The length of its written history, and the variety of civilisations sharing a common history (going back to the Roman mare nostrum), of course, go some way to explaining this wealth of studies. However, the Mediterranean has only recently been studied as a global study area, rather than as separate sub-regions.

1.1.1 In Search of Human Unity in the Mediterranean

The very word “Mediterranean” was coined in the eighteenth century and refers only to the fact of “being inland”Footnote 1 (Bourguet et al., 1998). The Mediterranean was little studied before the nineteenth century, owing to its lack of scientific significance. It was thought of rather as a space of division and separation between very different populations. It was not until three French scientific expeditions to Egypt (1798–1799), the Morea (1829–1831) and Algeria (1839–1842) that the Mediterranean became an object of study in itself. The research thus carried out attempted to give consistency to the Mediterranean area by seeking to identify common elements on both sides of this geographical area, which would make it possible to affirm the unity and homogeneity of the region.

The botanists on the Egyptian expedition discovered species common to Syria, the Barbary Coast and Southern France. Pyramus de Candolle (1820) considered the Mediterranean to be one of the world's 20 botanical regions. This led to the notion of a Mediterranean region, although it was not recognised as having any specific character. The Mediterranean was said to be a transitional botanical region between the three continents, along a north–south and east–west gradient.

It was not until the 1830s that a scientific discourse developed on the cultural unity of the Mediterranean, based in particular on ancient history (Roman and Philhellenic legacies). Thus, this geographical area is less and less perceived as a barrier or a space of division between civilisations, but as an interface linking the different shores, where the sea facilitates the exchange of ideas and trade more than the land (Reclus, 1876).

The Vidalian approach, prominent at the end of the nineteenth century, broke with tradition by positing that the physical homogeneity (climate and landscape) of the Mediterranean had an influence on the lifestyles of its populations. The sea is becoming less important than the land (Sorre & Sion, 1934). This idea of the Mediterranean as a human unit cumulates with the work of Braudel (1949). He was one of the first to articulate the concept of a Mediterranean world. He viewed the Mediterranean as a permanent place of maritime and land exchanges, which changes its meaning throughout history (Roncayolo, 2002). In particular, he emphasised the continuity throughout history, but also the homogeneity of the societies surrounding the Mediterranean Sea. Landscapes are considered as major elements of this homogeneity, in accordance with a certain geographical determinism of the Vidalian approach, which has been vigorously criticised (Péguy, 1986). However, Braudel deserves credit for taking on the Mediterranean and establishing it as an object of study in the human and social sciences.

Other scientific fields have been involved in this question of the geographical, historical and cultural unity of Mediterranean societies. In particular, anthropologists have researched the common characteristics of social life, marking a coherence of the Mediterranean world. For example, Peristiany (1968, 1976) proposed the concept of “Mediterranean modes of thought”. Pitt-Rivers (1963) compared the social structures of rural communities in several Mediterranean countries across the different shores. He argued that Mediterranean unity stems from geomorphological and climatic characteristics. Continuities over time would therefore be little affected by political or religious changes. Others identified concepts and values characteristic of social relations on both sides, such as honour and shame, the male virility complex and patronage (Albera, 2006; Peristiany, 1966; Tolosana, 2001).

These aspects can also be found in the field of history of the family. Peter Laslett (1983) proposed four patterns of family and household formation in historical Europe, one of which was explicitly identified as Mediterranean. Subsequently, the testing of this “Mediterranean model” quickly became one of the objectives of family and marriage historians in Southern Europe and the Middle East (Sacchi & Viazzo, 2014). Historians of the family have tended to dismiss the possibility of commonalities between Mediterranean marriage and family patterns (Sacchi & Viazzo, 2014). On the other hand, some have hypothesised a long-standing regional contrast between Mediterranean Europe and the rest of the continent, and a very long-term continuity of the different shores of the Mediterranean (Smith, 1990).

The human and cultural unity of the Mediterranean has been widely debated since the 1960s. Anthropologists criticise, for example, the relevance of human unity as an anthropological mirage and a sum of stereotypes of northern researchers, especially of British anthropology (Herzfeld, 1987; Pina-Cabral, 1989). Some researchers reject the category of the Mediterranean (Horden & Purcell, 2000) and question the very existence of the object (Kayser, 1996).

At present, there is a growing consensus that rejects the vision of the Mediterranean as a culturally homogenous area (Huebner, 2016). The human unity of the Mediterranean no longer structures the frames of reference (Deprest, 2002). Nevertheless, it is possible to recognise a certain scientific coherence to the Mediterranean without denying its heterogeneity. Bromberger and Durand (2001) believe, for example, that it would be a mistake to present the Mediterranean as the ultimate place of mixing and blending. Exchange, migration and spatial proximity do not necessarily imply merging. However, without speaking of homogeneity, they recognise a “family resemblance”Footnote 2 between Mediterranean societies and cultures. This can be explained by the circulation of ideas, human migration, a similar ecological context and monotheism. Bromberger and Durand (2001) noted that the Mediterranean is defined more by its differences than by its similarities. The specificity of the Mediterranean space would be defined precisely by these differences, neither too close nor too far away. The Mediterranean would then constitute a stimulating study area for “understanding the whole range of possible relations with the other which is neither too close nor too far, neither too small nor too big”Footnote 3 (Bromberger and Durand, 2001), and would ultimately be an implicit tension between unity and diversity of its characteristics (Von Kondratowitz, 2013).

1.1.2 The Mediterranean and Population Sciences

This overview, which is deliberately not exhaustive, reveals the extent to which the Mediterranean area has been the subject of much research and intense debate since the nineteenth century. Thus, in several social sciences disciplines (geography, history, anthropology, etc.), the Mediterranean continue to constitute a relevant study area. A specific field of research has emerged (Mediterranean studies), and there are research structures dedicated to it, such as the Maison Méditerranéenne des Sciences de l'Homme (MMSH) of the University of Aix-Marseille in France.

However, we note that this liveliness of studies and debates on the Mediterranean is almost absent from the population sciences, the scientific field under which this publication falls. The Mediterranean does not seem to be a geographical frame of reference for population studies. The latter are largely concentrated on portions of the Mediterranean area of varying size, but much less so on the Mediterranean as a whole (on a pan-Mediterranean scale). The national scale remains the most commonly adopted, with many monographs on a particular Mediterranean country or sub-national regions. There are also regional studies which deal with a specific region (in the broad sense) of the Mediterranean, such as the Maghreb (Fargues, 1990; Le Bris, 2021; Ouadah-Bedidi et al., 2012; Sebti et al., 2009), the Near East (Courbage, 2008; Fargues, 1995), the Southern and Eastern shores of the Mediterranean (Courbage, 1997, 1999, 2015; Fargues, 2000a; Rashad, 2015; Tabutin & Schoumaker, 2005), the Balkans (Deslondes, 2004; Lerch, 2018; Sardon, 2001), and the countries of Southern Europe (Avdeev et al., 2011; Doignon et al., 2016; Pfirsch, 2011).

Thus, compared with the numerous national and macro-regional monographs, there are relatively few pan-Mediterranean demographic studies (Anastasiou et al., 2020; Attané & Courbage, 2004; Carella & Parant, 2016; D’Addato, 2010; Doignon, 2019, 2020; Doignon et al., 2016; Fargues, 2000b; Fargues and Salinari, 2011; ITAN, 2015; Matthijs et al., 2016; Oliveau and Doignon, 2014; Salvini, 2023; Troisi & Von Kondratowitz, 2013; Wilson, 2005; Zagaglia, 2013). This is probably due to the demographic contrasts across the Mediterranean, which many researchers in the population sciences consider to be too great to constitute a coherent study area. Indeed, the shores of the Mediterranean are generally presented in terms of contrast: a declining, low-fertility and ageing Northern shore facing a fast-growing, highly fertile and young Southern and Eastern shore. However, this dichotomy does not take into account the dynamic nature of demographic phenomena in terms of time and space. Mediterranean demography is in a state of flux, whether in terms of structures, dynamics or demographic behaviour: the rapid spread of declining mortality and fertility rates, the deceleration of population growth, the circulation of sociocultural or family models and ideas, etc. Consequently, the disparities that so well characterised the populations of the Mediterranean in the past are not necessarily of the same intensity today. By shedding light on the spatial dichotomy between the shores in light of recent demographic changes, it becomes possible to rethink the Mediterranean not just as a space of contrasts, but as an area in the process of homogenisation.Footnote 4

1.1.3 A Lack of Pan-Mediterranean Demographic Studies

In light of these various elements, we note that there is a lack of recent general publications, or writings in general, providing a synthesis or inventory of the various demographic phenomena on a pan-Mediterranean scale. For instance, the few existing demographic studies on the Mediterranean as a whole often focus on a particular demographic aspect. At this level, international migration remains perhaps the most studied demographic phenomenon at the scale of the Mediterranean as a whole, as the latter is conceptualised as a coherent migration system (Ambrosetti et al., 2016; Wihtol de Wenden, 2019).

From a certain perspective, this lack of analysis of the demography of Mediterranean populations as a whole is harmful. Population is the foundation of many other socio-economic phenomena. Studying the population allows us to understand the intimate aspects of a society, such as the fact of dying, giving birth, marrying, migrating, etc. Often considered as basic, these phenomena are nevertheless essential for a detailed understanding of contemporary societies and their evolution. This understanding would be more than welcome in the case of the Mediterranean, at a time when there are various initiatives for regional integration and cooperation in the Euro-Mediterranean region: Plan Bleu, the 5 + 5 Dialogue, the recent 2019 Summit of the Two Shores of the Mediterranean, the Union for the Mediterranean, the NATO Mediterranean Dialogue, the Centre for Mediterranean Integration, and the European Institute of the Mediterranean (IEMed). By updating our knowledge of population dynamics in the Mediterranean, we would be able to generate interest and fuel discussions in several fields, such as the social sciences, geopolitics, political sciences, etc. This is all the more important when demography is used to study international relations. The clash of civilisations (Huntington, 1993) for example conceptualises the youthfulness of Arab-Muslim societies (against the ageing societies of Europe) as a determining cause of a war between civilisations. Today's demography is thought to be timeless, making political unrest and conflict in the Mediterranean inevitable at the same time. The reduction of demographic contrasts is therefore a strong argument against the clash of civilisations and its conflicting predictions.

The aim of this publication is to provide an overview and detailed description of the demographic trends of the last 70 years for the populations of the Mediterranean as a whole. From this point of view, we are in line with the approach taken some 20 years ago by Attané and Courbage (2004) in their publication Demography in the Mediterranean region: situation and projections.

1.2 Approach and Commitment

1.2.1 What Do We Mean by the Mediterranean?

Firstly, we need to define the project's study area. The delimitation of the Mediterranean has been the subject of much debate, with no real consensus having been reached. First of all, it is pertinent to consider the Mediterranean Sea as the essential element of the Mediterranean area, especially as it is easy to circumscribe. On the other hand, the task becomes more complicated when it comes to delimiting the Mediterranean land: we know where to start, from the Mediterranean shore, but we do not know a priori where to stop. Thus, there are several delimitations of the Mediterranean.

The Mediterranean has a characteristic climate with the concomitance of the dry period and the hot period, and a very marked seasonal alternation. However, with a width of 3,800 km, there are great climatic differences, both in terms of rainfall and temperature. Indeed, the summer drought lasts between 2 and 3 months on the Northern Shore of the Western Mediterranean, compared to 6 to 7 months on the Eastern Shore (Clément, 2002).

Geographers of the nineteenth century choose to use the olive tree as the boundary of the Mediterranean. This tree is said to be a shared symbol in the Mediterranean as it is a gift from Athena to the Athenians, an idealised image of Abraham in the Old Testament, the tree that introduces the Light in the Qur'an, and a representation of justice and peace in many Mediterranean societies. However, the biogeographical and bioclimatic limits would be insufficient and too stringent (Roncayolo, 2002). They have the disadvantage of “revealing only one facet of the Mediterranean environment and would always be an imperfect approximation”Footnote 5 (Clément, 2002, 37).

Yves Lacoste (2001) proposed a geopolitical delimitation of the Mediterranean from Morocco to Turkmenistan and Iran, passing through the Caucasus region and the Gulf countries, and including the countries that are part of NATO (for example the United Kingdom).

Finally, other researchers, such as Robert Ilbert (2006), considered that there is no fixed limit between the Mediterranean and the non-Mediterranean. The Mediterranean would be characterised precisely by an absence of limits, since it would be a territorial space limited by edge and not by a limit.

In this publication, we will delimit the Mediterranean in the following way. A first simple geographical criterion is to choose the countries with direct access to the Mediterranean Sea. However, national borders changed during the twentieth century and problematic cases easily emerged, for example in the Balkans. At the end of the First World War, the Kingdom of Serbs, Croats and Slovenes, i.e. today's Croatia, Slovenia, Serbia, Macedonia and Bosnia-Herzegovina, was created and later called Yugoslavia. Until the period between 1991 and 1995, this country had access to the sea and could be considered Mediterranean. However, it is currently divided into several different countries. Thus, what are now Serbia, Kosovo and Macedonia would no longer belong to the Mediterranean region due to their lack of access to the sea. However, we believe that the Mediterranean character is not lost in the course of political change. Some countries do not have direct access to the sea, but are nevertheless extremely close: regions of Portugal, the West Bank, Jordan, Kosovo, Serbia, Macedonia and Bulgaria are often located less than 250 kms from the sea (Fig. 1.1).

Fig. 1.1
A contour map represents the distance to the Mediterranean Sea on a scale of 500 kilometers. The distance ranges from 0 to 1000 kilometers.

Distance to the Mediterranean Sea

Full size image

For example, France and Algeria are considered Mediterranean. However, regions in the north of France are as close to the Mediterranean Sea as some regions of the Caucasian countries, Russia, Poland and Ukraine. Without extending the scope so far, we chose to include countries with at least a portion of their territory within 250 kms of the Mediterranean Sea. We made an exception with Austria, Switzerland and Hungary, which are not oriented towards the Mediterranean. Thus, our study area consists of all countries with direct access to the sea, plus Portugal, Jordan, Kosovo, Serbia, Macedonia and Bulgaria. Andorra, Monaco and San Marino, located less than 250 km from the Mediterranean Sea, are not included in the study area as their populations are too small.

Mediterranean countries are commonly classified according to the shore to which they belong: the Southern Shore (African shore), the Eastern Shore (Asian shore) and the Northern Shore (European shore). We will use this categorisation for the representation of the data, to which we will add four regional clusters, an intermediate categorisation between countries and shores: North Africa, the Near East, the Balkans and Southern Europe (Table 1.1, Fig. 1.2).

Table 1.1 Composition of the geographical categories used in the publication
Full size table
Fig. 1.2
A location map of the Mediterranean region marks various countries, capital cities, main cities, sea or ocean, and regional clusters. The main clusters are northern Africa, the near east, the Balkans, and Southern Europe.

Location map of the Mediterranean

Full size image

1.2.2 International Comparisons Over Time

The general approach of this publication is to give a global vision of the demographic phenomena of the Mediterranean region. To this end, we have five commitments that we try to maintain throughout the different chapters.

First, we will not limit ourselves to describing the current demographic situation. Indeed, focusing only on recent developments would run the risk of losing the global perspective. Instead, we choose to place them in the long term, thus making it possible to put recent trends into perspective or, on the contrary, to underline their unprecedented character compared to past trends. Depending on the data available, our study will cover the period from 1950 to present day.

Secondly, we will carry out a global international comparison of the Mediterranean. Our unit of comparison is the country, not preconceived regional categories. We will not compare the Balkans with North Africa for example, but all Mediterranean countries with each other simultaneously.Footnote 6 The interest of this approach is to go beyond a vision of the Mediterranean in terms of regional blocks that oppose each other, just like the three shores. The Mediterranean is a maritime basin in which the shores move closer together and further apart (Brunet, 1995). Contrasts and disparities therefore change over time. By using existing regional categories, we would perpetuate differences that are perhaps less alive today. It is true that there is no longer a consensus on the human unity of the Mediterranean, but refusing to compare all the Mediterranean countries is tantamount to refusing to consider possible trans-Mediterranean formations and their dynamics (Deprest, 2002). Refusing a priori to compare would not allow the discovery of new differences and new regionalisations, or the emergence of unexpected scenarios. We believe that there is a need to update our analytical frameworks, particularly the regional categories, to ensure that they are currently consistent.

The third commitment is to describe the main demographic trends. It is clear that there are specific challenges for each country and region cluster. However, our aim is neither to analyse the specificity of each country nor to comment on short-term developments. On the contrary, we want to study the major movements that have characterised the demographic evolution in the Mediterranean over the last 70 years. Our aim is not to demonstrate that the Mediterranean is homogenous from a demographic point of view. Studying the Mediterranean countries as a whole will perhaps highlight the structural contrasts between the different shores. But it is also possible that it will show contrasts between unexpected groups of countries, thus renewing the demography of the Mediterranean. However, it is difficult to know until this global international comparison exercise has been carried out.

Moreover, we have chosen not to carry out any systematic forward-looking reflection, as the objective is above all to offer a synthetic vision of the major past and current demographic trends. However, we will look to the future and present forward-looking reflections where we consider it useful.

Finally, most of our analysis is done at the country level. Of course, we are aware that there are large disparities within countries, and that a sub-national analysis would reveal these. However, we have made this choice because we believe that there is already enough to comment on by staying at the level of the Mediterranean countries. Again, however, sub-national analyses will be offered where necessary to clarify a more specific point. In fact, comparison on a sub-national scale requires ad hoc data that are not available everywhere for all subjects and at administrative levels that can be harmonised on the Mediterranean scale.

1.2.3 Grid Convergence to Analyse Demographic Phenomena

To analyse the major trends of the Mediterranean countries as a whole in the long term,Footnote 7 we used the concept of convergence, i.e. the reduction of disparities and inequalities within a study area. We wanted to measure the evolution of disparities between Mediterranean countries. Have the gaps between these countries narrowed or increased? Are Mediterranean countries more similar today than in the past? However, lower disparities between countries do not imply an absence of disparities. Thus, observing convergence should not be interpreted as a deterministic process where all populations are destined to be perfectly similar in the short or long term. Converging trends should simply be interpreted as a reduction in disparities, not as a sign of continued convergence in the future.

The spread of the demographic transition in the world has led to a convergence of several demographic phenomena, such as birth rate and mortality (Wilson, 2001, 2011). It will come as no surprise that it is expected that there will be a convergence in the Mediterranean. However, our aim is not to demonstrate demographic convergence in the Mediterranean, but to analyse the extent to which this convergence is or is not taking place, and even to envisage cases of divergence. If there is convergence, our aim is then to analyse its intensity, pace, persistent disparities, etc.

Similarly, through the analysis of a possible convergence, our aim is not to prove any kind of demographic homogeneity (or unity) of the Mediterranean. We simply wish to highlight the converging and diverging areas, to provide elements for reflection on the fact that the Mediterranean is a more integrated demographic area today than in the past.

1.2.4 Data Used

In the absence of an integrated statistical system specific to the Mediterranean (Blöss-Widmer, 2019), we choose to use major international databases. Conventional demographic indicators are reliable for international comparisons as their definition is fairly widely shared. When analysing indicators such as birth rate, mortality or life expectancy at birth, we are not comparing incomparable things.

The main source of data for our publication is the World Population Prospects 2022Footnote 8 of the Population Division of the UN Department of Economic and Social Affairs. This database provides a wide range of demographic indicators for all countries around the world. Unlike a national statistical office, the UN does not collect data. It produces population estimates from existing data sources, such as surveys (national and international), population censuses, civil registration data, etc. A specific section of the World Population ProspectsFootnote 9 lists the sources used by the UN for each country. The quality data sources in some countries are scarce or incomplete, which makes demographic estimation even more difficult. In these countries, the data should therefore be interpreted with caution. The World Population Prospects thus consists of two parts: retrospective data since 1950, which includes UN population estimates, and population projections up to 2100. Moreover, UN estimates and projections have sometimes been criticised, in particular in relation to Mediterranean countries (Courbage, 1999). However, it should be stated that they have the undeniable advantage in providing good quality international comparative data over a long period of time, which is very useful when it comes to analysing major demographic trends, which is the objective of this publication.

In terms of territorial boundaries, the World Population Prospects considers current and UN-recognised states or territorial entities. The UN has therefore reconstructed the retrospective data so that it corresponds to the States (or territorial entities) in their current borders. This requires several clarifications in the case of the Mediterranean:

  • Western Sahara has been considered as a non-self-governing territory by the UN since 1963. For this reason, estimates for this territory are made from Moroccan data sources, such as the population census.

  • For the countries of the former Yugoslavia, the UN has reconstructed the demographic data of the Yugoslav period for each country within their current borders, using in particular the Yugoslav censuses which made it possible to distinguish the different territorial entities that made up Yugoslavia.

  • Kosovo refers to the United Nations administered region under security council resolution 1244. For statistical purposes, the data for Serbia do not include this area.

  • The Turkish Republic of Northern Cyprus is not recognised by the UN. Therefore, the data refer to the population of the whole island of Cyprus, without distinguishing between the two political entities.

  • The UN considers Palestine within its 1967 borders. Referred to as the “State of Palestine” in the data, Palestine thus comprises Gaza and the West Bank, including the Arab populations of East Jerusalem and excluding Israeli citizens residing in the occupied Palestinian territories. Population estimates are based, among other things, on data from the Palestinian Central Bureau of Statistics.

In this publication, for the sake of clarity, we use “Macedonia” for “North Macedonia”, “Palestine” for “State of Palestine”, and “Syria” for the “Syrian Arab Republic”. Similarly, for convenience, we use the term “country” to refer indiscriminately to the countries or territorial entities of the UN. These two choices are in no way an expression of opinion regarding the legal status of a country, its authorities or the delimitation of its borders.

When the World Population Prospects 2022 is not sufficient to study a subject, we use other international data providers to complement it. These are other UN databases (UNdata,Footnote 10 International Migrant Stock,Footnote 11 World Marriage Data,Footnote 12 etc.), and data from other international institutions (International Organization for Migration, WorldPop,Footnote 13 etc.), along with international surveys (DHS,Footnote 14 MICS,Footnote 15 etc.). The additional sources used are always indicated below the figures or in the text.

1.3 Structure of the Book

For a publication that aims to provide a global perspective of the population in the Mediterranean, we felt it was important to start by recalling how this population has been distributed in the area concerned. The second chapter therefore deals with settlement and urbanisation in the Mediterranean.

The third chapter focuses on the age and gender composition of the population, analysing the sex ratio of the population and the age pyramids of the Mediterranean sub-national regions.

The fourth chapter deals with demographic transitions, i.e. the evolution of birth and death rates. It proposes a typology of Mediterranean countries according to several characteristics of the process. It highlights the great diversity of demographic transitions in the Mediterranean.

Secondly, we wanted to take a closer look at each of the phenomena that make up the demographic transition. Thus, the fifth and sixth chapters analyse fertility and the formation and dissolution of unions respectively. The seventh chapter deals with mortality, with the evolution of life expectancy at birth, the decomposition of age-specific mortality, and an overview of the health transition.

The eighth chapter deals with international migration, where the Mediterranean is a migration system, which has undergone major changes since the 1950s.

Finally, the ninth chapter concludes the publication with an analysis of population dynamics, i.e. the evolution of the number of inhabitants in each country, and its components (natural and migration). It will also take UN projections to consider the future distribution of the population around the Mediterranean.

There will be a general summary conclusion, the main objective of which will be to take stock of the analysis of the demography of the Mediterranean countries as a whole and of the question of the convergence of this area over the last 70 years.