Save

Counting Couples: The Marriage Banns Registers of the City of Amsterdam, 1580–1810

Social and Economic History

In: Research Data Journal for the Humanities and Social Sciences
Authors:
René van WeerenDepartment of History and Art History, Utrecht University, Utrecht, Netherlands, r.vanweeren1@uu.nl

Search for other papers by René van Weeren in
Current site
Google Scholar
PubMed
Close
and
Tine De MoorRotterdam School of Management, Erasmus University, Rotterdam, Netherlands, demoor@rsm.nl

Search for other papers by Tine De Moor in
Current site
Google Scholar
PubMed
Close
Open Access

Abstract

Marriage is generally regarded as a decisive moment in the life course of individuals. As the social, but also the legal status of women and men changes as soon as they enter marriage and – by extension – their preceding wedding engagement, registers are and were being kept to record this life event in most societies. The difficulty in studying the long-term development of marriage patterns is the need for, among other things, detailed information about the marriage formation process. Most of the research on marriage patterns is based on a limited amount of data. Data either cover only a limited period (at most several consecutive decades), a limited number of variables, a relatively small number of marriages, and/or a relatively small town or region. The Amsterdam marriage banns registers are an exception to the above, in terms of content, focus area, and volume. In this article, we present the dataset results of the Citizen Science project ‘Ja, ik wil!’ [‘Yes, I do!’], involving over 500 participants retrieving a wide range of socio-economic data on over 94,000 couples from the rich source of the historical Amsterdam marriage banns registers, covering every fifth year between 1580 and 1810.

Abstract

Marriage is generally regarded as a decisive moment in the life course of individuals. As the social, but also the legal status of women and men changes as soon as they enter marriage and – by extension – their preceding wedding engagement, registers are and were being kept to record this life event in most societies. The difficulty in studying the long-term development of marriage patterns is the need for, among other things, detailed information about the marriage formation process. Most of the research on marriage patterns is based on a limited amount of data. Data either cover only a limited period (at most several consecutive decades), a limited number of variables, a relatively small number of marriages, and/or a relatively small town or region. The Amsterdam marriage banns registers are an exception to the above, in terms of content, focus area, and volume. In this article, we present the dataset results of the Citizen Science project ‘Ja, ik wil!’ [‘Yes, I do!’], involving over 500 participants retrieving a wide range of socio-economic data on over 94,000 couples from the rich source of the historical Amsterdam marriage banns registers, covering every fifth year between 1580 and 1810.

Online publication date: 7-7-2021

  1. Related data set “Ja, ik wil – Amsterdam marrage banns registrations 1580–1810” with doi www.doi.org/10.25397/eur.14049842 in repository “eur Data Repository”

1. Introduction

Marriage, including present-day (formal) partnership registration and same-sex marriage, has, throughout history, generally been regarded as a decisive moment in the life course of individuals, as is demonstrated by the recent multivolume series A Cultural History of Marriages (Behrend Martínez, 2019; Ferraro, 2019; Ferraro & Pedersen, 2019; Klaiber Hersch, 2019; Puschmann, 2019; Simmons, 2019).

Although the popularity of marriage in the EU has clearly been declining over the past two centuries, in the regions around the North Sea changes in marriage behavior already occurred much earlier. The extremely high average current (2018) ages at first marriage, for example, of 34.1 for men and 31.6 for women in the Netherlands (Centraal Bureau voor de Statistiek, 2019), are not novel, but the result of a very fundamental change in marriage behavior dating back to the late medieval period, already noted by Hajnal (1965) and coined back then as the European Marriage Pattern (emp). In the Netherlands, a specific subtype of marriage pattern, the Western European Pattern, can be discerned. This subtype is characterized by 1) higher ages at first marriage (especially for women) than in other parts of Europe and the world, 2) a comparatively small age gap between bride and groom for first-married couples, 3) a considerable number of men and women staying unmarried and 4) neolocality (the formation of an own household, independent of the couple’s parents) of the newlyweds, which was most pronounced in the North Sea area (Low Countries and England). De Moor & Van Zanden (2010b) linked the emp to the exceptionally strong socio-economic development of this area compared to other parts of Europe, although this assumption is not entirely undisputed (Dennison & Ogilvie, 2014). The empirical basis of research into marriage behavior in the early modern Low Countries can now be expanded through a dataset based on the vast collection of marriage banns registrations of the city of Amsterdam, from 1580 until 1810, just before the end of the French occupation.

2. Problem

One of the challenges when studying the long-term development of marriage patterns is the lack of detailed information about the marriage formation process. As the social, but also the legal status of women and men changes as soon as they enter marriage and—by extension—their preceding wedding engagement, registers have long been kept (and are still being kept) to record this life event in most societies.

For the Western Christian part of the world, the structural registration of marriages is based on the papal decree Tametsi in 1563, summoning local priests to keep records of all baptisms and marriages within their parish. However, Tametsi only stated that baptisms and marriages should be recorded in specific registers kept by the parish priests, without specifying exactly how to register these life events. Marriage registers, therefore, differ considerably in content, depending on the methods local clergymen and (later on) officials used across Europe. The simplest registers, for example, just contained the names of the engaged couples and markings indicating the unhindered completion of each of the mandatory three weekly engagement proclamations, the presence of a third mark indicating that no legal objections had prevented the marriage from being solemnized (see Figure 1). Also, as the keeping and preservation of these registers were mainly up to local clergy and officials, not all registers were preserved. Some were taken by priests when they left their parishes, while others were lost or destroyed over time, due to war, neglect, city fires, or natural disasters. German bombardments in May 1940, for example, destroyed 582 registers containing baptism, marriage, and burial registrations in a fire, causing the loss of about 50 percent of all of such records for the Dutch province of Zeeland (Van Waarden-Koets, 2010).

Figure 1
Figure 1

Marriage registrations registered in Amsterdam, January 1566. In the left column are the names of the grooms, in the right the names of their brides. The marks behind the names of the bride indicate the number of proclamations that had passed without any third-party objections being raised. Only couples for whom three unhindered proclamations had taken place were allowed to marry. The crosses marked in the space to the left of the names of the grooms indicate that the marriage was solemnized in an Amsterdam Catholic church, which was the only permitted religion in Amsterdam until May 1578.

Citation: Research Data Journal for the Humanities and Social Sciences 6, 1 (2021) ; 10.1163/24523666-06010002

amsterdam city archives, archive no. 5001, inv. no. 969, p. 8.

Before the Reformation, ecclesiastical rulings by the Catholic Church were the only legislation on marriage in the Western Christian part of the world (Donahue, 2016; Lettmaier, 2017; Van der Heijden, 2016). As soon as a region fell under Protestant rule, marriage was no longer merely the prerogative of the church, but also subject to civic law (Lettmaier, 2017; Van der Heijden, 2016; Witte, 2012). Protestants nonetheless retained some authority over marriage: an ecclesiastical marriage celebrated between Protestants in a Protestant church immediately also became a valid civic-law marriage. Non-Protestant marriages, however, could only be ecclesiastically celebrated if the marriage had been confirmed by the civic authorities, i.e., in the case of Amsterdam by the Commissioners of Marital Affairs (Commissarissen van Huwelijkse Zaken) (Stadsarchief Amsterdam, 1586). For the Northern Low Countries, this change took place in the final quarter of the 16th century, mainly between 1572 and 1576 (Israel, 1995). The city of Amsterdam remained faithful to Catholic and Spanish rule until rather late: only after a relatively peaceful coup d’état in May 1578 (commonly known as ‘Alteratie’) did Protestantism become the predominant and preferred religion within the city, subsequently also implementing Protestant rulings on marriage procedures (Israel, 1995).

Most of the historical research on marriage patterns is based on a fairly limited amount of data, whether limited in timespan (at most several consecutive decades), number of variables, number of marriages, and/or the relatively small size of a town or region. Although attempts have been made to compile larger datasets on early modern marriages (Dennison & Ogilvie, 2014; Szołtysek, 2016), the data collected on pre-1800 marriages are hardly usable for large-scale analysis because either their numbers are small and insufficiently representative (as Carmichael et al. [2016] demonstrated for the data collected by Dennis and Ogilvie [2014]), or because they focus primarily on rural areas (Szołtysek & Poniat, 2018, p. 477, Table 3). The extensive data collections from the Mosaic and napp dataset, for example, specifically lack data on the early modern Low Countries (Szołtysek & Poniat, 2018, p. 474, especially Figure 1), an area that is known for its exceptional economic development in the early modern period (De Moor & Van Zanden, 2006; Prak, 2012).

3. The Source

The Amsterdam marriage banns registers are a unique source among their kind, because of the combination of volume, specific geographical focus, timespan, and detail of the registered original data. As such, it is comparable in quality and representativeness to other historical demographic datasets, such as those from the Balsac project on the Quebec parishes and the China Multi-Generational Panel Dataset – Liaoning (Lee & Campbell, 2016). These datasets include a much larger number of individuals than those in the Amsterdam registers, but this is because their data were extracted from a large number of small, individual parishes within an extensive region (over 1.5 million square kilometers in Quebec and 145,900 square kilometers in China, respectively), whereas the Amsterdam dataset is based on a centralized system of registrations within a single city.

In terms of the total number of registrations, the Amsterdam marriage banns registers are most comparable to the Barcelona dataset of Five Centuries of Marriage (Cabré, n.d.). The latter dataset, however, was based on a collection of marriage registers of over 250 parishes, and the text of the registrations offered far less detailed data on registered individuals compared to the Amsterdam registers. In terms of specific information, the campop-dataset (Wrigley et al., 1997; Wrigley & Schofield, 1981) seems to be the one most similar to the Amsterdam registers, but also consists of a compilation of numerous parish registers instead of a single town; besides, the period covered within the mentioned datasets varies per included parish. Therefore, the registers of Amsterdam, covering an uninterrupted timespan of centralized marriage registration over 230 years for an entire city, and providing a wide array of details on over 900,000 individuals, can be regarded as unique (see Table 1).

T1

The earliest marriage registers date back to 1565 and were of the simplest form, as described earlier: two names and a mark for each completed and unhindered proclamation. Over time, these marriage registers changed little (see Figure 2).

Figure 2
Figure 2

Registrations of marriages confirmed before the Commissioner of Marital Affairs, 24 January and (partially) 30 January 1790. The composition of these registers changed very little in comparison to the register of 1566 as displayed in Figure 1. The main differences are that in the 1566 register, couples were registered as soon as they announced their intended marriage and the register also served to keep track of the number of proclamations issued. With the introduction of separate marriage banns registers, this became redundant and couples were from then onwards only recorded in the marriage registers once their marriage was formally confirmed in a Reformed church or before the Commissioners of Marital Affairs.

Citation: Research Data Journal for the Humanities and Social Sciences 6, 1 (2021) ; 10.1163/24523666-06010002

amsterdam city archives, archive no. 5001, inv. no. 961, p. 217.

Until the Reformation, legislation on marriage was the prerogative of the Catholic Church. Once an area became subject to Protestant rule, the marriage regulations laid down in the Politieke Ordonnantie of 1580 (Cau, 1655) came into effect. As a result of the Ordonnantie, marriage legislation became subject to worldly instead of ecclesiastical law, with the former’s extensive conditions and prohibitions for a legal marriage. This resulted in an increased need for the registration of individual details on the bride and groom, to ensure that a couple met the conditions to be married, often resulting in the introduction of separate marriage banns registers. In Amsterdam, this change took place in 1578. Within three months after the Protestant coup d’état of 26 May 1578 in Amsterdam’s city council (the so-called Alteratie), separate marriage banns registers were introduced. Just as under prior Catholic rule, each couple had to register their marriage intention at least three Sundays before the actual wedding could take place (Cau, 1655). This waiting period of a minimum of fourteen days (the shortest span of time comprising three consecutive Sundays) and the weekly proclamations were designed to inform fellow citizens about the intended marriages, to allow anyone who might have legal objections to bring those to the attention of the local authorities. To preclude couples eloping, these proclamations also had to be made in previous towns of residence (see, e.g., Haks, 1985, especially from page 125).

In Amsterdam, three types of registers were used, because of the difference in legal implications between Reformed and other marriages:

  1. A)Marriage banns registers of the church (kerkregisters; Stadsarchief Amsterdam, archive 5001, inv. nos. 401–660), containing marriage banns between two Reformed partners. Marriages between marriage partners celebrated in a Reformed Amsterdam church were immediately also valid legal marriages. Banns for this type of marriage were proclaimed in the Reformed churches of Amsterdam.
  2. B)Marriage banns registers of the civic authorities (puiregisters; Stadsarchief Amsterdam, archive 5001, inv. nos. 661–761), containing marriage banns of couples of whom at least one was not Reformed. These weddings were performed by the local magistrates (in Amsterdam by the Commissioners of Marital Affairs [Commissarissen van Huwelijkse Zaken], aldermen specifically appointed to these affairs, and were placarded weekly at the facade [pui] of the town hall, hence the nickname ‘puiregisters’). For non-Reformed marriages, only the civic marriage had legal status.
  3. C)Marriage banns registers containing registrations for weddings that would be performed elsewhere (extraordinaris registers; Stadsarchief Amsterdam, archive 5001, inv. nos. 762–765). Probably because of Amsterdam’s strong population growth in the course of the seventeenth century, as of 1636 these separate registers were abandoned, and thereafter such registrations were registered in the regular marriage banns registers mentioned under A and B. The need for proclamations elsewhere and the marriage location were from then on indicated above the concerned registration.

All marriage banns registers were kept by the Commissioners of Marital Affairs and their secretaries. The registers of the church start in 1578 and run until the introduction of the Civic Registration (Burgerlijke Stand) in Amsterdam in the course of 1811. The registers of the magistrates date back to 1581 and run until the start of the French occupation in 1795. It should be noted, however, that because of the égalité principle of the French Revolution, a decree was issued that from 1 June 1795 onwards, all marriages, regardless of the religious denomination of bride and groom, had to be confirmed before the local magistrates to become valid, hence annulling the former privileged position of Reformed marriage (Huussen, 1975). As a consequence, all marriages between 1 June 1795 and 28 February 1811 were registered in the same register (for which the former church registers were used). From 1 March 1811 onwards, marriages were registered in the dedicated registers of the Civic Registers (Burgerlijke Stand).

4. Data Collection Methods

Although the great potential value of the Amsterdam marriage banns registers has been recognized by other researchers in the past, their datasets comprised only a limited selection of the source (Hart, 1955a, 1955b, 1975; Kuijpers, 2005), which left out the many opportunities this source offers for long-term analysis of marriage patterns in one of the most vibrant cities of historical Europe. The Hart collection also has not been digitized, making it less suitable for further analysis. As part of the vidi-project ‘Nature or nature? A search for the institutional and biological determinants of life expectancy in Europe during the early modern period’, data from all marriage banns registrations of every fifth year from 1580 to 1810 of all Amsterdam marriage banns registers (church, municipality, and extraordinaris registers) were digitized with the help of a large group of volunteers. For this purpose, the Citizen Science project called ‘Ja, ik wil!’ [transl.: ‘Yes, I do!’] was set up to manage the data collection procedure, via the already existing Dutch crowdsourcing platform ‘Vele Handen’ (https://www.velehanden.nl). This platform offers multiple (primarily) historical projects for which citizens can volunteer to contribute, and allowed for the rapid recruitment of a total of approximately 500 skilled and enthusiastic volunteer transcribers, many of whom were loyal participants throughout the two-year data entry period (for an extensive analysis of this project, including a description on how participants were supported by a project forum, participant meetings, and paleography education, and an analysis of the success factors of the project, see De Moor et al., 2019). In the design phase before the project’s start, it was also decided to exclude the registrations between 1580 and 1600 from data entry via the Vele Handen platform because of the technical complications created by the structure of the original registers, and to limit data entry to registers of every fifth year between 1580 and 1810 to make it possible to complete this endeavor within the project’s duration. As banns registers were only ‘standardized’ from 1605 onward (by printing the standard fixed-text formulas, hence automatically limiting the maximum number of registrations to the three partially pre-printed registrations per page, see Figure 3), the same application design could be used for all registrations between 1605 and 1810 (that is, about 97 percent of all registrations). Taking a sample of every fifth year has its limitations. However, as our research questions concern the entire early modern period, we find the five-year grid sufficiently fine-grained to analyze developments over several decades or even centuries, as well as to analyze the overall effects of disasters like the plague and its aftermath.

Figure 3
Figure 3

Register of the marriage banns register of the church, May 5–8, 1604. This page marks the introduction of printed text in the marriage banns registers. The printed text concerns the text that applied to every registration and referred to the couple’s request to be registered, to have their marriage intention proclaimed on the subsequent Sundays, the declarations of both bride and groom that they were not bound to any other partner, were not blood-related to a prohibited degree, and that there were no other legal issues that might prevent their intended marriage. As a result of using this pre-printed text from May 1604 onward, each page contained a maximum of three registrations, whereas in the pre-1604, fully handwritten registrations, a page could contain as many as six registrations.

Citation: Research Data Journal for the Humanities and Social Sciences 6, 1 (2021) ; 10.1163/24523666-06010002

amsterdam city archives, archive no. 5001, inv. no. 411, pp. 216–17.

The data in the registrations from 1580–1605 were ultimately distilled by a selected group of expert-volunteers directly from the sources, subsequently transformed into the same data format, and then combined with the data collected through the Vele Handen platform in the final dataset. Although the introduction of pre-printed pages standardized the registers’ layout, the content of the handwritten parts of the registrations differed in included details over time (see Appendix, Figures A1A6).

Using the standard methodology of the Vele Handen platform resulted in a high-quality dataset: each scan was entered by two separate volunteers. A third person with a higher level of expertise compared the results of the two entries—supported with software to enhance the efficiency of this process—and made the final decision regarding the correct transcription of the registers. To facilitate efficient data entry, the data fields on the data entry forms (and subsequently the instructions for the volunteers) generally followed the same order as in the source. The content of the registrations generally remained uniform throughout the whole period, with only two exceptions. The first exception was that, although no specific reason has been found, the structural registration of grooms’ occupations ceased in 1715. The other major change involved the structural registration of the religious denominations of both groom and bride between 1755 and 1795, which was related to the implementation of legislation aimed at discouraging marriages between Reformed and Catholic partners (see also Haks, 1985) and was ended in 1795 as a result of the French occupation of the Low Countries. A full overview of the development of the registers can be found at http://www.collective-action.info/Ja-ik-wil/Ondertrouwregisters-ontwikkeling.

5. Data

  1. “Ja, ik wil – Amsterdam marrage banns registrations 1580–1810” deposited at eur Data Repository – doi:www.doi.org/10.25397/eur.14049842
  2. Temporal coverage: 1580–1810

The dataset contains a total of 94,302 marriage banns registrations in the sample years between 1580 and 1810, hence containing data on about 188,000 individuals (in a small number of cases, the registration was incomplete, usually lacking data on the bride). The checked data entry from the ‘Yes, I do!’ project forms the basis of the dataset; an overview of the data entered is displayed in Table 2. The decision to end with 1810 as the final sample year is directly related to the introduction of the Civic Registration in 1811. This 1811 transition was not only an administrative change, but also confirmed the change in meaning of marriage from 1795 onwards: ecclesiastical marriages, including Reformed, lost their legal status, and the moment of marriage became more important than the moment of engagement (Huussen, 1975). Consequently, the marriage acts (instead of the marriage banns on which our dataset is based) became the most important documents in marriage procedures. Given these changes in the types of sources and meaning attached to them, our dataset can thus not straightforwardly be connected with post-1810 data such as those from the links-project (International Institute of Social History, n.d.).

T2

In addition to the data entered by volunteers, the dataset also contains three other types of data generated in other ways:

  1. a)Metadata generated by the platform used to identify each scan, each registration, and each checked data entry record (Columns B-D).
  2. b)Data from a preexisting digital name index on grooms, brides, and any possible previous spouses, provided by Amsterdam City Archives [Stadsarchief Amsterdam] (Columns co, cs-cv).
  3. c)Data added by standardizing or processing the entered data, required to make the data suitable for statistical analysis (see Table 3), performed by researchers involved in the project

T3

In the codebook (https://www.doi.org/10.25397/eur.14049842), the origin of the data in each column has been indicated.

After combining the data collected through the Vele Handen platform with the pre-1605 data collected separately, a final check of the project was performed looking for any remaining inconsistencies, solving some issues related to scans that were hard to read, and standardizing part of the data columns before depositing the dataset and related documents into the repository.

Although most registrations concern unique registrations, a small number concern duplicate registrations, attributable to three main reasons:

  1. 1)The marriage was called off beforehand, because of legal objections (e.g., proven bigamy) or failure to meet legal requirements (e.g., lacking parental consent). In such cases, the original registration could have been canceled, with the couple later re-registering once the hindrances had been overcome or solved.
  2. 2)The couple that had registered did not appear and was declared as ‘not serviced’ [‘niet geholpen’]. These couples sometimes later re-registered and completed the procedures.
  3. 3)In the case of marriages between a Reformed and a non-Reformed partner, the marriage banns were registered in both the church and municipal banns registers.

Cases as described under 1) and 2) are recognizable within the source documents, as these registrations were usually crossed out and/or explicitly indicated as canceled or not serviced (see Figure 4). Following data entry instructions, these entries have been coded by participants as canceled in the data field [Geannuleerd]; in this way, canceled registrations can be easily filtered from the dataset. Duplicate marriage banns registrations belonging to the third category are difficult to distill, as these registrations were not consistently indicated in the source as being registered in two separate registers at the same time. As the number of such registrations was relatively small and the selection of these duplicate registrations needed to be done manually, these data have not been corrected in the dataset.

Figure 4
Figure 4

Cancelled marriage banns registration, Amsterdam, 6 January 1775. This couple intended to marry in Sloterdijk, a small village located at that time just outside Amsterdam, and had already registered their marriage there. As the groom lived in Maarssen (near Utrecht) and the bride in Amsterdam, proclamations had to be made in Amsterdam, Sloterdijk, and Maarssen. From the note in the margin, it appears that both partners decided not to marry in the end and to release each other from the mutual promises to marry.

Citation: Research Data Journal for the Humanities and Social Sciences 6, 1 (2021) ; 10.1163/24523666-06010002

amsterdam city archives, archive no. 5001, inv. no. 620, p. 12.

Volunteers were instructed to enter data from the sources as-is, to avoid (mis)interpretation. The basic output generated by the Vele Handen-platform export module consisted of records combining a) metadata related to the input system, b) data from the indexes already made available through Amsterdam City Archives, and c) the data entered as-is by the volunteers. In order to analyze the entered data, the data needed to be standardized (see Table 3). For fields that were mainly numerical in nature (like ages), the standardization was performed by filtering out fields with incomplete data (due to illegible or incomplete ages in the source registration). For the three text fields (places of origin groom and bride and occupation groom), a two-step process was needed: the original entry was first standardized to the present-day equivalent, and then the present-day equivalent was coded using available reference documents or coding systems (see Table 3).

6. Data Usability and New Possibilities

The detailed registration of historical demographic data on individuals on such a scale as the data sample collected in this dataset not only offers researchers the opportunity to distill aggregated data or data on specific individuals (as far as their surname allows actual identification) but also to perform detailed historical long-term analyses, based on a large number of variables, for example, research on specific groups (religious, occupational). By its nature, size, and continuity, covering a time span of 230 years, the Ja, ik wil – Amsterdam marriage banns registers 1580–1810 dataset is a valuable research resource in academic discourse on marriage patterns, mainly initiated by Hajnal’s (1965) well-known seminal article on the European Marriage Pattern (emp). As Amsterdam belongs to the core area of the emp, the dataset offers substantial empirical data to test Hajnal’s original hypothesis as well as those of other scholars in this discourse, as the emp theory is not undisputed (Dennison & Ogilvie, 2014). Also, the dataset may offer valuable information in other discussions as well, for example, on the role of marriage patterns in economic development and the Great Divergence debate (De Moor & Van Zanden, 2010b; Foreman-Peck, 2011; Greif, 2006; Pomeranz, 2000; Voigtländer & Voth, 2013) or the influence of marriage patterns and household formation on factor markets (Puschmann and Solli, 2014; Szołtysek et al., 2011; Van Bavel et al., 2009; Van Zanden et al., 2019). Given the nature of Amsterdam as a migrant society (Kuijpers, 2005), the dataset may also shed light not only on the effects of migration to emp-areas on the marriage patterns of groups of migrants (e.g., the Sephardic and Ashkenazi Jewish communities, mid-European seasonal laborers), but on the influence of those groups on marriage patterns within an emp core environment as well. Existing datasets within the emp core area of this size are scarce or based on very small samples (e.g., Dennison & Ogilvie, 2014; see also Carmichael et al., 2016), with only the campop-database offering data for measuring the effects of marriage patterns and household formation in the emp core area (e.g., Wrigley, 2014).

As mentioned before, other researchers have used these marriage banns registers in the past for historical analysis, but only a very limited selection of them. The detailed dataset that is now available will allow researchers to perform detailed analyses, especially if these data are combined with other data containing entries on the level of a person or an aggregate location. Looking at the data summary, the most effective expansion would be to match the data from the places of origin with the data already collected by Hart (n.d.), combined with the iso 3166 coding (International Organization for Standardization, n.d.), which would result in a fine-grained overview of migration to Amsterdam. Another option for large-scale standardization would be the address fields in the current dataset (for the groom, parents groom, bride, parents bride), although the fact that specific addresses were only commonly used for the data from the end of the eighteenth and the beginning of the nineteenth century limits the possibilities.

Despite the unique combination of historical data this dataset offers, research options using this dataset are of course not without limitations. A first limitation is caused by the nature of the source itself: as the registers served to register the banns of intended marriages, the dataset by definition contains no information on Amsterdam citizens who never got engaged to be married. The substantial part of the population that remained single, both among men and women, can thus not be captured with this dataset (De Moor, 2014; Devos et al., 2015, 2016). Also, the registers contain information on individuals for only one (or more, if the individual remarried) static moment(s) in their lifetime, that is; their details at the moment they registered their marriage banns. The dataset therefore cannot be used in itself to study the social or economic life stories of individuals.

As registration was performed and recorded on an individual basis, the dataset in itself is also not sufficient to study intergenerational developments within a specific group, as the dataset does not provide a way to connect consecutive generations of a specific family; to achieve this, one would need to perform additional genealogical research or use a form of family/population reconstruction. However, as marriage procedures were compulsory and equal for all citizens, regardless of their status, the registers are far more inclusive than, for example, tax registers as the basis for social-demographic reconstructions, as the latter only included those who had sufficient means to be taxed, excluding those below the taxing threshold. Although in 1795 a citywide census was held, the officers performing the counting already admitted that the census would appear to be incomplete:

The population density within the Jewish part of town is at some locations that high, with every possible room, even in the attics, occupied by a large number of inhabitants, the immodesty of the number of people inhabiting these houses being so big, that the district supervisors are not able to guarantee that all persons, especially children, have been accounted for.

farret et al., 1795, p. 4

Also, this census was performed only once at the end of the period of the marriage banns registers dealt with in this article, hence providing population data for only that specific year.

A challenge that every researcher faces when using data transcription from historical sources is the possibility of faulty transcriptions, causing differences between the actual data in the source and the data finally entered in the dataset. Although this risk can never be excluded entirely, using the above-described ‘three pairs of eyes’ methodology for data entry minimized this risk. Nonetheless, it should be taken into account that the data in the source itself should be analyzed and interpreted with some precautions: the recorded data were mainly provided by the persons registering themselves. Ages were registered in the database as mentioned in the source, often without being checked against other potential sources, such as baptism records for the brides and grooms. Ages given by the bride and/or groom could therefore sometimes differ several years from their actual ages. This difference could either be unintentional, as not everybody was equally experienced with handling figures (De Moor & Van Zanden, 2008, 2010a), or intentional, to avoid the need for parental consent, or to artificially decrease the age gap between the groom and bride. A small age difference (preferably with the groom being the elder of the two) was regarded as favorable for a good Christian marriage (e.g., Wittewrongel, 1661, pp. 45–46), while a large one was viewed as making the marriage less likely to succeed, attributed, among other things, to an assumed reduction of the couple’s mental flexibility in making compromises (De Vos, 1771, pp. 269–276, especially p. 273).

A potential cause of unintentional differences between actual ages and ages reported in the marriage banns could be ‘age heaping’, which entails a significant statistical overrepresentation of ages ending on zero or 5, indicating that people reporting their ages tended to round up to the closest multiple of five, The Whipple index calculated based on these statistics forms an indication of the numeracy of the population concerned (e.g., A’Hearn et al., 2009; De Moor & Zuijderduijn, 2011; see also De Moor & Van Zanden, 2008, 2010a). Earlier analysis on age heaping, however, based on four sample years originating from the same archival source as the current dataset, demonstrated that the Whipple index for those years for Amsterdam brides and grooms was considerably lower than for populations from comparable archival sources elsewhere (De Moor & Van Zanden, 2008). If we calculate the Whipple index for the entire dataset, the figures concur with those earlier findings. The index figures of just below 119 for men and just over 116 for women indicate that no significant age heaping seems to have happened. Although the Whipple index is generally accepted and used as an adequate indicator of numeracy within a population (e.g., A’Hearn et al., 2009; Szołtysek et al., 2018), it does not rule out faulty age reporting by individuals completely. Additional comparison of reported ages with baptism records of the marrying individuals (for at least a representative number of sample years) might shed more light on (average) discrepancies between reported and actual ages. Such a check would, however, require an additional research project, and far exceeded the possibilities and scope of the Ja, ik wil! project.

Also, it should be taken into account that ages are not available for all individuals, due to the original purpose of age registration in this source. The Amsterdam banns registers, administered by the Commissioners of Marital Affairs and their staff, served first and foremost as a ‘checklist’ for the magistrates to determine whether or not all conditions were met to perform the final wedding ceremony. Age was therefore registered merely to check whether the bride- and groom-to-be were of proper age and whether or not parental consent (if under 25) or consent from next of kin (if under 30) was required. If widows or widowers remarried, however, registration of their ages was irrelevant, as they had already been of age when they married their previous partner. Therefore, ages for remarrying widows and widowers are mostly lacking in this source and, consequently, in the dataset. Nonetheless, except for the first two sample years, ages are available for at least 60 percent of all brides and grooms, and figures for some years even exceed 80 percent (see Table 4). The sudden percentual drop for the year 1665 can be explained by consecutive plague pandemics causing a sudden increase in the number of remarrying widows and widowers (see Figures 5a and 5b). For those who were remarrying without an age mentioned, it would in some cases be possible to deduct their actual age at remarriage from the age mentioned at a previous marriage. The age of Harmanus van Weeren, for example, who married four times during his lifetime, was not mentioned for his last three marriages but could be deducted from the banns registration of his first marriage, for which on 2 June 1735 it was recorded that the groom-to-be was 35. In the current dataset, however, such deductions have not been made.

T4
Figure 5a
Figure 5a

Number of grooms-to-be per sample year with indication of their existing marital status (red = non-widower, blue = widower)

Citation: Research Data Journal for the Humanities and Social Sciences 6, 1 (2021) ; 10.1163/24523666-06010002

dataset ja, ik wil – amsterdam marriage banns registers 1580–1810
Figure 5b
Figure 5b

Number of brides-to-be per sample year with indication of their existing marital status (red = non-widow, blue = widow)

Citation: Research Data Journal for the Humanities and Social Sciences 6, 1 (2021) ; 10.1163/24523666-06010002

dataset ja, ik wil – amsterdam marriage banns registers 1580–1810

A similar precaution should be taken into account when using this dataset for research on migration. In the case of brides and grooms who had previously lived outside Amsterdam, the previous place of residence was mentioned. In the case of those already residing in Amsterdam for a considerable number of years prior to their registration, either no previous place of residence was mentioned or they were registered as ‘van Amsterdam’ [‘from Amsterdam’]. As a result, places of origin are indicated for each individual, either explicitly or implicitly (in the case of Amsterdam-resident marriage partners where the absence of a place of origin indicated that they were citizens of Amsterdam). Nonetheless, some issues regarding the use of this source for migration studies require specific attention.

Analysis of the places of origin mentioned shows that many of the individuals recorded in the source, and subsequently in the dataset, lived outside Amsterdam before they married; Amsterdam, therefore, can be characterized as a ‘city of migrants’ (Kuijpers, 2005; Van Weeren & De Moor, 2019). Kuijpers (2005) and Lesger (2006) state that the places mentioned in the banns registers are in most cases the birthplaces of the grooms and brides involved. However, the source both Kuijpers and Lesger refer to (Van Dillen, 1929, p. xxxiv, Table iii, fn. 1), does not reference any supporting literature when stating that the places recorded in the banns registers “apparently are the birthplaces rather than the places of origin”. As the marriage procedures prescribed engagement announcements in the places of residence of both groom and bride, it is more likely that the places mentioned are the places of origin (which may be but need not be by definition, the same as the places of birth) of the couple. The registration of the place of origin in the banns registers served a specific purpose: it was meant to show that the marriage partner had lived outside of Amsterdam and therefore might need to provide adequate proof that the banns had also been announced unhindered in the previous place of residence, to prevent couples eloping from other cities to avoid objections from third parties to their intended marriage (see on this, e.g., Haks, 1985). Consequently, brides and grooms that had already left their place of birth in the past and had moved to other towns before moving to Amsterdam did not have their place of birth registered, but their most recent place of residence instead.

Another issue regarding the places of origin limiting the potential for analysis is the identification of the locations mentioned. Although the majority could be identified with a unique present-day location (in some cases using the impressive ‘Herkomstregister’, composed by Simon Hart and his team in the 1970s [Hart, n.d.]), in some cases they could refer to several possible locations. Without further details, it seemed impossible to pinpoint the exact place of origin (e.g., a reference to ‘Vianen’ could refer to four separate locations: Vianen just south of the city of Utrecht, Vianen in the province of North Brabant, Vianen in the province of Zeeland, or Vianen located in present-day Flanders [Belgium]). In such cases, the value in Column K of the dataset was for now set to ‘TBA’. In addition to this, some locations could not ultimately be identified with any present-day location: this may be caused by either spelling mistakes (as locations were recorded phonetically) or by the mere fact that names of some locations have disappeared into oblivion in the course of history. These unidentifiable locations are indicated in Column K of the dataset also with the value ‘tba’.

For other aspects, the data recorded may limit research options. Examples are the occupations of grooms, which were only registered in the marriage banns registers until c. 1710 and are therefore almost entirely lacking for the period 1715–1810. Although the reason for this change is unclear, it may be that in the pre-1710 registers the occupation was registered to serve as an additional means of differentiating a person from his namesakes, but that the explosive growth of the Amsterdam population over the 17th century (Amsterdam’s ‘Golden Age’) caused the registration to lose its value for identifying the groom (brides’ occupations were not registered, as their work was often not regarded formally as an occupation) and hence their registration in the marriage banns registrations was abandoned around 1710.

Also, research involving the religious denomination of the registrants is only possible for the period 1755–1795. Structural registration of the religious denominations individuals belonged to only came into effect in 1755. This was directly related to measures taken by the Reformed government in the Low Countries to prevent mixed marriages (i.e., between a Reformed and a non-Reformed partner). One of the main effects on marriage procedures was an extended waiting period for engaged couples between the consecutive mandatory three announcements of their intended marriages: instead of announcements being proclaimed on three consecutive Sundays, announcements of ‘mixed’ marriages had to be announced with an interval of six weeks between the respective announcements. In addition, the couple also had to wait for another six weeks after the third and final announcement to be allowed to finally have their formal wedding day (Allart, 1795). By registering the religious denominations, the Commissioners of Marital Affairs and their staff could check whether the appropriate waiting period had been respected for each couple. This registration ended soon after the French occupied the Low Countries in 1795. As a result of the égalité-principle of the French Revolution, a governmental decree by the Staten Generaal published on 30 April 1795 stated that as of 7 May 1795 all marriages needed to be completed before the proper civic authorities to become legally valid marriages. Ecclesiastical Reformed marriages consequently lost their validity for civic law (Huussen, 1975).

The data collected in this dataset has been used for several academic papers by members and affiliated researchers of our research team (De Moor et al., 2020; Störmer et al., 2017), a book on marriage in early modern Amsterdam directed at a broader, non-academic audience (Van Weeren & De Moor, 2019) and a study of the long-term development of marriage patterns worldwide and its implications for the position of women (Van Zanden et al., 2019). As a side-effect of this data collection, research on the learning curve of participants in this project as well as other projects on the Vele Handen platform was performed, resulting in a publication on the dynamics of Citizen Science (De Moor et al., 2019).

7. Concluding Remarks

Given the number of records, the time span covered, the relatively fine-grained interval of five years between each of the sample years, the detailed per issue registration, and the fact that data have been collected per engaged couple rather than in an aggregate way, we believe the dataset in its current version offers fellow-researchers a firm basis for addressing a wide variety of demographic research on early modern urban society in Western Europe, enabling them to combine several specific selection criteria within the data to narrow their search, but also to combine data from this dataset with their dataset to complete or expand their dataset and research options.

Thanks to the detailed registration, the dataset will allow detailed research on such topics as migration and marriage patterns in early modern Amsterdam and the relation between these data and the remarkable development of Amsterdam and the Low Countries over this period. As such, it can offer the necessary empirical basis for debates on, for example, the role of household-formation-related changes in economic development.

Acknowledgements

The authors would like to thank all participants in the Ja, ik wil! project for their contribution to the project and the data collection for this database. We also would like to thank Ellen Fleurbaay and Hans Visser of the Amsterdam City Archives for providing index data and to the people of Picturae, who provided the technical infrastructure and support for the Vele Handen platform. Finally, we would like to thank both anonymous peer reviewers for their valuable suggestions for improving the initial draft of this article.

The data collection was part of the vidi-project ‘Nature or nature? A search for the institutional and biological determinants of life expectancy in Europe during the early modern period’, funded by the Dutch Organization for Scientific Research (nwo) (vidi – de Moor – 276-53-008).

References

Archival Sources

  • Allart, J. A. (1795). Groot Placat boek, Inhoudende de placcaten ende Ordonnantien Van de Hoog Mog. Heeren Staaten Generaal der Verenigde Nederlanden ….[etc], 8. National Archives, library signature H 21 E-1, pp. 543544.

    • Search Google Scholar
    • Export Citation
  • Cau, C. (1655). Groot-placaetboek, vervattende de placaten, ordonnantien ende edicten van de Staten Generaal der Verenigde Nederlanden ….[etc], 1, National Archives, library signature H 21 D-1, columns 329343.

    • Search Google Scholar
    • Export Citation
  • Stadsarchief Amsterdam. (1586). Memorieboek van Commissarissen van huwelijkse zaken en injuriën 1586-2 November 1759. Archieven van de Schout en Schepenen, van de Schepenen en van de Subalterne Rechtbanken (no. 5061), inv. no. 3051.

Publications

  • A’Hearn, B., Baten, J., & Crayen, D. (2009). Quantifying quantitative literacy: Age heaping and the history of human capital. The Journal of Economic History, 69(3), 783808. https://www.doi.org/10.1017/S0022050709001120.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Behrend-Martínez, E. (Ed.). (2019). A cultural history of marriage in the age of enlightenment (1650–1800). A cultural history of marriage series, Vol. 4. Bloomsbury.

    • Search Google Scholar
    • Export Citation
  • Cabré, A. (n.d.) Five centuries of marriages. Retrieved 29/09/2020 from http://dag.cvc.uab.es/infoesposalles/.

  • Carmichael, S. G., de Pleijt A., Van Zanden, J. L., & De Moor, T. (2016). The European Marriage Pattern and its measurement. The Journal of Economic History,76(1), 196204. https://www.doi.org/10.1017/S0022050716000474.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Centraal Bureau voor de Statistiek (2019). Bevolking: geslacht, leeftijd en burgerlijke staat, 1 januari. Retrieved 26/03/2020 from https://opendata.cbs.nl/statline/#/CBS/nl/dataset/7461BEV/table?fromstatweb.

    • Search Google Scholar
    • Export Citation
  • De Moor, T. (2014). Single, safe, and sorry? Explaining the early modern beguine movement in the Low Countries. Journal of Family History, 39(1), 321. https://www.doi.org/10.1177%2F0363199013515167.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • De Moor, T., Boele, A., Carmichael, S., & Farjam, M. (2020). A match made in Amsterdam: An examination of the determinants of marriage patterns. Submitted article.

    • Search Google Scholar
    • Export Citation
  • De Moor, T., Rijpma, A., & Prats López, M. (2019). Dynamics of engagement in citizen science: Results from the “Yes, I do!”-project. Citizen Science: Theory and Practice, 4(1), 117. http://www.doi.org/10.5334/cstp.212.

    • Search Google Scholar
    • Export Citation
  • De Moor, T., & Van Zanden, J. L. (2006). Vrouwen en de geboorte van het kapitalisme in West-Europa. Boom.

  • De Moor, T., & Van Zanden, J. L. (2008). Van fouten kan je leren: een kritische benadering van de mogelijkheden van “leeftijdstapelen” voor sociaal-economisch onderzoek naar gecijferdheid in het pre-industriële Vlaanderen en Nederland. Tijdschrift voor Sociale en Economische Geschiedenis, 5(4), 5588. http://www.doi.org/10.18352/tseg.537.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • De Moor, T., & Van Zanden, J. L. (2010a). “Every woman counts”: A gender-analysis of numeracy in the Low Countries during the early modern period. Journal of Interdisciplinary History, 41(2), 179208. https://www.doi.org/10.1162/JINH_a_00049.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • De Moor, T., & Van Zanden, J. L. (2010b). Girl power: the European Marriage Pattern and labour markets in the North Sea region in the late medieval and early modern period. Economic History Review, 63(1), 133. https://www.doi.org/10.1111/j.1468-0289.2009.00483.x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • De Moor, T., & Zuijderduijn, J. (2011). The art of counting: Reconstructing numeracy of the middle and upper classes on the basis of portraits in the early modern Low Countries. Historical Methods, 46(1), 4156. https://www.doi.org/10.1080/01615440.2012.706795.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • De Vos, W. (1771). Verhandeling over het Huwelyk, …. [Genootschap onder de zinspreuk Floreant liberales artes].

  • Dennison, T., & Ogilvie, S. (2014). Does the European Marriage Pattern explain economic growth? The Journal of Economic History, 74(3), 651693. https://www.doi.org/10.1017/S0022050714000564.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Devos, I., Schmidt, A., & De Groot, J. (Eds.). (2015). Single life and the city 1200–1900. Palgrave Macmillan.

  • Devos, I., Schmidt, A., & De Groot, J. (2016). Introduction. Unmarried and unknown: Urban men and women in the Low Countries since the early modern period. Journal of Urban History, 42(1), 320. https://www.doi.org/10.1177/0096144215610770.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Donahue, C., Jr. (2016). The legal background: European marriage law from the sixteenth to the nineteenth century. In S. Seidel Menchi, E. Eisenach, & C. Donahue (Eds.). Marriage in Europe, 1400–1800 (pp. 3360). University of Toronto Press.

    • Search Google Scholar
    • Export Citation
  • Farret, J. P., Verster, A. G., & Van Swinden, J. H. (1795). Rapport over de telling van het volk van Amsterdam, Overgegeven aan de Representanten van het zelve Volk, op den 20sten October 1795. Pieter den Hengst.

    • Search Google Scholar
    • Export Citation
  • Ferraro, J. M. (Ed.). (2019). A cultural history of marriage in the Renaissance and early modern age (1450–1650). A cultural history of marriage series, Vol. 3. Bloomsbury.

    • Search Google Scholar
    • Export Citation
  • Ferraro, J. M., & Pedersen, F. (Eds.). (2019). A cultural history of marriage in the medieval age (500–1450). A cultural history of marriage series, Vol. 2. Bloomsbury.

    • Search Google Scholar
    • Export Citation
  • Foreman-Peck, J. (2011). The Western European marriage pattern and economic development. Explorations in Economic History, 48(2), 292309.

  • Greif, A. (2006). Family structure, institutions, and growth: The origins and implications of Western corporations. American Economic Review, 96(2), 308312. https://www.doi.org/10.1257/000282806777212602.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hajnal, J. (1965). European marriage patterns in perspective. In D. V. Glass & D. E. C. Eversley (Eds.). Population in history. Essays in historical demography (Vol. I: General and Great Britain, pp. 101143). Aldine Transaction.

    • Search Google Scholar
    • Export Citation
  • Haks, D. (1985). Huwelijk en gezin in Holland in de 17de en 18de eeuw. Utrecht: HES Uitgevers.