Searching for a Wet Nurse: Prato, 1395–98

Corinna Peres and Juliane Schiel, Vienna University
<>; <>

The process of entering a work relationship is very complex both on an individual and on a structural level. There are those searching for employment, those looking for someone to employ, and those assisting (or hampering) the search process through information, networking, and mediation (or the refusal of help). All parties involved are constrained by certain conditions, both subjective and objective ones. And while some are in the position to choose, others remain without options. This data story deals with the (potential) entry into a very particular work relationship: wet nursing a child. The story takes place in late medieval Tuscany and explores how wet nurses were found and hired at the time. It is reconstructed from the correspondence of the merchant couple Margherita and Francesco Datini. With the use of digital tools for text annotation and data analysis, we study the forms and levels of coercion involved in the relationship between those who wished or needed to work as a wet nurse and offered their services, and those who wished or needed to find a wet nurse for their own child or the child of someone else.

A Merchant Couple in Late Medieval Italy: The Datini Correspondence

The Datini archive in Italy is the most extensive merchant’s archive of the European Middle Ages. For the period from 1363 to 1411, it contains about 135,000 letters (125,000 business and 10,000 private letters), more than 550 account books, about 300 partnership contracts, 400 insurance policies, and several other notes, contracts, and cheques, documenting the economic and private life of a merchant’s household in the 14th and 15th centuries.

The following letters are part of the private correspondence between the merchant couple Datini. From 1384 to 1410, over a period of more than 25 years, Francesco Datini (1335–1410) and his wife Margherita (1357–1423) communicated through letters when the expanding business made it necessary for Francesco to temporarily leave the household in Prato. During his prolonged stays in other Italian cities, especially in Florence and Pisa, he managed his growing, medium-scale business in the Western Mediterranean. Francesco’s business spanned from Prato to Florence and Pisa, but also to Genoa, Avignon, Barcelona, Valencia, Mallorca, and Ibiza. In his absence, his wife Margherita supervised the household affairs in Prato and, from the 1390s onwards, sometimes even those of the extended household in Florence. When Francesco stayed in the Florentine branch, Margherita often remained in Prato and resided only occasionally in Florence. The spatial separation between the couple is a recurrent theme in their correspondence. Especially in the years 1397–1398, when Francesco was busy with his business in Florence, they saw each other in person only for very short periods. Letters were therefore the only means to stay in touch.

All letters start with the invocation of God and the date of writing and end with the sender’s name and the place where the letter was written. They are all written on paper and one to two pages in length. Some of letters were self-written, and some dictated to a scribe. Each letter was transported by one of the Datini’s carters. Transporting a letter from Prato to Florence took a few hours by donkey.

The epistolary exchange of the Datini couple – in total 433 letters, 251 by Margherita and 182 by Francesco – is a treasure trove for the study of labour relations. They document almost on a daily basis how the servants and employees of the Datini household behaved, what they asked for, and what they needed. The wife and husband complained about their servants and employees and sometimes one of the two defended them against the other. They passed their orders on to the servants and judged day by day whether they performed their tasks in a satisfactory manner.

Salaried Breastfeeding in Late Medieval Tuscany

The following selection of the correspondence sheds light on a very particular form of female work: wet nursing. In late medieval Italy, parents with enough financial means – initially aristocrats and in the 15th century also commoners – used to give their new-born babies to a wet nurse. Finding the right match was a delicate and often lengthy process for both sides. Medical thought and religious conviction placed enormous demands on the potential wet nurse, while ensuring the baby’s wellbeing was time sensitive and often required pragmatic solutions.

In general, wet nurses were paid more than female servants. The exact wage for breastfeeding, however, depended on the location and on the assumed qualities of the wet nurse. In rural areas, women were paid approx. 3–5 lire per month. In the region of Prato, wet nurses received about 4–4.5 lire, according to Margherita (case 2). Higher wages were paid in Florence or for women breastfeeding in the employers’ household. For the time period 1390 to 1399, Christiane Klapisch-Zuber estimates that a Florentine balia in casa (ital. for “in-house wet nurse”) earned on average about 6 lire per month, while female servants ordinarily received 2.5 and male servants 3.65 lire per month (Klapisch-Zuber 1988: 269; for currency exchange rates around 1400, see James and Pagliaro 2012: 402). Wet nurses working in an urban household could thus earn at least twice as much as female domestic servants.

The good financial remuneration for wet nursing often incited poor rural women to offer their services in order to increase their families’ income. They did so after having lost a child in infancy, or decided to give away their own new-born babies in order to either take in a Florentine baby for breastfeeding or to offer their services to households in Florence. From the money they earned from wet nursing they could in turn pay the wet nurse looking after their own child in the countryside, while still making substantial profit.

Profit could also be made when household masters rented out their female slaves “with milk” (cum lacte). The (mostly illegitimate) children of these household slaves were often given away to foundling hospitals. The more common household slavery became in late medieval Tuscany, the higher the number of abandoned babies in foundling hospitals and female slaves “with milk” whose services were offered by urban household masters, competing with rural and often impoverished women from the countryside who sought to increase their family’s income (Heers 1996: 199–204, Klapisch-Zuber 1988: 213–225). Christoph Cluse has described the capitalisation of female slaves’ body functions as “a new escalation level of exploitation of human beings,” which created a precarious market situation for breast milk (Cluse 2005: 85). In his testament, Francesco Datini himself donated part of his fortune to the Florentine hospital Santa Maria Nuova, with the instruction to build a new foundling hospital, the later Ospedale degli Innocenti – an institution that itself was in constant need of women “with milk.”

Three Cases: The Datinis as Mediators in the Search for Wet Nurses

The sample of 22 letters presented here, all written by either Margherita or Francesco, consists of three cases of negotiating wet nursing arrangements (1395–1398). In all three cases, Francesco and Margherita Datini, whose own marriage remained childless, were asked to act as brokers either for finding a baby to breastfeed (case 1) or for finding a nursing woman (cases 2 and 3). All cases show that searching for a wet nurse (or for a baby to breastfeed) not only involved the baby’s parents and the nursing woman, but a whole network of friends and relatives as well as business partners and dependent labourers. Francesco and Margherita tried to recruit wet nurses from Prato and its surroundings, an area well-known for the search of balie (Klapisch-Zuber 1988: 219).

In the first case (March 1395), Schiatta, a labourer of the Datinis, tried to sell the services of his wife as wet nurse and asked Margherita for help. Margherita judged Schiatta’s wife to be a low quality wet nurse and refused to help, not wanting to put her reputation at risk. In the second case (April 1397), Lodovico Marini, the relative of one of the co-owners of Francesco’s business (Manno d’Albizo degli Agli), searched for a wet nurse for his new-born baby. Margherita tried to sell the services of the wife of one of Francesco’s workers (Ceccarello), praised her good qualities as a wet nurse, and pre-defined the contractual terms for a potential agreement. By the time the woman in question accepted to negotiate a contract, however, Lodovico had already found another wet nurse. The third case (mainly August 1398) involved Francesco’s business partner Manno d’Albizo degli Agli himself, who was searching for a wet nurse for his illegitimate baby son. In this case, the Datinis spared no effort to find the perfect wet nurse. None of the potential foster mothers seemed to be good enough, however, and the Datinis came under pressure the more time passed. Finally, Margherita found an interim solution while continuing to look for a wet nurse of higher quality.

Yet, who felt under pressure when and why? What is the difference between feeling under pressure and being coerced? Which forms and layers of coercion are at stake here, and how can we as historians identify and differentiate those forms and layers of coercion and set them in relation with each other?

The selected letters reflect the perspective of the mediating party. They are written from a particular standpoint, according to the rules of private letter writing, and contain statements on, and not from, the other parties involved. The analysis that follows seeks to uncover not only forms and layers of pressure and coercion experienced by the mediating Datini couple, but also by those who were desperately in need of a wet nurse and those who (un-)willingly offered their wet nursing services.

The analysis consists of three parts: a quantitative data evaluation, a semantic text analysis, and a historical contextualisation of the most comprehensive third case. All letters have been processed into machine-readable txt-files, keeping the original spelling of the hand-written manuscripts. In all three parts, we analyse the letters both with the help of digital text mining and annotation tools and through close reading and hermeneutic interpretation.

Quantifying Coercion Through Quantitative Data Evaluation

The three cases of wet nurse mediation involving 22 letters and about 3,000 words give us an insight into the mechanisms of negotiating and entering a nursing arrangement (Figure 1). 16 of the 22 letters and 2,282 words are written by Margherita, six letters and 623 words by Francesco (Figure 2). The fact that more than three-quarters of the letters are written by Margherita suggests an uneven power relation and an unequal distribution of responsibility and workload in the couple: Francesco giving (short) orders, Margherita doing the actual mediating work and reporting back in great detail. Yet, we need to take into account here that for the periods under study at least two of Francesco’s letters are missing (from 16 and 20 August 1398), and that they might have contained relevant information on case 3 (Cecchi 1990: 21).

Figure 1
Figure 1. Percentage of words the Datini couple used to write about the wet nurse mediation compared to other themes in the sample of 22 letters.
Figure 2
Figure 2. Number of words Francesco and Margherita used respectively to write about the wet nurse mediation in the sample of 22 letters.


This data story focuses on the marital correspondence for pragmatic reasons, as the letters exchanged between Francesco and Margherita all exist in printed form and were thus easier to access for digital analysis. However, Francesco’s (handwritten) correspondence with his business partners and other people in his network would complicate the picture further. For example, Francesco exchanged 23 letters with Manno alone during the ten days when Margherita was searching for a wet nurse for Manno’s baby (case 3). This correspondence will be taken into account for the hermeneutic interpretation in the third part of the analysis; yet it is not included in the quantitative data evaluation and the digital semantic analysis.

When we look at the marital correspondence of the Datinis per case, we see that the commitment of the Datinis strongly depended on their relationship with the searching party, ranging from an insignificantly low to a very high involvement in the search. The closer the bond, or the more relevant the person in need for Francesco’s own network and business, the higher the probability that the Datinis would help out. Power and social relations played a crucial role in this type of work arrangement.

In the first case, one letter and 109 words from Margherita were enough to inform Francesco about Schiatta’s request to find employment for his wife as a wet nurse. The Datinis indirectly refused to help, did not forward Schiatta’s request, involved no other person in the search, and even sought to avoid circulating the information among their labourers. Schiatta’s subordinate position as an agricultural labourer and Margherita’s poor assessment of his wife’s milk quality, supposedly based on her age and social position, deterred the Datinis from offering their mediating services.

Figure 3
Figure 3. Number of words Francesco and Margherita used respectively to write about case 2.

In the second case, one letter and 78 words from Francesco and three letters and 587 words from Margherita dealt with the possible mediation of a wet nurse from Prato for Lodovico Marini’s baby (Figure 3). Margherita asked several people for advice but put her efforts mainly in mediating the wife of one of her labourers in Prato (Ceccarello). Besides Lodovico’s nephew, Manno d’Albizo degli Agli, on the one side, and the potential wet nurse, her husband, and her father, on the other, a labourer of the Datinis (Schiavo) and two members of a related family (Niccolò di Piero di Giunta del Rosso and his son Agnolo) were involved in the search. The search radius remained restricted to the Datini lands in Prato. The incentive to mediate, however, was not necessarily the relationship to Marini himself, but rather to his nephew Manno, Francesco’s important business partner and contact person in Pisa. While staying at Francesco’s Florentine residence, Manno personally asked Francesco for help on behalf of his uncle.

Figure 4 Figure 4. Number of words Francesco and Margherita used respectively to write about case 3.

The number of letters and the involvement of persons reached its peak when Manno made the same request one year later – this time for his own child, an illegitimate son most probably born from one of his female slaves. That case involved 12 letters and 1,586 words from Margherita’s side and at least five letters and 545 words from her husband’s side (Figure 4). This time, Francesco was in constant contact with Manno, Manno’s mother Monna Bice, and Zanobi di Taddeo di Gaddi, a Florentine merchant banker in Venice, who happened to desperately wait for Manno’s slave, the supposed mother of Manno’s baby, to breastfeed his own child. Margherita, in return, conducted negotiations, often in parallel, with eight potential wet nurses in the region of Prato. Besides them, she and her husband involved in the search two of their carters (Argomento and Nanni), two of their female servants (Bartolomea d’Aleso and Chaterina d’Andrea), their labourer Saccente, their relative and business partner Niccolò di Piero and his wife Lapa, and seven other persons from their personal network. Furthermore, Margherita reported that numerous neighbours and unnamed women were randomly asked for further suggestions in public places (piazza). It seems that in order to find a suitable wet nurse for one of Francesco’s most important business partners in Pisa, Margherita left no stone untouched not only in Prato and its surroundings but also beyond. The search radius was by far larger than in the previous case, as the annotated map illustrates.

The listing of letters, words (Figure 5), activated contacts, and locations mentioned per case draws a very clear picture: one letter, 109 words, and no additional person or place involved in case 1; four letters, 665 words and six additional people in Florence and Prato involved in case 2; and 17 letters, 2,131 words, 22 additional people, and several places like Montemurlo at a distance of up to 7.5 km actively involved into the search by the Datinis in case 3. The more important the person in need was for the Datinis, the more pressure the couple felt (and the more pressure Francesco put on Margherita) to provide a high quality wet nurse. However, the quantitative analysis tells nothing about the pressure or coercion experienced by the other parties, nor how the social pressure felt by the Datinis related to other forms of coercion in this situation. For this, we will now turn to a semantic analysis of the letters.

Figure 5
Figure 5. Number of words the Datini couple used per case.

Analysing Semantics of Coercion Through Text Annotation and Text Mining Tools

For the semantic analysis, we imported the txt-files of the letter sample together with the cross-corpora annotation model developed with the members of the working group “Grammars of Coercion” into CATMA, a computer assisted text annotation and text mining tool. We regrouped the letters by case and writer into separate text collections, while the cross-corpora annotation model set the basis for a joint tagset used to annotate the whole sample. Each statement of the letter sample then received a tag declaring whether the statement contains information (a) on the mediating party helping (or impeding) to find a wet nurse (mediator’s perspective) [action phrases, attributes given to potential wet nurses], or (b) on the party of the wet nurse searching for employment (worker’s perspective) [action phrases]. Properties designating the type of action and values qualifying the action phrase were not used in the annotation process as they cannot be sufficiently included in the Analyze Module offered by CATMA.

With the Analyze Module in CATMA, we generated queries based on the annotated letter sample. The Query Builder allows for various searches. We ran queries by tag, by frequency of words/phrases, and by grade of similarity (i.e., finding all the words that are similar to a degree of 70% to the one typed in). Some of the results were further transformed into visualisations. For figures 7, 8, 10, 13, and 14 we used the “Distribution Graph,” whereas for figures 6, 9, 11, and 12 we chose the “Wordcloud” and the “Double Tree.” The “Distribution Graph” displays how often a search tag or phrase is used in the annotated sample of letters. For every 10% of the letters’ length (per case or writer) (x-axis), the occurrences (y-axis) are summarized in a dot of a specific colour. Thus, on the x-axis, “0” means the first word from the letters considered in the analysis and “100” the last word of the sample. If a tag or term occurs in different 10%-paragraphs, it gets several dots according to its respective search hits in the narrative sequence. These dots are linked in a development graph. The “Wordcloud” is also an overview of the frequency of tags or words. Instead of positioning the search hits in the narrative sequence (graph), it summarizes them in a cloud. The higher the frequency of a tag or word in the Datini letters, the larger it appears in the (colourful) “Wordcloud.” The “Double Tree” is not so much about the frequency of words and phrases, but about their textual contexts, as it allows for an interactive keyword-in-context visualisation. Each branch of a typed in word or phrase leads to another context of keywords. For further information on the digital tool used for the analysis, please consult the Compact Manual provided by CATMA.

As the whole letter sample deals with the (potential) entry into a wet nursing arrangement, the annotation of action phrases allows for a reconstruction of the main activities and the time sequence of that process:

(1) mandare […] e fare cerchare per trovare una balia (“to send […] and make someone search for a wet nurse”) [employer’s and mediator’s perspective]
(1) venire per tôre uno fancullo a balia (“to come get a baby to wet nurse”) [worker’s perspective]
(2) fare cherchare le chondizione sue e chom’ela sta (“to make someone find out about their health and circumstances”) [employer’s and mediator’s perspective]
(3a) trovare una balia buona (“to find a good wet nurse”) [employer’s and mediator’s perspective]
(3b) dare [il fanciullo] ad una di queste […] che ce lo terebe uno mese o due, tanto noi ne trovasimo una buona (“to give the baby to one of these wet nurses […] who could take the baby for a month or two, until we find a good one”) [employer’s and mediator’s perspective]
(4) toglere il fancullo (“to accept/take the baby”) [worker’s perspective]

The (potential) wet nurses either came and offered themselves (venire) or just appeared and could be found (apparire). Once the agreement was made, they were given (dare) the baby or took (tenere, togliere) the baby for nursing.

The main activities connected to the Datinis’ involvement as brokers were mandare (to send) and trovare (to find). Whom to send – (transitory) wet nurses, the baby, servants – was a constant part of the Datinis’ correspondence. Also, as the couple’s main concern was a successful outcome of the mediation, the verb trovare (to find) was used repetitively by both Francesco and Margherita. Yet, while Francesco sought to serve his partners and friends (servire), which led him to use the verb twice as much as his wife, Margherita was primarily concerned with the search process itself. (Fare) cercare – i.e., to send and command someone to search and to coordinate the search – clearly dominates the reports she sent to her husband.

Figure 6
Figure 6. Based on all 22 letters. The search hits are: 32 mandare / “to send” (F: 10, M: 22), 25 trovare / “to find” (F: 6, M: 19), 13 cercare / “to search” (F: 1, M: 12), 10 avvisare / “to inform” (F: 2, M: 8), 7 servire / “to serve” (F: 5, M: 2), 4 inpacere / “to burden” (F: 1, M: 3), 2 piacere / “to like” (F: 0, M: 2).


The word clouds visualising the verb frequency in Francesco’s (left side) and Margherita’s (right side) letters (Figure 6) clearly illustrate the different roles Margherita and Francesco had in the search process. Francesco used the mediation of wet nurses to strengthen his own personal and professional network, while Margherita, instructed by her husband, was fully responsible to bring the order to a successful conclusion. When Francesco decided “to serve” someone, Margherita was obliged to follow his request. She was only free to decline when the request came from someone other than her husband, such as in case 1.

At times, however, the internal hierarchy between Francesco and Margherita could prove much more complex, as illustrated by the use of imperatives. Margherita reports in the first person how she commanded (farò cerchare, mandai), evaluated (a me non piaceva), decided (Io mi sono diliberata, patti ò fatto), and employed (ò presta). Even though she regularly received and followed instructions from Francesco and others, she was very autonomous in organising the search for a wet nurse. In case 3, she even reversed the direction of instructions in the couple: her use of imperatives increased when she finally found an interim solution for which she had to rely on Francesco’s cooperation. She had already secured an “interim” balia and instructed Francesco to send Manno’s baby in a text passage full of imperatives to ensure that Francesco would not ruin her meticulously prepared plan (Figure 7).

Figure 7
Figure 7. Based on the 17 letters of case 3. By “Imperatives (Direct Speech)” we mean requests and commands directly exchanged between the couple (e.g., di’ a Manno – “tell Manno”). “Imperatives (Indirect Speech)” are requests and commands forwarded to a third party and only reported in the letters, e.g., when Margherita reports how she made her subordinates search for a wet nurse (ò fatto cerchare). Interestingly, Francesco’s letters contain no imperatives in indirect speech; all his commands were addressed directly to Margherita.


A similar differentiation can be made between the emic labels that Margherita and Francesco used when talking about (potential) wet nurses. Margherita used femina, donna (woman), fanciulla (girl, maid), balia (wet nurse), and chosa (thing), while Francesco never used the term chosa. For him, it was a person with milk (persona ch’arà late) whom he almost always referred to as balia (wet nurse) (Figure 8).

Figure 8
Figure 8. Labels for wet nurses based on all 22 letters. Please note: the diverse numeric variation of emic labels used by Francesco (left) and Margherita (right) causes a diversely high y-axis (occurrences).


A coherent system of word usage cannot be detected, but there are still certain tendencies. When Margherita knew the candidates personally, even only superficially, she described them as women or girls (femina, donna, fanciulla). The job title balia (wet nurse), in return, was mostly used for women that were recommended to her by others but whom she had not met yet. Finally, when she referred to her failure to procure a suitable wet nurse, she also used the abstract term chosa or chose (thing/s), often collocated with the indefinite pronouns niuna (none) and nulla (nothing) and the past participle of the verb to find (trovato) (Figure 9).

Figure 9
Figure 9. Based on Margherita’s 16 letters of the whole sample. The words marked in red designate collocations, i.e., niuna chosa (“no thing”) can be preceded by trovato (“found”), and followed by che (“that”), coparirà (“appears”), m’è (venuta alle mani) (lit. “fell into my hands”).


The different word usage by Francesco and Margherita can again be explained by their respective roles as mediators. Francesco was mainly interested in getting someone able to breastfeed the baby of a partner, relative, or friend. The emic label balia referring to the actual job to be done (bàlia ch’abia late da potere dagli popa) was therefore his first choice. Margherita’s main concern, in turn, was to find the best wet nurse she could in order to consolidate her and her husband’s good reputation as brokers. The labels she used were consequently much more differentiated.

This internal hierarchy is not limited to the broker position. The same holds true both for the employer and for the worker positions. According to the Datinis’ presentation, the fathers of the babies in need and the husbands of the women offering their services initiated the search process and/or led the negotiations, while the children’s mothers and the “women with milk” remained mostly invisible. In fact, contractual arrangements were a male business in late medieval Italy, as Christiane Klapisch-Zuber has pointed out (Klapisch-Zuber 1988: 226). The father of the child conducted and formalized the negotiations with either the husband or the father of the wet nurse-to-be, and vice-versa, the husband of a wet nurse-to-be approached the household masters in search for employment. In our sample, the children’s mothers (as well as the children themselves) and the (potential) wet nurses were never mentioned by name, neither by Margherita nor by Francesco. The breastfeeding women were identified by their place of residence and, more importantly, by their husband’s and/or father’s name. The commodification of a female body function turned linguistically male.

When Margherita collected information about a potential wet nurse, she did so not primarily through direct questioning but by asking persons who had given their baby to them in the past, or by talking to the neighbours and the husband of the woman. In most cases, it is therefore almost impossible to say to what extent the women themselves decided to work as wet nurses and in how far they were forced into that form of employment by their husbands or other family members. A list of the statements attributed to the potential wet nurses themselves still shows a wide range of possible, although generally very limited actions (Figure 10):

  • Case 1: …bene ch'ella dicha che gl'abia pocho (“[The woman is old and the milk too abundant,] despite the fact that she says she has very little.”)
  • Case 2: …se dilibera di fare quello ch'io gli dirò, penso Lodovicho ne sarà be' servito (“…if she agrees to do what I say, I think Lodovico will be well served.”)
  • Case 2: …se togle questo fancullo (“…if she accepts this baby [I want it brought straight here at all costs.]”)
  • Case 3: …e' non vuole venire per esso (“She doesn’t want to come to Florence to get the baby.”)
  • Case 3: …à promeso che, se lla fanc(i)ulla sua muore istanotte, che sta per morire, ch'ella vi verà a mano a mano che l'arà sopelita (“she promised that if her dying baby succumbs tonight, she will come as soon as it is buried.”)
Figure 10
Figure 10. Based on the 17 letters of case 3. For the digital analysis, the potential wet nurses mentioned in the Datini letters were given a number and a short identification note. Any objectification is not intended by the authors. The “Actions” are directly performed by the wet nurses during the mediation process (e.g., no ci vole venire / “she does not want to come here”), according to the Datinis’ report. It is worth noting that the baby’s mother, the servant Domenica, and the wet nurse in Montemurlo do not (or cannot) take any active role in the process. The only wet nurses with a restricted radius of action are the “interim wet nurse” and the “pick-up wet nurse.”


However, whether a woman remained completely silent (i.e., wet nurses 4–6) or appeared in the letters as a direct interlocutor of Margherita, her carters, or her servants, the broker’s assessment of the wet nurse’s qualities clearly determined whether she was forced to promise and to accept the job or could take the liberty to refuse. We therefore need to turn to the attributes used during the assessment process in order to better understand the hierarchy of the intersecting factors considered in the assessment and the resulting degrees of scope of action and freedom of choice for the women in question.

Attributes used by Margherita or Francesco to qualify a woman as a wet nurse (Figure 11):

  • late frescho (“fresh milk”)
    • e i' latte suo àne due mesi: potrebe alevare di largho ongni fancullo (“she has had her milk for only two months; she could feed any baby with no trouble at all”) [Case 2]
    • abia i late frescho, in però non à u mese il faculo (“she must have fresh milk, because the baby is not more than a month old”) [Case 3]
    • una balia cho' late frescho (“a wet nurse with fresh milk”) [Case 3]
    • bàlia ch’abia late da potere dagli popa a pel chamino (“a wet nurse who has milk to breastfeed the baby on the way”) [Case 3]
  • una buona (“a good one”)
    • questa fanculla […] e' l'è d'uno richo lavoratore e d'una persona molto da bene (“because she is the daughter of a rich and respectable artisan”) [Case 2]
    • chon tutte le chondizione buone che si apartiene a niuna buona balia (“with all qualities that make an excellent nurse”) [Case 2]
    • una chosa profetta / buona (“a promising thing” / “a good nurse”) [Case 3]
    • è una buona femina (“she is a good woman”) [Case 3]
    • il marito l'à dato la parola ch'ella faccia quello ch'io voglio (“her husband has given his word that she will do as I wish”) [Case 3]
  • nonne ingrosa (“won’t become pregnant”)
    • e sì è la chondizione sua che nonne ingrosa mai insino che l'à ventotto mesi o più (“she is at the stage where she won’t become pregnant for twenty-eight months or more”) [Case 2]
  • d’uno richo (“of a rich”)
    • questa fanculla […] e' l'è d'uno richo lavoratore e d'una persona molto da bene (“because she is the daughter of a rich and respectable artisan”) [Case 2]
Figure 11
Figure 11. Based on all 22 letters. The wordcloud includes all general and specific statements in English translation which Francesco (left) and Margherita (right) made to qualify a woman as a (potential) wet nurse. The search hits are: 12 “a good one” (F: 1, M: 11), 10 “fresh milk” (F: 4, M: 6), 1 “with rich relatives” (F: 0, M: 1), 1 “won’t become pregnant” (F: 0, M: 1).


Attributes used by Margherita to disqualify a woman as a wet nurse (Figure 12):

  • vechia (“old”)
    • la donna è vechia e 'l late abundato (“the woman is old and the milk too abundant”) [Case 1]
  • balia con fanculo (“wet nurse with own child”)
    • perch'io non potrei mai credere che, quando elle ànno i fanculini d'un ano, ele non ne diano a loro (“because I could never believe that when a wet nurse had her own one-year-old child she wouldn’t be feeding it as well”) [Case 3]
  • latte ispopato / abundato (“little” / “too abundant milk”)
    • 'l late abundato (“the milk too abundant”) [Case 1]
    • a' latti ispopati (“someone with little milk”) [Case 3]
    • non n'à latte (“has inadequate milk”) [Case 3]
  • inferma / invalida (“ill” / “handicapped”)
    • trovai ch'ella moriva di fame tutto l'anno (“I discovered that she had been dying of hunger all year”) [Case 3]
    • dice che senpre mai quanto più la tenea, più lei diventava trista tra le mani (“she says that the longer the baby stayed in her care, the unhealthier she became”) [Case 3]
    • mi dice ch'ella non n'à se nonne uno ochio (“she also says she only has one eye”) [Case 3]
    • è inferma (“is unwell”) [Case 3]
  • non piacere (“to not like”)
    • a me non piaceva (“I didn’t like her”) [Case 3]
    • quella che gl'à non n'è chom'io vorei (“the woman he has is not the one I would prefer”) [Case 3]
  • tropo a lunge (“too far away”)
    • mi pare uno pocho tropo a lu(n)ge (“it seems to me she is a little too far away”) [Case 3]
Figure 12
Figure 12. Based on all 16 letters written by Margherita, as Francesco does not disqualify any of the potential wet nurses. The search hits are: 4 “ill/handicapped”, 4 “little/too abundant milk”, 3 “not likeable”, 1 “with a child of her own”, 1 “too far away”, 1 “old”.


Interestingly, Francesco never wrote about the (missing) qualities of a specific woman. The evaluation process was Margherita’s business alone, which emphasises again our observation that Margherita and Francesco had different roles in the mediation of wet nurses. Very obviously, the quality of the milk (fresh vs. too little/too abundant) appeared as the decisive and most frequent factor to qualify a woman as a wet nurse (Figure 13). Poor health seemed to be an equivalent for milk of poor quality and functioned as a relevant criterion for exclusion (Figure 14). A starving, sick, or disabled woman could never be accepted as wet nurse for a Florentine baby. The other most relevant attribute to qualify a woman as wet nurse was a good character. Yet, the attribute was generally used to second the main criterion of good milk. A buona femina might help out in case of urgent need but would not get the job without clear information on the quality of her milk or her general state of health. All the other criteria mentioned in the Datini correspondence underscored or relativised an existing assessment but were never the decisive factor.

Figure 13
Figure 13. Based on Margherita's 12 letters concerning case 3.


Figure 14
Figure 14. Based on Margherita's 12 letters concerning case 3.


Only those women with a positive (pre-)assessment appear in Margherita’s letters also as women with a certain degree of decision-making independence. Indirect speech from women with a negative assessment usually revealed incorrect statements (for example in case 1: “The woman is old and the milk too abundant, despite the fact that she says she has very little.”). (The only exception to that rule is surprisingly the interim wet nurse. She is selected by Margherita without a description of her qualifying attributes. Margherita might have qualified her according to the same list of criteria in an oral form, yet, as the letters contain no information, we just don’t know.)

To sum up, physical condition came first, followed by personal behaviour and social reputation, while the social or economic status played a subordinate role. Thus, a potential wet nurse might have been able refuse the job if she disposed of good and fresh milk fully available for the baby in need. Having a rich husband or an excellent reputation could help, but proved not to be the decisive factors.

Contextualising Coercion in the Case of Manno

The quantitative data evaluation and semantic analysis of the letter collection have pointed to a series of striking features and remarkable characteristics in the Datini correspondence. The two approaches helped visualise how wet nurse arrangements were made at the time. Yet, they don’t teach us why the Datinis put so much more effort into finding a wet nurse for Manno (case 3) than in any of the other cases. In order to answer the question of why, we need to contextualise the stories through close reading and hermeneutic interpretation.

In this last section of our data story, we focus on the third case, the search for a wet nurse for Manno’s illegitimate baby son. A detailed analysis of the relationship between Francesco Datini and Manno d’Albizo degli Agli based on their unedited letter correspondence will be offered by Corinna Peres in her dissertation "Versklavte für Haus und Hof. Drei Milieustudien zum mittel- und oberitalienischen Raum (1350–1550)". Here, we give some background information in order to better understand the open ends left by the data analyses in the first two sections.

Francesco and Manno knew each other at least since 1383. At the time, Francesco, who had met and married his wife Margherita while exiled in Avignon, was moving his business and family back to his hometown Prato. Soon after, he set up a new branch of his business in Pisa by taking over a company formerly run by Agnolo degli Agli, a senior relative of Manno. Agnolo, who had first taught Francesco how to run the company, gave the young Manno in the care of Francesco for apprenticeship. In 1392, Manno became junior partner of Francesco’s company in Pisa and conducted the business on-site, which soon became one of the most important import and export ports of the Datini company (James 2012: 4–7; Origo 1957: 135; 431). In 1398, Manno, who remained a bachelor all his life, begot a son with a woman who remains nameless in the correspondence but who seems to have been one of his domestic slaves. (Unlike Elena Cecchi, we read schiava di Manno, and not schiava di Marino in Francesco’s letter from 19 August 1398.) Zanobi di Taddeo di Gaddi, a Florentine merchant banker working in Venice, had chased away the wet nurse of his own child and was now waiting for the mother of Manno’s son to breastfeed his own baby, according to the Datinis’ correspondence. As Francesco had no branch of his business in Venice, he sold his products on the Venetian market by doing business with Florentine merchants who had a trading post in Venice. Zanobi was one of them (Mueller 1997: 267–268).

In late medieval Italy, merchant families used to conduct their business by splitting financial risks, operational responsibilities, and economic profit among the shareholders of a company. While the senior partners (socius stans) put up the main share of the capital needed and were in charge of the coordination and accounting of the business at home, the junior partners (socius procertans) brought in the smaller share of the capital, travelled, and conducted the negotiations on the spot. Knowledge was transferred within one or two families from one generation to the next, and often from uncle to nephew. Starting their career as adolescents, the sons of a senior partner often accompanied their (younger) uncles to the company’s trading posts, learned about buying, accounting, and promotion, and later became junior and then senior partners of the same (or a closely related) companies of their families (Schiel 2014: 126–127).

In our case, the two families Datini and degli Agli were obviously bound together in an intergenerational compagnia. Francesco had been junior partner of Agnolo’s company, and Manno became Francesco’s apprentice and then junior partner in the business where Francesco had the greater share (Origo 1957: 105–106, 128). Zanobi might have been the contact person for both families and their joint company in Venice. Francesco was thus bound in several respects: he felt obliged to help his commercial foster son Manno, who had made his business in Pisa a great success, and he felt obligated towards the older generation of Agnolo and Manno’s mother Monna Bice, whom he owed much of what he had built up after moving back to Italy. Also, he wanted to satisfy Zanobi, his contact person in Venice, as his company’s business depended strongly on the Venetian market. While the bonds towards the degli Agli had become family-like relations, the obligation he felt towards Zanobi was probably a much more professional one, implying another form of pressure than what he felt towards the degli Agli. Indeed, when Francesco wrote about his obligation towards Zanobi, he always used the verb servire (“to serve”). “To serve,” however, is the verb that, throughout the marital correspondence set two parties in an asymmetrical (work) relation to each other. The verb servire always created some sort of hierarchy.

  • io no’ llo volesi servire (“I didn’t want to help him”) [Case 1: Margherita doesn’t want to help her labourer Schiatta and his wife in finding a baby to nurse.]
  • penso Lodovicho ne sarà be’ servito (“I think Lodovico will be well served”) [Case 2: Margherita thinks that Ceccarello’s wife would do a good job as wet nurse for Lodovico’s baby.]
  • Troppo arò charo di servire Zanobi (“It is very important for me to serve Zanobi”) [Case 3: Francesco stresses that finding a wet nurse for Manno’s baby is important for him in order to serve Zanobi.]
  • no macherà che tu sarai servitto (“you will be served without fail”) [Case 3: Francesco reports that Giovanna di Gieri da Chapali has recommended Margherita as a successful wet nurse broker to Zanobi.]
  • cci vole servire u mese o 2 (“who would like to serve for one or two months”) [Case 3: Francesco accepts the interim wet nurse suggested by Margherita for Manno’s baby.]
  • mandarci quella che ti vole servire (“send us that one that wants to serve”) [Case 3: Francesco asks (again) if the interim wet nurse can be sent to Florence.]
  • tropo voletieri vorei servire Zanobi perché merrita ogni bene (“I would gladly serve Zanobi“) [Case 3: Francesco stresses again that he really wants to serve Zanobi by finding a wet nurse for Manno’s baby.]

The obligation Francesco (and Margherita) felt towards Manno, as well as the obligation Margherita felt towards Francesco (and Manno), in return, seems to have been a rather moral and emotional one. Here, we find per amore di (“for the sake of”) instead of servire. Margherita repeated several times that she would act as if the baby was hers or Francesco’s child (io ne farò di questo fatto chome se fose mio figliuolo o tuo), stressing that she would do so notwithstanding the fact that it was an illegitimate son (e non dubitino, perché e’ sia bastardi).

Yet, what was at stake here more generally, both for Francesco and for Margherita, was on the one hand the possibility to gain honour (onore) within their community and their network for a successful mediation, and on the other the risk of bringing shame (verghogna) upon their name and doing damage to their reputation. Margherita refused to help the labourer Schiatta as “it would be wrong and shameful” (è chosa d’averne pechato e verghogna). Francesco commanded Margherita “to gain honour (…) for yourself and for myself as well, do not bring shame on you because of it” (Or fa di farnele onore, e similemette a me no te ne fare verghongnia), and Margherita took on his order by resuming the same semantics.

Manno, for his part, was also in a delicate situation. As an unmarried bachelor, the birth of a baby son from one of his slaves opened the possibility to raise an heir for the family’s business and, at the same time, could cause idle talk and a bad reputation. According to Roman law, children usually inherited the status of their mother. When a domestic slave gave birth to a child, the child was a slave as well, unless it was recognized as legitimate by a free-born father. In most cases, however, as discussed in the introduction, those children were given to foundling hospitals without specifying their descent. Slave-born children were sometimes recognized only if the household’s master needed a legitimate heir (McKee 2004). In these cases, the slave mother was often sold or rented out to another household. This is what seems to have happened in the case of Manno.

Manno promised the slave mother to the merchant banker Zanobi as soon as he had found a wet nurse for his own child. Yet, fearing gossip, Manno, Francesco, and Margherita troubled to make her as invisible as possible during the whole search process. Francesco wrote that he didn’t want to send the baby with his mother (Manno’s slave) as he “did not want her to start gossiping with another person” (no voglio ch’ela entri a novella chon persona). He feared that a woman of her kind “would want to look and touch without reaching the objective” (vorebe vedere e tochare e no se ne verebe mai a fine), and Margherita shared his reasoning (“you have done right not to send her”).

Manno might have felt pressure from his family to produce offspring for the survival of the family’s company. The surprisingly active role mother Mona Bice played in the whole search process might point in that direction. At the same time, he might have felt pressure from the child’s mother, whom he seemed to believe capable of damaging his reputation. Asking his commercial foster father and close business partner for help was certainly a good choice to keep the whole situation confidential and receive support for the search process. It was in Francesco’s own interest as a close business partner to protect the good reputation of Manno and his family. Moreover, Francesco, whose marriage remained childless, also had several children with his household slaves. While most of them died at a very young age, his daughter Ginevra, born in 1392, was raised as legitimate daughter of the Datinis, while Ginevra’s mother Lucia was married to one of the Datinis’ servants and was finally freed (James 2012: 8–9).

The pressure felt by the (potential) wet nurses was obviously of a very different kind. Some hoped to work as wet nurses because the better salary could allow them to combat hunger and disease. According to the Datinis, the woman who had breastfed the child of Lapa and Niccolò and the other woman recommended by Niccolò would have accepted any conditions in order to have enough to eat. In a similar way, the woman who promised to come as soon as her dying baby was buried, and the one living in Montemurlo seem to have acted out of economic hardship. Others were obliged to work as wet nurses because their husbands, masters, or employers asked them to do so. The one who finally picked up the baby from Florence and left her own baby girl under the care of Margherita and her servants had to obey to her husband’s plans. Domenica would have had to accept the job if Margherita had chosen her to be the best wet nurse, as her husband Saccente was one of the Datinis’ workers. The slave mother of Manno’s baby, finally, might have known about Manno’s delicate position and might have tried to take advantage of the situation. Yet, similar to Domenica, she was simply sent to another place when her master Manno wished and decided to do so.

Defining and Differentiating Coercion

As categories of circumstances leading to loss of freedom or agency, economic hardship and asymmetrical power and labour relations clearly rank on a different level than the cultivation of professional and private relations and the defence of a good reputation. A poor woman refusing to work as wet nurse risked starving to death or falling sick. A person living and working in an asymmetrical power and labour relation risked punishment or being driven away. Someone refusing to help a family member or business partner find a wet nurse or someone who recommended a bad wet nurse might have experienced libel and slander among his or her peers and might have suffered economic losses, but the refusal or failure never had existential consequences for the person concerned. We therefore suggest associating the first two categories of circumstances with forms of coercion, while the other two categories are better associated with forms of social and emotional pressure.

Yet, besides these categorical distinctions many other factors like gender and age intersect and contribute to an increase or decrease in coercion and pressure. An impoverished woman just having lost her new-born baby but being herself young and healthy had a good chance to escape hunger by working as wet nurse in a Florentine household. She might have been forced to do so by her husband or have volunteered herself in order to leave the countryside and her family and live in the city of Florence. A slave woman giving birth to her master’s child might have been able to improve her social, economic, and even legal situation like in the case of Francesco’s Lucia, or she might have been rented out as a wet nurse to improve her master’s income, or sold and silenced by transferring her to a place far away from her previous social environment. A merchant’s wife could gain great independence and autonomy in running a household and sometimes could even reverse the direction of orders, or she could serve as an extended arm of her husband’s interests with little scope to set her own priorities. Besides gender inequalities, the age factor seemed to have played an important part in the Datini sample. While young merchants had to obey and respect the older generation, a young mother had much higher chances to work as a wet nurse than an old one. In all cases, however, the individual relationship, behaviour, and situation of the persons involved played a role in deciding whether someone was able to take advantage of a (potential) wet nursing arrangement or not.

The combination of close and distant reading of the 22 Datini letters has allowed some transversal observations. By juxtaposing quantitative data evaluation, semantic analysis, and historical contextualisation, coercive mechanisms in labour and social relations can be studied outside the boxes of “free” and “unfree” labour, of a domestic slave as opposed to a household servant or an agricultural labourer. The search for a wet nurse cut across many different legal and social groups and resulted in similar dynamics at different levels. Even though Francesco’s situation couldn’t be more different from the one of a rural wet nurse from the countryside of Prato, the use (and non-use) of the verb servire reveals how different forms of coercion and pressure can be made comparable via the underlying social dynamics involved. Those whose services were in demand had a greater range of action but also experienced a higher degree of social and emotional pressure. The woman living at Piazza della Pieve and the one living as far as Montemurlo had the liberty to refuse the job, yet the one felt Margherita’s impatient anticipation for her baby to eventually die and the other the expectation to be willing to travel the long way to Florence. Margherita had much autonomy in her way of assessing wet nurses due to her excellent reputation as a wet nurse broker. Yet, when she was successful, the honour gained from it was also Francesco’s honour, while an eventual failure would have brought shame primarily on her.