# Italian Processetti and Allegati: The Marriage Packet That Reaches Back to the 1700s

The marriage act is one page. Filed beside it is a folder of birth extracts, death certificates and sworn statements that reaches back into the 1700s. It exists because Napoleonic law made a couple prove which of their ancestors were dead before it would let them marry. Here is what is inside, and where it does not exist.


The marriage act is one page of a bound register, and it is efficient to the point of meanness. A date. A groom, with his age and his father's name. A bride, the same. Four witnesses. The mayor's signature. If the clerk was having a good day you also get occupations and the street.

You have pushed the family back one generation, to the parents, and stopped.

Filed somewhere else in the same archive, under the same number, there is a folder for that marriage. It may contain the groom's baptism extract copied out of a parish register from 1794. It may contain the death certificate of his father, and of his father's father. It may contain a statement sworn by seven people who knew the bride's mother, because her birth record could not be found and somebody had to say where she came from. It may contain a notarised declaration that the groom, who had been working in America, is not already married there.

None of that is in the marriage act. All of it was required to make the marriage legal.

**The short answer.** *Processetti* and *allegati* are the supporting documents a couple had to produce before an Italian civil marriage, filed as a packet beside the marriage register. Article 79 of the Italian Civil Code of 1865 lists what they had to hand over: birth extracts, the death acts ending any previous marriage, the acts proving an ancestor's consent, the certificate of banns, and anything else needed to justify their freedom to marry and their family condition. The Italian State Archives says plainly that these packets carry information not obtainable elsewhere and let you reach generations living in the second half of the 1700s. The reason they are so rich is not generosity: the law required consent from a specific named ancestor, so a couple had to document which ancestors were dead in order to establish whose permission they actually needed. And the packets exist only where civil registration existed, which between 1816 and 1865 meant the south, plus one duchy.

## The dead relatives are there because the law needed to know who was dead

This is the single fact that makes the whole record type make sense, and almost nothing written about *processetti* says it.

The French Civil Code of 1804, introduced into Italy with the Napoleonic annexations, set a consent regime with a cascade in it. Article 148 provided that a son who had not reached twenty-five, and a daughter who had not reached twenty-one, could not marry without the consent of their father and mother, and that in case of disagreement the father's consent sufficed. Article 149: if one of the two parents was dead, or unable to express their will, the consent of the other sufficed. Article 150: if the father and mother were both dead or unable, the grandfathers and grandmothers replaced them, with the grandfather's consent settling a disagreement inside one line, and a disagreement between the two lines counting as consent.

The unified Italian Civil Code of 1865 kept the same machinery almost word for word. Its article 63 repeats the ages, twenty-five for a son and twenty-one for a daughter, and the tie-breaking preference for the father. Article 64 sends the decision up to the grandfathers and grandmothers if both parents are dead or unable. Article 65 spells out a final rung: where there are no parents, no adopter and no grandparents, or none of them able to act, a minor needs the consent of the *consiglio di famiglia*, the family council. Article 63 also adds a rung the French text did not have, requiring the adopter's consent as well as the parents' for an adopted child under twenty-one.

Read that as a research instruction rather than as family law. To work out whose signature the marriage needed, the registrar had to establish who was alive. To establish that a father was dead, somebody had to produce his death act. If the father was dead, the consent climbed to the paternal grandfather, so if he was dead too, his death act was needed as well.

**The genealogy in the packet is a by-product of proving negatives.** Each dead ancestor in the folder is there because his death was the reason the authority to consent moved one rung further up the family. A packet with three death acts in it is a packet where the consent had to climb three rungs, and the shape of the paperwork is telling you the shape of the family's mortality before you read a single name.

The 1865 code then makes the consent document useful in its own right. Article 81 provides that where the ascendants' consent is not given in person before the civil officer, it must appear in an authentic act precisely identifying both spouses, and that the act must express the name, surname, profession, residence **and degree of kinship** of the people giving it. A consent act is therefore a small declared pedigree, written by people who had a legal reason to get the relationships right, and the same requirement appears in article 73 of the French code before it.

The Napoleonic period has one extra document worth knowing by name, because it records a family quarrel. Articles 151 to 153 of the 1804 code required a child who had passed the consent age to ask their parents' advice anyway, by a formal instrument called an *acte respectueux*, a respectful act. Between that age and thirty for sons or twenty-five for daughters, an *acte respectueux* that produced no consent had to be renewed twice more at monthly intervals, and a month after the third the marriage could proceed regardless. After thirty, one act and one month were enough. A packet holding three of them, a month apart, records a couple who married over a living parent's stated objection and waited out the statutory delay to do it legally. The 1865 code dropped the requirement, so this is a document of the 1806 to 1815 window rather than a general feature of Italian marriage files.

## When the birth record could not be found, five or seven people swore to it

Article 70 of the French code required the civil registrar to be handed the birth act of each of the future spouses. Then it makes provision for the situation every genealogist eventually meets: a spouse unable to obtain it may supply it instead with an *acte de notoriété*, which in the Italian files is an *atto di notorietà*, issued by the justice of the peace of the place of his birth or of his domicile.

Article 71 sets out what that document contains, and the list is remarkable. The declaration of **seven witnesses**, of either sex, relatives or not, giving the forenames, surname, profession and domicile of the future spouse, and those of his father and mother if they are known; the place and, as far as possible, the date of his birth; and **the reasons that prevent the birth act being produced**. The witnesses sign alongside the justice of the peace, and where any of them cannot sign, the act has to record that. Article 72 then required the act to be presented to the tribunal of first instance where the marriage was to be celebrated, which granted or refused *homologation* depending on whether it found the declarations and the stated reasons sufficient.

Article 80 of the Italian code of 1865 carries the same instrument forward with two changes that matter. The number of witnesses drops from seven to **five**, and they may now be relatives of the spouses, which the Italian text says explicitly. And the required contents grow by one line: as well as the names, professions and residences, the place and as far as possible the time of birth, and the reasons the birth act cannot be produced, each declaration must state **le cause di scienza di ciascun testimonio**, the grounds of each witness's knowledge.

That last requirement is the most valuable sentence in Italian marriage law for a genealogist. It obliges every witness to say *how* they know: that they were the midwife, that they lived in the same street, that they served with the father, that they are the child's godmother. A record of relationships between named people, sworn before a magistrate, is not a by-product here. It is a mandatory field.

Three more things come out of the *atto di notorietà* for a researcher.

**The reason clause is a historical record in itself.** A statement that the registers burned, or that the family came from a village now in another state, or that the parish was suppressed, is a documented explanation for a records gap you would otherwise be guessing about. That is the closest thing Italian genealogy has to formal [negative evidence](/blog/negative-evidence-in-genealogy): an official finding, ratified by a magistrate, that a specific record could not be produced and why.

**Five or seven witnesses is a community sample.** Named people, with professions and addresses, who were prepared to swear they knew a family's circumstances. That is a larger and better documented associate list than most researchers assemble in a year of work, and it is the raw material of the kind of [FAN club research](/blog/breaking-genealogy-brick-walls-fan-club-research) that breaks a wall when the direct line runs out.

**The birth date in it is sworn memory, not a record.** Both codes say the date should be given as far as possible, which is the law conceding the point. Treat a date from an *atto di notorietà* as testimony, weigh it as you would any [family story](/blog/how-accurate-are-family-stories), and prefer any actual register entry you later find over it.

There is a second, rarer *atto di notorietà* worth recognising, because it looks similar and proves something else entirely. Article 78 of the 1865 code allowed dispensation from both publications, for very grave causes, on production of an act in which five people, again even if relatives, swore before the *pretore* that they knew the couple well and could assure on their conscience that none of the impediments listed in articles 56 to 62 stood in the way. The code required the magistrate to read those articles aloud first and to give the declarants a serious admonition about the importance of what they were attesting and the gravity of the consequences. A packet containing one of these is a packet from a marriage somebody was in a hurry to conclude.

## The statute itself lists what should be in the folder

You do not have to guess at the contents, because article 79 of the 1865 code enumerates them. The spouses had to present, to the civil registry office of the commune where they intended to marry:

| What article 79 requires | What it looks like in the folder |
| --- | --- |
| The extracts of their birth acts | A birth or baptism extract for each spouse, hand-copied from a register |
| The death acts, or judgments proving the dissolution or nullity of any previous marriage | A death act for a first husband or wife, or a court judgment |
| The acts proving the consent of the ascendants, or of the family or guardianship council, where the law requires it | A consent act naming the relative and their degree of kinship, plus the death acts that explain why it was that relative |
| The certificate of the publications made, or the decree of dispensation | The banns, sometimes from two communes where the couple lived apart |
| All other documents which, in the variety of cases, may be necessary to justify the freedom of the spouses and their family condition | Everything else |

That last line is the one to dwell on. It is a statutory blank cheque, and it is the legal reason a marriage file can contain almost anything. Military service declarations for a man of conscription age, a poverty declaration, a notarial act legitimating children born before the wedding, a correction to a surname the register had misspelled, all enter under "all other documents."

Legacy Tree's Kim Running, writing in April 2018 about her own family's 1904 packet from a Sicilian civil registration series in the province of Palermo, gives a good example of how strange that category gets. Her great-grandfather had been in New Orleans before the wedding, so the file contains a notarial act declaring that while abroad he had not married and was still free to do so. That single document sent her to his 1898 passenger arrival with two brothers. The same packet held a statement that the bride's mother had not heard from her own husband in four years and did not know whether he was alive or dead.

The publications themselves are worth knowing in their original form. Article 63 of the 1804 code required the registrar to make two publications, eight days apart, on a Sunday, at the door of the town hall, each stating the forenames, surnames, professions and domiciles of the couple, whether they were of age or minors, and the same details for their fathers and mothers. Article 64 required an extract to stay posted on that door throughout the eight days between them. So a banns certificate is not filler. It is a second, independently drafted statement of both sets of parents.

Legacy Tree also gives a cataloguing convention worth knowing, though it is a working habit rather than a legal definition: packets from 1804 to 1815 tend to be catalogued as *allegati*, those from 1816 to 1865 as *processetti*, and those from 1866 onward as *allegati* again. Search an archive's catalogue for both words rather than for the one you expect.

## The record only exists where civil registration did, and after 1815 that was mostly the south

Here is the limit that decides whether any of this applies to your family, and it is the reason nearly every worked example you will find online comes from Naples, Sicily or Calabria.

The Italian State Archives, on its Antenati portal, sets out the sequence. Napoleonic civil registration, the *stato civile napoleonico*, was introduced in Italy from 1806, following the annexation of many regions to the French Empire and the introduction of the Civil Code, and it remained in force until 1815. It produced registers of births, marriages and deaths kept in duplicate, and the second original generally ended up in the State Archives of the relevant province. Antenati then notes that alongside those registers there are often numerous *allegati* or *processetti matrimoniali* carrying information not obtainable elsewhere, such as the paternity and maternity of the parties or their consent to the marriage, and that these allow you to trace back to earlier generations who lived in the second half or at the end of the eighteenth century.

Then the empire fell, and so did the system. After the fall of the Napoleonic Empire, Antenati records, civil registration on the French model was abolished in almost all the restored states. It was kept only in the Kingdom of the Two Sicilies, corresponding roughly to present-day southern Italy, and in the Duchy of Modena and Reggio. In Sicily it was introduced in 1820. In the Grand Duchy of Tuscany a mixed system was set up instead, by the *motuproprio* of 18 June 1817.

**What that means if your family was northern.** For most of the north and centre, there is a gap of roughly fifty years, from 1816 until unification, in which civil registration did not operate and no *processetti* were generated. For that period the records are the Catholic parish registers, and they have no equivalent supporting packet. An article about *processetti* is not your article if your family was from Piedmont, Lombardy or the Veneto and married in 1840.

Nationwide civil registration returned with unification. Antenati dates it precisely: the Italian *stato civile* was established by Royal Decree of 15 November 1865, number 2602, and has been in force since 1 January 1866 in every Italian commune. It requires registers of births, citizenship, marriage banns, marriages and deaths, always in duplicate, one copy staying with the commune while the second went, until 2001, to the competent tribunal and afterwards to the State Archives for permanent conservation.

Antenati also makes a coverage point worth quoting back to anyone who tells you the parish registers are enough. Civil registration covers all citizens, without distinction of sex or religious faith, which makes it more complete than military documentation, which concerns only men, and more complete than parish registers, which exclude atheists, the unbaptised and anyone professing another faith.

## Where the packets are, and why they are hard to search

The packets are archival, not indexed, and that is the practical obstacle.

The second original of each register series, and the associated packets, sit in the State Archive of the province, and a large and growing share are digitised on Antenati, the portal run by the Directorate-General for Archives at the Italian Ministry of Culture. For the post-1866 Italian registers, Antenati notes that they are accompanied by annual and decennial alphabetical indexes, which is what makes the marriage act findable.

The packets are not indexed that way. Legacy Tree's description matches what you find in practice: *allegati* are typically stored separately from the marriage register, as loose papers in a folder, filed in the same order as the marriage records, so you locate them by the marriage act number rather than by name. Find the marriage first, write down its number, then page through the packet series for that year looking for the same number. Most runs are in numerical order. Some are not, and then it is an image-by-image search.

So the workflow inverts the one you use on an indexed collection, in the same way that [Russian Empire revision lists](/blog/russian-empire-revision-lists-revizskie-skazki) invert it: you do not search for a person, you find the community and the year and then read. And as with the revision lists, the payoff for reading rather than searching is that the good material is in the parts nobody indexed.

## What a processetto does not prove

Every strength above comes with a matching limit, and stating them is the operating manual rather than a disclaimer.

**Almost everything in the packet is a copy.** The birth extract was hand-copied out of a parish register by a priest or a clerk, at the request of the bridal party, for this specific purpose. Legacy Tree makes the point directly and it is the right one: because the documents were hand-copied, transcription errors can be made. A date or a name in a packet is one copying generation away from the register it came from, and where it disagrees with the register, the register usually wins.

**A packet is not proof the marriage happened.** The documents were assembled in preparation for a marriage. Occasionally the marriage did not take place. Check the marriage act itself before you treat the packet as settling anything.

**The absence of a packet is not the absence of a record.** Survival is uneven, cataloguing is uneven, and digitisation is a great deal more uneven than the Antenati interface suggests. A year with no *allegati* online is a year whose *allegati* may still be on a shelf in the provincial archive.

**The document proves what the law required, not what was true.** An *atto di notorietà* proves that five or seven people were willing to swear to a birth. It does not prove the birth. That distinction, between what a court accepted and what actually happened, is the same one that decides how much weight to give a [Scottish kirk session minute](/blog/kirk-session-records-illegitimate-father-scotland) naming a father, or [an English will](/blog/pre-1858-english-wills-what-they-prove) naming children, and it is worth carrying into every record type you meet.

## What KleioBase does with a packet like this, and what it does not

Disclosure, since we sell a tool.

A *processetto* is the hardest shape of document for a conventional family tree program to hold, because it is not one record. It is eight or twelve records about six or ten people, produced on different dates by different authorities, bound together only by a marriage number. KleioBase is a research companion that turns exactly that kind of scattered material into one connected archive: each page you upload becomes a record linked to the people it names, the people are matched against everyone already in your [knowledge base](/docs/knowledge-base), and the Research Companion can be asked a question with up to five of those records attached at once, so "what does this packet establish about the groom's paternal line" is a question you can put to the whole folder rather than to one page.

The upload workspace has a Context box for notes before processing, which is where you tell it that the page is an 1838 *atto di notorietà* in Italian rather than a birth register, and what surnames you are expecting. Extraction goes through a Review step where every field is editable before anything is committed, which matters on a document class where the underlying text is already a hand copy. Each record is kept as a fixed snapshot of what that one document said, so when the packet's birth extract and the parish register disagree about a year, both claims stay visible against their own source instead of one silently overwriting the other.

Now the honest limits. Three of them.

KleioBase does not hold Antenati's images and cannot search them for you, so finding the marriage number and paging to the packet is work you still do yourself in the portal. Nothing in the software understands the consent cascade: it will extract a grandfather's death act as a death, and it will not tell you that the death act is in the folder because it is what moved the consent up a generation. And the transcription is only as good as the hand it is reading, which on these packets can be a hurried clerical copy of an already difficult original. A deep scan tags names it was not confident about, and even that does not catch every misreading, which is why the image stays next to the transcription and why the Review step is not optional.

## Where to start with the marriage you already have

Find the marriage act first, in the civil register or on Antenati, and write down its number and year. If you are working backwards from an emigrant and do not yet know the commune, that is a different problem with its own method, covered in [which column names the town](/blog/find-immigrant-ancestor-town-of-origin); a *processetto* cannot help you until you have a place. Confirm that the commune is one where civil registration was actually running in that year, which for anywhere outside the old Kingdom of the Two Sicilies, Modena and Reggio, or Tuscany's mixed system means checking whether you are inside the 1806 to 1815 window or after 1866.

Then look in the archive's catalogue for a separate series under *allegati* or *processetti* for that commune and year, and page to your number. Read the whole folder rather than the document with your name on it, because the value is disproportionately in the death acts and the sworn statements about people you were not looking for. And when you meet an *atto di notorietà*, read its reason clause before its names: it is telling you which record was already missing in 1838, which is usually the same record you have been failing to find.

The mayor who assembled that folder was not building you a pedigree. He was satisfying himself that two people were legally free to marry and that the right relative had said yes. That the fastest way to establish it was to make a family document its own dead, one generation at a time, is an accident of administrative caution that has outlived everyone it inconvenienced.

Canonical: https://kleiobase.com/blog/italian-processetti-allegati-marriage-packets