Stasi / Facebook / Big Data DAAD Day 17 – Stasi networks and Facebook Social Graph

Similar to how Facebook uses a user’s connections to determine information about them, the Stasi considered all information they could gather about a person in their intelligence campaigns. They incorporated not only the contents of a target’s mail, but information about professional and personal relationships, love affairs, and organizations to which they belonged. (Reinicke, 106)

Source: ProPublica/BStU

This attention to not only the contents of a person’s postal communications, but their social connections can be evidenced in this hand-made graphic discovered at the BStU by Julia Angwin. The graphic shows forty-six connections, linking a targeted citizen to other potentially interesting persons (e.g. an aunt), places (e.g. “church”), and meetings (e.g. post, phone). Through whichever means necessary the Stasi made and utilized these connections in order to gain additional insights and institute methods of control.

Visual explanation of Facebook Social Graph. Source: Business Insider

This image reinforces the notion that the Stasi’s work was mostly a manual, often experimental, labor. It also tells us their work in surveillance and control, which is seen as a miserable, tyrannical sort, is now standard business practice among the software-automated surveillance capitalist models of the internet used by Facebook, Google, and others. Just as Department M workers like Gerd Reinicke mined the post for intelligence and PiD, the same basic processes the Stasi developed, now used as software by Facebook, are also at work in China, Iran, and other countries to filter and control what people can say and read through elaborate censorship software systems.

  1. Hanna Labrenz-Weiß. Abteilung M (MfS Handbuch). Hg. BStU. Berlin 2005.
  2. Reinicke, Gerd. “Mitlesen für den Klassenkampf: Postkontrolle der Stasi”. In: Heimliche Leser in der DDR: Kontrolle und Verbreitung unerlaubter Literatur (ed. Siegfried Lokatis). Berlin, 2008.
  3. Angwin, J. (2017). You Know Who Else Collected Metadata ? The Stasi. ProPublica, 6–8.

Stasi / Facebook / Big Data DAAD Day 16 – Notes from a former Department M surveillance operator (part2)

During the seven years Gerd Reinicke worked for Department M the scope of surveillance grew steadily. This is particularly evident in the 1980s, when as the government became more and more concerned of the “class struggle,” they dramatically increased the number of operators in Department M, more than doubling the size from 1980 to 1989.

Stasi Dept. M – Total employees (data from Abteilung M (MfS Handbuch). Hg. BStU. Berlin 2005.)

Reinicke and his coworkers in the “Evaluation and Information” unit constantly tried to increase the “Trefferquote” or “hit rate” of material worthy of confiscation by memorizing large numbers of postal codes, names, and handwriting. Their shift managers carefully sampled the work of individual employees, checking for missed letters, which could result in punishment and disciplinary measures.

Once a citizen or group had been classified as “operationally interesting” then all their mail was withheld for inspection. Citizens of the GDR who sent private messages critical of the government or its leaders through the mail first, had their mail confiscated, and second could be brought to court on charges of “anti-state agitation” according to Section 106 of the Criminal Code of the GDR. (Reinicke, 105)

  1. Reinicke, Gerd. “Mitlesen für den Klassenkampf: Postkontrolle der Stasi”. In: Heimliche Leser in der DDR: Kontrolle und Verbreitung unerlaubter Literatur (ed. Siegfried Lokatis). Berlin, 2008.

Stasi / Facebook / Big Data DAAD Day 15 – Notes from a former Department M surveillance operator

In the essay, “Eavesdropping for the class struggle: Postal control of the Stasi,” from the book, “Secret readers in the GDR: Control and distribution of illicit literature” (ed. Siegfried Lokatis), Gerd Reinicke describes his work in the “evaluation and information” section of the Stasi’s Department M. He explains his job was to assess and classify the contents of mailings, personal letters and printed materials, and in extreme cases, make decisions about confiscation.

This work was not easy. The stream of mail was constant and so he had a limited time frame for his work. He operated in secret, but also had to consider the public perception of his decisions. While technically illegal, the government watched all mail moving within the country, as well as that leaving or arriving in the GDR. As far as the population was concerned it was an “open secret,” because they observed the effect of the government’s policy of opening mail to remove valuable items, as well any writing or media considered to be “Politisch-ideologische Diversion (PiD)” (English: “Political-ideological Diversion”)—a task which was not as clear as it might seem.

Upon his hire was told to collect information on people and facts that might be “operatively interesting” to intelligence services. But he also received instructions to remove “anything unnecessary.” This work reflects the dual role that Department M played for the SED as a surveillance and control mechanism. On one hand, they watched for communications between spies or dissidents, but they were often a censorship apparatus.

This flowchart shows the three basic categories by which mail was classified, and subclassified, specifically, as “messages suspicious of being intelligence communications,” “messages suspicious of being PiD,” and “messages determined to be ‘interesting content’.” The assignment into of any of these categories determined further routing into the Department M machinery.

Heimliche Leser in der DDR: Kontrolle und Verbreitung unerlaubter Literatur (ed. Siegfried Lokatis)

These broad definitions also underline the ambiguity in how their surveillance and censorship was officially applied. For example, in dealing with Western and dissident literature, these instructions were followed in different ways. Sometimes operators were generous in their interpretation of “class struggle appropriate” materials, while other times he characterizes their subjective interpretation as “petty.”

  • Hanna Labrenz-Weiß. Abteilung M (MfS Handbuch). Hg. BStU. Berlin 2005.
  • Reinicke, Gerd. “Mitlesen für den Klassenkampf: Postkontrolle der Stasi”. In: Heimliche Leser in der DDR: Kontrolle und Verbreitung unerlaubter Literatur (ed. Siegfried Lokatis). Berlin, 2008.
  • Stasi / Facebook / Big Data DAAD Day 14 – Department M OibE groups

    Thanks to the implementation of index cards across the MfS, Department M had access to sender and receiver addresses, contacts, relatives, etc. to inform their surveillance work. By 1970, the department had instituted the following operational steps for examining mail. Numbers 3–6 were considered highly secretive and operations were performed under the codename “Eagle Flight”.

    1. Check if the sender or recipient existed in the police card index
    2. Establish an “Operational Comparison Card” to record information found in the mailing
    3. In the case of a suspicious mailing, carry out a search of the mail box
    4. Create photographic documentation of letterheads
    5. Compare the handwriting or typewriting font, spelling, etc. of the letter head. Identify any stealth writing.
    6. Form a special “OibE” group to check mail boxes.

    Offizieren im besonderen Einsatz (OibE) (English: Officers in Special Operations) groups were formed specifically to monitor, and in some cases, empty mailboxes under the cover of the Deutsche Post uniforms and vehicles. They photographed persons of interest as they dropped mail off, and then opened the mail boxes and examined the contents of letters before they even reached the post office. Photographic records of their work exist, and were presented in the exhibition, “Ein offenes Geheimnis. Post- und Telefonkontrolle in der DDR” (English: “An open secret. Postal and telephone control in the GDR”) (Labrenz-Weiß, 22–24)

    Source: BStU

    1. Hanna Labrenz-Weiß, Abteilung M (MfS Handbuch). Hg. BStU. Berlin 2005.

    Stasi / Facebook / Big Data DAAD Day 13 – The Stasi’s Electronic Databases

    The number of cards in the Stasi indexes grew to an incredible 39,000,000 by 1989. In order to provide constant access to this massive material data system they created various special storage units and containers as it expanded. The current mechanical powered card indexes in place at the BTsU were found after the end of the GDR and are still in use in order to meet requests of citizens who wish to view the data the Stasi kept on them.

    Mechanical index card file organization system used by the Stasi.

    Beginning in the 1960s, the MfS also used various electronic databases to track citizens.* These databases were never fully implemented as a primary source of storage because computing and software available in the GDR at the time, even machines stolen or purchased in the west, were unreliable. For six years the Stasi mirrored essential data from the F16 personal card into an electronic format. In 1975, Department XII, responsible for central data collection, information and storage, introduced the “System of automatic pre-selection” (SAVO) which allowed workers to verify that a person was “not yet recorded” in the system. This considerably reduced the time it took for verification, however it still required a substantial 12–21 days for a process to complete.

    By 1986 SAVO 2.0 was introduced providing printing functionality, increased capacity—now containing all data from the F16 card—and reducing search times to 1–6 days. It also provided teletype connections eliminating the need for couriers and allowing multiple agencies (passport control, tourism, etc.) to use the same data source simultaneously. (The Key to Power, 6) While this greatly modernized the surveillance workflow of the Stasi, it continued to act only as a duplicate of the F16 card, which was still faster for priority inquiries, and less susceptible to errors.

    A computer used by the Stasi. Source: BBC

    The logic for these implementations is echoed in modern database planning and use. While redundancy is employed for backups and increasing access speed across global networks, there is still a “master” copy of all data, or in the case of distributed system, automated processes (software) that merge and maintain distributed data into a master. Like the Stasi’s decision to completely mirror portions of the F16 system, modern systems use whichever implementation is most reliable for the master, while summarizations of that data may be “pushed” to other indexes for faster retrieval.

    The Stasi’s use of multiple systems, as well as the F16 and F22 card also preempt the concept of “joins”, which is a powerful feature in relational database management systems (RDBMS), to allow various pieces of data to be spread out across multiple indexes. Like the registration number on the F16/22 cards, records from multiple discrete RDBMS can be “joined” by matching a single identification field in order to supply additional, or limit access to other data, when performing queries.

    Multiple other computerized database systems were developed or used by the Stasi, including SOUD, a collection of information on the “enemy” shared between the GDR, Bulgaria, Cuba, Poland, the USSR, Czechoslovakia, Hungary, and Vietnam. The most expansive of all their electronic indexes, was the central database of personal data (ZPDB). Introduced in 1981, by the end it contained records on over 1,320,000 people, 417,000 situations, 558,000 objects, and numerous related information.

    The desire to track not just people, but occurrences and objects is directly visible in the largest person tracking organization today. In Facebook’s “Social Graph”, introduced in 2007, any person, place, thing, event, anything that can be named, is recorded in a massive attempt to commodify everything and our connections to it. Like the Stasi ZPDB, everything in the Facebook Social Graph is indexed by a unique identification string as single record, which Facebook considers a “node,” with links or “edges” that connect them. These definitions are based on Facebook’s desire to do data science on the their information, applying statistics and other mathematical operations from graph theory in order to understand their data and predict (and manipulate) the behavior of users.

    *Regrettably, all electronic databases and their magnetic storage was destroyed not long after the end of the GDR. Meaning, BStU staff now use the original paper F16 card system to provide services to the public.

    1. The key to power: Card indexes and other finding aids practices of State Security. Archivar 64. Jahrgang Heft 04 November 2011.

    Stasi / Facebook / Big Data DAAD Day 12 – The VSH card system

    The VSH index card system was yet another data storage method developed by the Stasi for monitoring and controlling the population. Introduced in 1974, the system used a red F401 card for all “secretly policed” persons, and a corresponding white F402 information card once that person was captured and began informing for the MfS. The system contained information not included on the Kerbloch index, for example, specifics deemed to be “operationally significant” like that person’s activity in the MfS, and ensured the flow of information on persons of interest across operational departments, districts, and facility services. The VSH files enormously expanded the number of persons surveilled, and were the most important cards used by operational service units until the end of the MfS. (The Key to Power, 6)

    Front of the pre-compaction, search and map card. Source: BStU

    Back of the pre-compression, search and map card. Source: BStU

    The VSH cards were divided into 15 categories ( PK ):

    • PK 1: Staff and agents of the enemy intelligence services
    • PK 2: Senior staff of the centers of political and ideological diversion
    • PK 3: Members of terrorist organizations and individual terrorists
    • PK 4: Members of Zionist hostile immigrant, clerical and other organizations
    • PK 5: persons who carried out orders from hostile intelligence agencies, centers of the political-ideological diversion, Zionistically hostile immigrant, clerical and other organizations against the states of the socialist community
    • PK 6: People unmasked as enemies of the opponent
    • PK 7: people who had not offered the information that was real
    • PK 8: persons who carried out provocative actions on the territories of the states of the socialist community as well as against their representatives and citizens
    • PK 9: Persons who were expelled from the states of the socialist community
    • PK 10: People who had committed particularly dangerous crimes
    • PK 11 diplomats and employees of diplomatic and consular missions of NATO , Japan and the People’s Republic of China
    • PK 12: correspondent of NATO -Mitgliedsländer, Japan and the People’s Republic of China
    • PK 13: Employees of trade and economic representatives
    • PK 14: Members of organizations that have been engaged in international smuggling on a large scale
    • PK 15: Persons whose activities were directed at a damage to the economy of the states of the socialist community
    1. The key to power: Card indexes and other finding aids practices of State Security. Archivar 64. Jahrgang Heft 04 November 2011.
    2. BStU: ( VSH ) of the Central Evaluation and Information Group ( ZAIG )
    3. BStU: Pre-compacting, search and identification card ( VSH ) of Division 1

    Stasi / Facebook / Big Data DAAD Day 11 – Kerblochkarten

    A Stasi Kerbloch card

    The Stasi used several other card systems to assist their postal surveillance. The Kerblochkarten (English: Kerbloch cards) system allowed storage of data encoded as a series of hand-made notches. Like other punch card systems, groups of cards could then be automatically processed by a machine based on the presence of notches and a predefined key. The Kerbloch card system made it possible to perform statistical evaluations and targeted searches on up to 350 cards/persons at a time. Importantly, this card amended the functionality of the the F16 cards and allowed the Stasi to quickly search for potential informants by querying characteristics relevant to a particular job. While anyone who has used a computer spreadsheet software today would take this sorting and searching functionality for granted, this was a highly useful tool for the Stasi, which by 1989 had amassed over 39 million index cards. (The Key to Power, 5)

    1. The key to power: Card indexes and other finding aids practices of State Security. Archivar 64. Jahrgang Heft 04 November 2011.
    2. BStU: Kerblochkarten (KK)

    Stasi / Facebook / Big Data DAAD Day 10 – Card systems of the Stasi

    The early days of the Stasi were marked by the development of new systems for monitoring the population. In the 1950s, having already incorporated a post-WWII file system used for denazification, the MfS implemented three new index card types to track citizens, all of which were stored in a central card index. These cards were distinguished by color and used to track persons considered hostile to the state (form A1 which was yellow), official staff (form B1, later replaced by the blue and grey F19), and persons who were arrested (the red, form C1).

    While official staff had titles and were on the payroll, the Stasi used hundreds of “unofficial staff.” These “Inoffizielle Mitarbeiter” (or IMs) were either foreign citizens who turned over valuable information like trade secrets, or domestic informants who spied on their colleagues, neighbors, and relatives (and sometimes even spouses). While money was a primary motivation, they were compelled for a number of other reasons, including blackmail, ideological commitments to the GDR, or gifts other than money, like lavish vacations, prostitutes, and official state awards. In fact, the Stasi implemented an impressive number of medals and rewards in order to convince both IMs and official staff to maintain loyalty and faithfully contribute their labor to the state. (Macrakis, loc 386) These awards are not unlike the loyalty cards, badges, and other gamification tactics that today’s corporations employ across their databases of customers in order to increase brand allegiance and sales.

    Similar to modern software development, the Stasi index card systems, like many of their methods over the years, were constantly being reworked and improved with each new iteration. For example, while the previous system used separate cards to track “enemies” and “friends,” as the Stasi learned to coerce those formerly consider hostile into becoming secret informants it became necessary to track persons who evolved into “frenemies of the state.”

    This situation presented a particular problem, as the Stasi needed to record and track personal data on these people, but their real identities needed to be protected should they agree to turn over valuable information. For these persons the Stasi implemented a separate “transaction card” (F22) that augmented the standard cards. The change updated the yellow card (F16, formerly A1), which contained the real name, address, birth date and place, occupational status, workplace or school, and the red card (F18, formerly C1), used for those arrested, which contained this information, as well as a mugshot, a visual description, and data related to their crime, and added a registration number if they were a Stasi informer or agent. (The Key to Power, 3–4)

    The central operation card “F22” Credit: BStU

    The new “transaction card” (F22) recorded Stasi informants only by this registration number, as well as a secret code name and name of the case officer who managed the agent. Finding the real name of a Stasi operative meant matching not only the registration numbers on these cards, but additional comparisons, because many agents purposely shared registration numbers to add extra layers of secrecy. (Macrakis, 78) This method is similar in theory to modern relational database lookup tables, where a user’s personally identifiable information (PII) is separated and protected from data dumps containing so-called “anonymized data” , or public key encryption, where data is transmitted securely by performing mathematical operations on it using a combination of a unique private and public key.

    1. The key to power: Card indexes and other finding aids practices of State Security. Archivar 64. Jahrgang Heft 04 November 2011.
    2. Macrakis, Kristie. Seduced by Secrets: Inside the Stasi’s Spy-Tech World (Kindle Location 1615). Cambridge University Press.
    3. Jason Koebler This Visualization of NYC’s Taxis Shows the Promise and Pitfalls of Big Data Motherboard (2014)
    4. Alex Hern New York taxi details can be extracted from anonymised data, researchers say The Guardian (2014)

    Stasi / Facebook / Big Data DAAD Day 9 – MfS Department M (1960’s developments)

    The implementation of zip codes across West Germany in 1961 made it much easier to monitor postal traffic between east and west. The Berlin Wall was erected in that same year, leading to a heightening of Cold War suspicions and increased emphasis on monitoring movement of information between east and west.

    The 1960’s also saw the various reorganizations and new leaders, including Lieutenant-Colonel Walter Mehrbach, whose years of experience in the postal service led to the development of the first training manual for Department M, and certification courses in mail surveillance at the MfS Law School in Potsdam. (Labrenz-Weiß, 21)

    As Department M grew in the 1960’s, so too did the amount of mail they examined and information they recorded. This demand for intelligence required constantly expanding data storage methods, which led the department to develop a card system for efficient alphabetical sorting and identification of people, places, and objects on both sides of the East/West German border. (Labrenz-Weiß, 19) I’ll dig in a little deeper over the next few days on these cards, for now a photo from my BStU archive research last summer:

    BStU: MfS, BV, Potsdam, Abt II 764 pg 6

    1. Hanna Labrenz-Weiß, Abteilung M (MfS Handbuch). Hg. BStU. Berlin 2005.