The first version of this article was a short one based on a comment I made on another Substack. This revised and expanded version is essentially a new article – thus the new publication date and title.
In A spatiotemporal reconstruction of the 1630 plague epidemic in Milan, Galli et al (2023) use daily death records from 1630 Milan to map when and where plague deaths occurred across the city. They show that the outbreak did not spread evenly but followed different timelines in different parishes, likely reflecting social and living conditions rather than a single, uniform wave of disease.
The results of this meticulous effort to reconstruct the event support the view that epidemics are not purely biological events, cannot be divorced from socio-political factors, and do not manifest the characteristics of “spreading” diseases in the manner health authorities across time and place would have the populace believe.1
My main interest in the study lies in the shape and behavior of the daily all-cause mortality (ACM) curve before, during, and after the excess death weeks. The data, shown below, are incomplete. Records showing deaths in hospitals and convents were not preserved or were not available. One register couldn’t be found in the public archives; hence, most of August is missing.
A complete dataset isn’t needed, however, for visual comparison with the Bergamo and New York “Twin Peaks” events in spring 2020, which have every appearance of being altered in magnitude, timing, or both.2
What stands out about the Milan 1630 curve (figure 1, reproduced with data from the study) is its organic shape and stochasticity. It’s full of noise, fluctuation, and variation across days of the week. It looks natural, even if the driving forces were social and political. The smooth, monotonic trajectories of the sort presented by COVID-Era dashboards and models are nowhere in sight.
Figure 1
Galli and colleagues don’t attempt to explain the daily spikes and drops in deaths, which could imply that they regard such “irregularity” as a normal feature of mortality records. The dramatic ups and downs might reflect specific triggers, or be a function of decedent records (or bodies) having been collected and documented in bursts (aka, reporting artifacts and “data dumps”).
Although the 1630 data largely show deaths in homes and on streets, not deaths in patient settings, there’s no reason to think the additional deaths would necessarily produce a less “noisy” curve. In fact, the missing institutional deaths would be expected to add variability, given differences in care, documentation, and death certification across homes, hospitals, and convents.
Comparison with Bergamo and New York 2020
The contrast between Milan 1630 and Bergamo and New York 2020, shown in figure 2, is stark.
Figure 2
With the 2020 curves, deaths shoot up in a bomb-like fashion without a break. Such steep, uninterrupted climbs — 12 consecutive days in Bergamo and 17 in New York — may defy precedent in contemporary death recording. Related analyses have raised similar questions about these curves, noting features more consistent with modeled or retrospectively-constructed series than with real-time death events.34
Heatwaves, cold snaps, power outages, earthquakes, and hurricanes can produce sudden jumps but not sustained “excess” curves for weeks on end. Part of the reason: outside of a war, there are few forces capable of generating sudden and dramatic excess death beyond a single day or several days. Daily mortality during New York City’s July 1917 and July 1999 heat waves, and August 2003 blackout (shown in figures 3, 4, and 5 below), are illustrative. The duration and intensity of the events isn’t strong or sustained enough to drive up deaths for very long.5
Figure 3
Figure 4
Figure 5
According to data reported in Vandetorren, et al, 2004 (image below), during the 2003 heat wave across France, Paris didn’t see more than two days of consecutive increases in deaths from all causes.
Comparison of daily mortality rate and mean temperature in Paris, France, for the years 2003 and 1999 through 2002 from “Mortality in 13 French cities during the August 2003 heat wave.”6
Even during periods of protracted elevated mortality, like some parts of the world experience in the winter months, the manner, cause, place, and patterns of incidence in death result in variability. See for example in figure 6 New York’s daily all-cause mortality during the relatively severe 2017-2018 flu season, which still shows expected stochasticity by day of the week.
Figure 6
What about New York City’s own 1918 “Spanish Flu” curve?
Data extracted from individual death certificates scanned into a database made public in March 20227 (figure 7) shows a period of seven days of consecutive increase between 7 October 1918 and 14 October 1918, with fluctuations before the 20 October 1918 peak.
Figure 7
Assuming that most death certificates were captured and accurately represent date of death, with some room for recording and other measurement error, we are still looking at a curve that is more realistic than the spring 2020 “Twin Peaks.”
Contemporaneous Comparison: Chicago and DuPage County (Ill.)
Two locations I have used previously for contemporaneous comparison with Bergamo and New York are Chicago (Cook County, population ~2.7 million) and DuPage County, Illinois (population ~930,000).
The 2020 all-cause daily death plots for both exhibit the same characteristics observed in Milan 1630: irregularity, noise, and obvious day-to-day fluctuation, even during periods of elevated mortality. When viewed alongside Milan 1630 (figure 8), all three curves present as organic and believable. They’re marked by pauses, reversals, and variability consistent with deaths occurring in different settings and the impact on recording.
Figure 8
Set against these curves (figures 9 and 10), the Bergamo and New York rises are highly anomalous. Their magnitude and speed are remarkable; most notable is the absence of expected stochasticity on the front end. Differences in density, climate, or demographics do not account for the contrast.8 9
Figure 9
Figure 10
The relevance of Chicago and DuPage County lies not in their similarity to Bergamo or New York, but in their resemblance to Milan 1630. What we see is jurisdictions separated by four centuries and radically different medical contexts, mortality curves which nonetheless share common structural features.
If steep, uninterrupted climbs were a natural consequence of a rapidly spreading lethal pathogen, human “reactions” and/or iatrogenic measures, such behavior should appear (at least intermittently) across settings with differing population density, demographics, and healthcare situations Instead, Chicago and DuPage, like Milan, display irregular, noisy trajectories consistent with how deaths are ordinarily experienced and recorded, even under stress.
Putting it another way: Why would the Bergamo and New York curves be “immune” to day-of-week effects or heterogeneity by place of death? It seems we are being asked to accept a degree of uniformity that even a highly coordinated hospital-based event (like I’ve described here) would struggle to produce. The prima facie features of the curves alone warrant scrutiny, apart from biological or iatrogenic explanations, and demand basic proof that the events occurred as depicted.
Why It Matters – and for Whom
The absence of expected stochasticity on the “front end” of the Bergamo and New York events should be regarded as a red flag and indicative of possible data construction rather than direct measurement, if not a strong signal that fraud has taken place. Even with incomplete data, the storied Milan Plague of 1630 has a more plausible shape.
To the extent that the “climb” of daily mortality curves in other locations exhibit characteristics similar to Bergamo and New York — whether in spring 2020 or thereafter — they also should be also be treated as highly suspicious and in need of substantiation with public death records.
Statisticians and epidemiologists should be concerned about the possibility of digital engineering being used to represent, model, and plan for impossible events.
Citizens and residents of the “geographical units” in which deaths occur and are recorded (i.e., city, county, province) have an interest in making sure authorities are not relying on estimates or projections, or otherwise presenting an aggrandized version of reality. As we know all too well, displaying unverified death numbers on a screen can easily be used to enact devastating emergency measures and justify harmful policies.
PDF version here.
- Also distilled well in this reaction to what the Italian researchers have shown: Engler, J. (2025, July 6). “The Milan Plague Epidemic of 1630 – sociopolitical, rather than a biological, event?” Sanity Unleashed. https://sanityunleashed.substack.com/p/the-milan-plague-epidemic-of-1630 ↩︎
- Hockett, J., & Engler, J. (2024, November 24). “Yes, We Believe the Bergamo (Italy) All-Cause Death Curve is Fraudulent.” Wood House 76. ↩︎
- Verduyn, T., Hockett, J., Engler, J., Kenyon, T., & Neil, M. (2023, November 1). “Does New York City 2020 Make Any Sense?” PANDA. ↩︎
- Hockett, J., & Engler, J. (2025, May 6). “Is the Bergamo 2020 death curve more ‘modeled’ than ‘measured’?” Wood House 76. ↩︎
- This isn’t to minimize the deaths that occurred but to point out that natural phenomena have limits, even when non-natural factors are involved (e.g., buildings without air conditioning, dereliction of duty in caring for the elderly). ↩︎
- Vandentorren, S., Suzan, F., Medina, S., Pascal, M., Maulpoix, A., Cohen, J.-C., & Ledrans, M. (2004). “Mortality in 13 French cities during the August 2003 heat wave.” American Journal of Public Health, 94(9), 1518–1520. https://pmc.ncbi.nlm.nih.gov/articles/PMC1448485/ ↩︎
- New York Genealogical & Biographical Society. (2022, March 16). “New York City public digitized vital records now online for free!” NewYorkFamilyHistory.org. https://www.newyorkfamilyhistory.org/blog/new-york-city-public-digitized-vital-records-now-online-fre ↩︎
- Hockett, J. (2023, October 13). “Three Cities, Same Virus?” Wood House 76. https://woodhouse76.com/2023/10/13/three-cities-same-virus/ ↩︎
- Hockett, J. (2025, February 26). “DuPage County (IL) vs Bergamo Province (Italy): Daily deaths, October 2017–May 2022.” Wood House 76. ↩︎
Leave a Reply