THE SPREAD OF BUBONIC PLAGUE OVER DISTANCES - Brill
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
CHAPTER FOUR
THE SPREAD OF BUBONIC PLAGUE OVER DISTANCES
Contiguous Spread and Metastatic Spread
The advocates of alternative theories, particularly Twigg,1 Scott and
Duncan,2 and Cohn,3 deny that bubonic plague spreads by leaps (per
saltum) by the movement of people or goods, technically often also
called metastatic spread. Although they must have seen Shrewsbury’s
presentation of the principle and mechanisms of metastatic spread of
plague by transportation of rats and rat fleas,4 they insist that bubonic
plague spreads only by contact between adjacent rat colonies, techni-
cally called contiguous spread. In the words of Twigg:
An important aspect of plague spread is the continuity across a rat popu-
lation. The more isolated the rat units the slower will be the movement of
plague amongst rats, and ultimately, to people.5
Among the specific examples he cites in accordance with his view that
this factor alone has the power of invalidating or undermining the rat-
and-flea-based bubonic-plague theory is the following:
The disease would be expected to have spread northwards from the coun-
ties of East Anglia, which it had reached in May 1349, spreading through
the rat population and causing local epizootics. Yet throughout this
large area the high mortality occurred only one month later over the
whole diocese. Plague is said to have reached York on 21 May but accord-
ing to Thompson (1914)6 the deaths of clergy over the whole of this vast
1
Twigg 1984: 57, 100, 185. However, Twigg acknowledges that plague can be spread
by fleas over various distances on pages 128–9, see below.
2
Scott and Duncan 2001: 79–80; Scott and Duncan 2004: 177.
3
In his monograph, Cohn accepts that bubonic plague can be spread by shipments
of grain, but denies all other types of spread by goods, textiles or clothing: see Cohn
2002: 29, 31–2. However, later, in a review of my monograph on the Black Death, he
denies flatly the relevance of the concept of metastatic spread and the fact of spread by
leaps, see Cohn 2005: 1354–5. Cohn repeats this view in a recent paper, Cohn and
Alfani 2007: 178.
4
Shrewsbury 1971: 29–34.
5
Twigg 1984: 100.
6
A.H. Thompson has written two valuable papers on the mortality of the beneficed
clergy in the Black Death which will be discussed at some length below in the chapter
on seasonality. See also the bibliography.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free access152 chapter four
northern diocese were on the increase soon afterwards. As we have seen,
bubonic plague can be excluded from the reckoning if these facts are cor-
rect, and there is no reason to doubt them.7
Objections against this account of the spatio-temporal pattern of spread
are not important at this point; I cite the passage to illustrate the implied
unconditional view that plague spreads according to a territorially
contiguous epizootic process, by direct contact between adjacent rat
colonies, and, consequently, that the Black Death could not have been
bubonic plague.
Presumably the advocates of alternative theories all adhere to this
view of the very slow territorial spread by contact between rat colonies
because it is obviously incompatible with the real spread rates of his-
torical plague epidemics. Since this can be seen as a sufficient condition
for rejecting the conclusion that historical plague epidemics were
bubonic plague, they can therefore triumphantly conclude that histori-
cal plague epidemics must have been a different disease, which justifies
their endeavours to identify an alternative disease. This erroneous view
is misleading other serious scholars in the field, derailing potentially
good research.8 The time is ripe for taking the advocates of these asser-
tions to task.
The metastatic mode of spread has long been empirically established
as a characteristic feature of the epidemiology of bubonic plague (as of
most other epidemic diseases). Since the advocates of alternative theo-
ries deny or ignore the facts on this crucial point in order to save their
theories, I will now not just paraphrase or summarise the scholarly lit-
erature on the mechanisms of spread of bubonic plague, but will supply
a broad collection of citations from the primary studies and standard
works on plague which should settle the matter definitively. However,
Cohn’s assertions on the matter in particular are so extraordinary that
they warrant some concluding remarks on the background of the evi-
dence presented.
In this chapter, the citations constituting the heart of the subject will
be presented chronologically to demonstrate the history and develop-
ment of this insight, how conspicuous this feature of plague epidemio-
logically was, how early plague researchers identified these mechanisms
7
Twigg 1984: 185.
8
Wood, Ferrel, DeWitte-Aviña 2003: 444; Drancourt, Houhamdi, and Raoult 2006:
234–41; below: 193–4, 461.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free accessthe spread of bubonic plague over distances 153
of spread and the consistency and firm tenability of these findings over
time. Observations of these features were first made several years before
the IPRC began its work.
The first standard work on bubonic plague based on modern medi-
cal studies in India, China and elsewhere was published by Simpson in
1905. Twigg cites and refers to it twice9 and must be presumed to know
it. Nonetheless, Twigg passes over in silence Simpson’s information on
the great importance of metastatic spread, for example, that
in Canton, many persons, especially the well-to-do, removed to the coun-
try, thus forming fresh foci for its dissemination; and in the same way the
outbreak in Hongkong no doubt arose from persons having migrated
from Canton to Hongkong.10
He notes that in China, “the infection on land has followed chiefly the
routes of busiest intercourse.”11 Simpson also provides much informa-
tion on spread by leaps in India in connection with the mass exodus
out of Mumbai, for instance, when the plague epidemic suddenly
blazed up: “Fugitives from Bombay and the Bombay Presidency were
not long in carrying infection to the other provinces of India.”12 W.G.
Liston, who joined the IPRC in 1905 as the Commission’s entomologi-
cal specialist, stated in a paper published earlier the same year that
infection could be conveyed from one place to another, either by infected
rats and fleas transported by ships and trains conveying merchandise or
by infected fleas carried on the clothing of man.13
In 1906, W.B. Bannerman, the director of the modern medical labora-
tory that had been built outside Mumbai at Parel and was placed at the
disposal of the IPRC,14 published a paper in The Journal of Hygiene
where he summarized plague-related research in India:
Spread of Infection by Means of Clothes
[…] Major Collie, I.M.S. (1898), relates the following striking instance
from his experience. “A man lost his wife from plague in Bombay, and ten
days later he brought her clothing and ornaments to his house in a village
near Harnai in the Ratnagiri Collectorate. In about a week dead rats were
found in this man’s house and neighbourhood, then one relative after
9
Twigg 1984: 135, 212.
10
Simpson 1905: 62.
11
Simpson 1905: 65.
12
Simpson 1905: 71.
13
IPRC 1907f: 713.
14
IPRC 1907g: 725; Lamb 1908: ii–iii.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free access154 chapter four
another sickened and died from plague, and ultimately the man himself
became the sixth victim. None of his relatives had been out of their vil-
lage. Eventually, this village and others suffered severely, and many lives
were lost. No plague cases had occurred up to the direct infection.”
Captain W. Ronaldson Clark, I.M.S. (James, 1899, p. 152) brings for-
ward several instances in which conveyance of infection by clothes seems
the most likely means of introduction in some villages in the Punjab
[…].
In the report of the fifth season of plague in the Punjab (Wilkinson
1904, a, p. 29) it is recorded: “That infection could be carried by clothes
was however recognised by the people of Ambala district fairly soon, and
in all towns the practice of burning the clothes of persons who died of
plague was introduced and acquiesced in by sweepers who had previ-
ously suffered severely from taking the abandoned clothes of the dead.”
In Patiala State it was noticed that infection was also conveyed by cloth-
ing and bedding of persons who had died from plague, and this was so
borne in upon the sweepers of Patiala that they refused to touch these
articles, which often had to be burnt where they lay in consequence
[…].
In the report in the Punjab for 1902 the following information is given
on what is regarded as “fairly definite” evidence. The mode of communi-
cation from one place to another was considered to be: “Human
communication in 88 instances. Probably human communication in
7 instances. Clothes in 25 instances. Probably clothes in 7 instances.” The
Government review in commenting on these results says (Wilkinson,
1902, p. 15), “It is therefore clearly established now that plague is spread
from infected to uninfected places almost entirely by personal communi-
cation on the part of persons who have resided in, or visited the infected
places during the existence of plague there.”
It serves to explain the infectivity of clothes so often noticed. It is prob-
ably the fleas in the clothes that are the danger, not the clothes
themselves.
We may sum up the above evidence as follows: —As regards introduc-
tion into a new area remote from a focus of disease, it seems certain that
human beings are the carriers of infection.15
Persuasive cases of spread of plague over considerable distances in
clothing were among the early observations made by British medical
personnel in India. Cohn and Twigg refer to Bannerman’s paper, which
they therefore presumably have read, but completely ignore these find-
ings, which are here presented in the form of a number of extracts that
easily could have been multiplied.
15
Bannerman 1906: 189–95, 208.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free accessthe spread of bubonic plague over distances 155
Similar observations were made during the plague epidemics in
Australia which occurred more or less simultaneously with these early
Indian epidemics. These observations were presented in a paper pub-
lished by the distinguished leader of the combat of plague in Australia,
Ashburton Thompson, in The Journal of Hygiene immediately after the
publication of the first reports released by the IPRC. His observations
pre-date shortly the establishment of the rat-flea theory of spread over
distances by the IPRC:
Extension of local infection by contiguity
The infection always first made its appearance at some one spot; never at
more than one. From that spot it extended continuously. […] this way of
spreading I distinguish as “extension by contiguity.”
Extension of local infection per saltum
While infection was apparently present only at the point of invasion or
central focus, it often made appearance quite suddenly at some distant
point, and there created a new focus […].
It appeared that the infection could be transported safely, be planted
afar, and could there take effect. In that there is nothing novel, for the
same happens with several communicable diseases. It becomes remark-
able and important when it is remembered that, in accordance with our
fundamental observation, the infection of plague is not maintained or
diffused by communication with the sick. This way of spreading I distin-
guish as “extension per saltum.”
For convenience of description outbreaks which result from extension
per saltum may be called “sub-epidemics.” In reality they have their own
independent status. […]. The numerous secondary outbreaks which we
have witnessed were not the less independent that they were isolated by
but a mile-wide ring of plague-free houses, nor the more independent
had they occurred (as we have seen) 70 or 2300 miles away from the
point of primary invasion.16
Ashburton Thompson introduced the concepts and terms of spread or
extension by contiguity and spread or extension per saltum, in English
“spread by leaps.” These concepts and terms were, as we shall see,
accepted and used by members of the IPRC, although they often relied
on accurate descriptions rather than on specific terms and concepts,
writing that infection was transported, imported, carried, transferred
or conveyed, or with other similar words and corresponding nouns,
from one place to another. As their studies progressed, they tended to
16
Ashburton Thompson 1906: 542–3, 546, 561.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free access156 chapter four
develop supplementary terminology and write for example of the,
“implantation” of infection and of infections which are “implanted.”17
Thus observations of these modes of spread of bubonic plague belong
to the initial phase of scientific plague research, before the IPRC had
set to work, as a consequence of their being very obvious features of the
characteristic modes of spread of this disease.
Ashburton Thompson’s paper can either be ignored and passed over
in silence, which is the strategy adopted by Twigg and Scott and
Duncan, or it can be viciously attacked without mentioning these
observations, which is the course chosen by Cohn (see above), in both
cases presumably because they recognized its disastrous consequences
for their own alternative theories which are based on the denial of
spread by leaps of bubonic plague.
The IPRC’s first observations on the spread of plague by the agency
of rat fleas were presented in a study of rat fleas’ willingness and ability
to use man as a host. Some central findings were:
We have made many observations which show that P. cheopis [= Xenopsylla
cheopis] will make use of man as a host, and may be captured in large
numbers on men in houses infested with rat fleas […]. We have on many
occasions caught rat fleas on our own persons, as well as on those of the
attendants […]. From the above experiments it is seen that we were able
to keep rat fleas alive for nearly four weeks by feeding them on human
blood […] (a) that they fed readily on man, and (b) that some of them
were still alive after 25 days.18
The willingness of rat fleas to attack men and the great number of fleas
involved were underlined by the outcome of another experiment.
A (vaccinated) person went inside a house in Mumbai which had been
evacuated after finds of dead rats and human cases of plague, and stayed
in one room only briefly. After he came out, a flea count was made,
with the following result: April 18, 51 rat fleas, April 19, 34 rat fleas,
April 20, 60 rat fleas.19 As Liston, a former member of the IPRC, later
commented:
It is quite possible that, had we not examined this man’s legs and removed
the fleas from them, he might have carried some of these fleas to his
home […].20
17
IPRC 1910d: 528, 534.
18
IPRC 1907e: 472–3.
19
IPRC 1907e: 474–5.
20
Liston 1924: 997. Cf. below: 164–5.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free accessthe spread of bubonic plague over distances 157
Scott and Duncan refer to this paper, but apparently did not see these
central statements.21
Summing up their findings, IPRC also stated that this study “shows
how readily and in what large numbers rat fleas may under certain cir-
cumstances come on to man.” They also concluded that the experi-
ments “indicate that rat fleas may often be transported in this way from
place to place, especially from plague-infected houses.”22
The next step was to study the arrival of plague in isolated villages on
the outskirts of Mumbai, specifically the villages of Sion, Wadhala,
Parel and Worli. A useful illustration of the IPRC’s work and method-
ology and the observations they made in these villages is provided by
the following example from Worli village:
Jankibai was a woman close on 70 years of age who lived in Worli. She left
the village in perfect health on the 14th February to attend the funeral of
her nephew, Laxuman Narayan, who lived at 94–6 Sonapur Street in
Bombay City. Motiram Ramji, another relative of the old woman, was ill
with plague in the same house in which Laxuman had died. After the
funeral Jankibai remained on at Sonapur Street nursing this latter patient.
Dead rats had been found in the house at Sonapur Street, and many dead
rats, found in the neighbourhood at this time, were proved by us to be
plague infected. Motiram died on the 18th February. Jankibai in the
meanwhile had developed fever. She came back to Worli on the 19th suf-
fering from plague, as evidenced by high fever and a well-marked bubo
in the groin. She went to a house in Block I, building No. 54, where we
saw her on the 21st February.
Two days after Jankibai came to the village, we placed two guinea-pigs
in her house […]. One of the two guinea-pigs which had been placed in
Jankibai’s house after isolation in the laboratory died of plague on the 9th
March […].
Taking all the facts into consideration it seems to us that the guinea-
pig which contracted plague in Jankibai’s house probably derived its
infection from infected fleas imported by Jankibai from Sonapur Street
in Bombay City. This case closely resembles one recorded in Parel village.
In this instance, it will be remembered, another such woman, coming
from the city, appeared to have brought infected fleas to that village (Case
1; Parel village).23
21
Above: 88, 90.
22
IPRC 1907i: 888.
23
IPRC 1907h: 869–70.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free access158 chapter four
The IPRC concluded their study of Worli village thus:
Three cases of plague which occurred in the village were investigated. It
is probable that these three cases all contracted the infection outside the
village.
In two instances there is evidence which points to infection having
been introduced in the clothing or persons of people [sic! my insertion],
and of this spreading in one instance to a guinea-pig, and in the other
instance to rats.24
Thus, at the very beginning of their studies of the mechanisms and pat-
terns of spread of bubonic plague the IPRC ascertained that plague
spread over distances, in this case from “Bombay City” into villages on
the outskirts of the city. They were also able to establish empirically at
a substantial level of tenability that in two of these cases the infection
had been imported in the clothing of a person and in a form that caused
spread of plague to a guinea-pig or rats, which strongly incriminates rat
fleas as the agent of transportation of the infection, riding in clothing
from the city into houses of the village, attacking their preferred hosts,
namely the black rat or alternatively other available rodents. This
explains also that “the carrier of the infection may not contract the
disease, as the Sion and Wadhala cases show.”25
Clearly this material exposes some central flaws of Scott and Duncan’s
work, since they repeatedly refer to this report for their views on the
role of rats in plague epidemiology,26 while maintaining throughout
their monographs that plague spreads only by contiguous contacts
between rat colonies, not by leaps. If they had really read the report,
they should have known better and ought to have eagerly acquired
more knowledge of the Commission’s work or at least have read Lamb’s
ninety-three page summary. However, as will be shown below, they
apparently had not read the IPRC’s reports, although it is a overriding
motive of their work to show that the results of the Commission’s work
is not compatible with the manifestations of historical plague epidem-
ics. Instead, they think it is adequate to build on Hankin’s pre-IPRC
paper of 1905 for their assertions regarding the epidemiology of plague
in India.27
24
IPRC 1907h: 873. The word “persons” could conceivably by a slip of the pen for
luggage.
25
IPRC 1907i: 891.
26
Scott and Duncan 2001: 72–3, 112, 357.
27
Scott and Duncan 2001: 77–8.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free accessthe spread of bubonic plague over distances 159
Inspired by their early observations, the IPRC moved briskly to
expand and deepen their knowledge on the spread of bubonic plague.
First, they did so by extending their studies of the spread of plague in
“Bombay City and Island.” In their next Report (no. XXIV), Section III
is called “The Transportation of Infection to a Distance,” which all
advocates of alternative theories have avoided or evaded, whilst it
should have been of great interest and of obvious crucial importance to
their projects. The IPRC begin by stating their principal explanatory
theory:
If we can exclude modes of spread of the infection of plague other than
the rat flea, we must conclude that the transportation of infection to a
distance is attributable solely to the conveyance of the infection in the rat
flea. A little reflection suffices to show that transportation of infection in
this medium is not only conceivable, but that under certain circum-
stances it may be a very likely contingency.
They go on to discuss “Transportation in Merchandise, Grain etc.” and
“Transportation of Rat Fleas in Clothing.” The last subtitle is too brief,
since it does not reflect an important experiment described in this sec-
tion which took into account an expanded concept of implicated
textiles:
Bundles of clothing, bedding, etc, were sent to the laboratory from houses
in the City in which plague cases had occurred. The bundles after being
opened out were kept in a flea proof godown28 for several days […].
Along with the clothing guinea-pigs were placed in the godown. In some
instances the animals were allowed to run free, in other cases they were
placed in pairs in cages, the control animal being protected from fleas
either by means of a layer of tanglefoot or by a curtain of wire gauze
[…].
The result of the experiment was that two rat fleas were caught on the
animals, that one of the free-roaming guinea-pigs died from plague,
and that one of the three rat fleas caught on the tanglefoot of one of the
cages was shown to contain plague bacteria in its stomach contents.
This permitted the IRPC to conclude:
(1) that rat fleas may be transported to a distance in bedding and cloth-
ing removed from plague houses; and
28
“So convinced were Colonel Bannerman and Captain Liston of the tenability of
the flea-transmission theory that they had designed and built at the Plague Research
Laboratory, Parel, special godowns or cabins, in which it was proposed to carry out
large series of experiments to test and prove this theory.” Lamb 1908: 2, 38.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free access160 chapter four
(2) that such rat fleas may prove infective if transferred to a susceptible
animal in the place to which they are carried.
We would point out that in whatever way rat fleas are transported,
whether in clothing or merchandise, they will select, when carried to
their new surroundings, either their true host, i.e., the rat, or the next
best available animal. If then infected fleas are imported into a house they
will by preference attack the rat rather than the human occupants of the
house. It is apparent from the account we have given of the rat infestation
of houses in Bombay, that under such circumstances opportunities for
transference of infected rat fleas to rats in houses are abundant. It would
appear, then, that the introduction of infected rat fleas into a hitherto
uninfected locality may lead to serious consequences by giving rise to an
epizootic amongst the rats.
It has been shown that infection may be transported to a distance by
means of rat fleas in clothing or merchandise and that such infection,
when imported into a hitherto uninfected locality, may give rise to an
epizootic in the rats.29
We may in effect complete this statement by mentioning that an epiz-
ootic among commensal rats is the point of departure of plague among
human beings in their proximity.
The following year, 1908, the IPRC published another study on fleas
containing specific subchapters on rat fleas and “The Mode of Dispersal
of Fleas.” Among their findings were:
That rat fleas may be attracted to man, jump on him but take some time
to feed on him. Plague infected fleas might in this way be carried from
one place to another without infecting the man, but would, when brought
near a rat, attack it in preference to man.
We would, however, like to draw attention to the ease with which rats
with their fleas can be transported in certain kinds of merchandise. We
have seen rats dive, as it were, into bags containing bran and disappear, so
that the bags could be moved without any evidence of the presence of the
rats within. M. rattus [= Rattus rattus, the black rat] from its habits is
particularly liable to be transported in this way […].
From what has been said above it will be apparent that merchandise
and grain, which have been visited by rats, may have fleas deposited on
them and these fleas may be transferred with these articles to distant
places […].
Fleas will be more readily carried on the clothes of a person, in that the
man who wears the clothes forms an attraction for the insects […] exper-
iments carried out in the godowns […] show how readily and in what
large numbers rat fleas may, under certain circumstances, come on to
29
IPRC 1907i: 886–90, 894.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free accessthe spread of bubonic plague over distances 161
man, and leave us certain that rat fleas must frequently be transported in
this way from place to place, especially from plague infected houses,
where they are more likely to take to man because of the absence of their
true host.
In whatever way fleas are transported, whether in clothing or mer-
chandise, etc., they will select when carried to their new surroundings,
their true host or the next best available animal.30
Also in 1908, G. Lamb, the Director of the IPRC, published a summary
of the Commission’s studies in the period 1906–May 1907 as published
in The Journal of Hygiene, some of them in the second half of 1907. This
summary is a good alternative source of knowledge of the opinions of
the IPRC on various issues, particularly on the subject of mechanisms
and types of spread discussed here. There is no excuse for the advocates
of alternative theories not having used at least this easy way to obtain
basic knowledge of their results, but none of them availed themselves
of this opportunity. Part VII is called “The Importation of Infection
from an Infected to an Uninfected Locality.” In this part, the main
points are presented in three easily readable pages.
In conclusion, then, we may state that all evidence points to the usual
method of the importation of plague from an infected to an uninfected
locality being by means of infected rat fleas and to the fleas being brought
in by human agency, namely, by healthy men, by plague cases or in their
belongings.31
In the following years, the IPRC continued to deepen and enlarge on
these findings. In their study of plague in Belgaum, a small town situ-
ated roughly 400 km south-east of Mumbai, in 1908–9, the Commis-
sion noted that plague probably was introduced into the urban centre
from an outside village and that there arose three effective centres of
spread in the town and that, in addition, plague was transferred in a
bag of bran over a distance of about two km from the centre to the fort
on the outskirts of the town. The Commission also emphasizes that
Given an epizootic such as we have seen raging in the Market it would be
difficult to imagine more favourable conditions for the dissemination
and spread of infection than this weekly gathering afford, a spread not
only to other parts of Belgaum but to surrounding villages. The commis-
sion agents, such as those whose stores, as we have seen above, were har-
bouring plague infected rats, import and export grain etc. by road and
rail and sell it to retail and other wholesale dealers. The grain is contained
30
IPRC 1908b: 253–5.
31
Lamb 1908: 81–3, cf. 91, 93. See also IPRC 1911b: 76.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free access162 chapter four
in gunny bags which are capable of harbouring not only rat fleas but rats
as well. It requires little imagination to picture what a serious danger a
market such that of Belgaum is, in times of plague, not only to itself but
to all places with which it has trade intercourse […]. There is always a
danger of importation of plague infected rats or fleas or both in the sacks
of grain and other stores that are continually being imported into these
godowns and when once introduced with the large rat population that
these buildings shelter there is nothing in he plague season to prevent the
disease spreading and an epizootic and epidemic resulting.32
As can be clearly seen, the fact of metastatic spread or spread by leaps
over small and intermediate distances is very obvious to the IPRC as is
the importance of grain and grain trade in the spread of infected rats
and rat fleas.
At about the same time, a similar study was carried out in Pune
(Poona), a city situated about 120 km south of Mumbai, with similar
results which however deepened the Commission’s understanding of
some aspects of the mechanisms of spread. As in the study of how
infection was introduced into villages outside Mumbai, special empha-
sis was put on the introduction of infection into the city. The pressure
of infection on the city was obvious; since the earliest cases of plague
were imported from Mumbai by the railway, people were either sick on
arrival and were taken to hospital directly from the train or fell sick
soon after their arrival. The IPRC suspected that the first indigenous
plague cases would be associated with the railway station. This hypoth-
esis was borne out: rats dead from plague were first found in the goods
sheds of the railway close to the station. Thus, the early history of the
epidemic and epizootic “started in the neighbourhood of the station,
the rats acquiring the disease from infection probably imported by the
railway.” The first indigenous plague case occurred in the sweeper’s
chawl, the railway servants’ quarters at the station and soon nearby
hotels and restaurants where people contracted plague and could act as
carriers of infected rat fleas to other parts of the city.33 The IPRC also
observed that a number of infective centres arose in the city that could
not be linked directly to persons coming from Mumbai or elsewhere, a
development with which they were familiar and which reflected the
fact that infected rat fleas carried in luggage or clothing would often
first seek out rats at their place of arrival and trigger an epizootic which
32
IPRC 1910c: 466–7, 470–1.
33
IPRC 1910d: 506–07.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free accessthe spread of bubonic plague over distances 163
would run its imperceptible course and in due time manifest itself in
human cases. This pattern was particularly likely in this case because it
took only four hours by train from Mumbai to Pune. At the same time,
the IPRC were closely monitoring the situation in the part of the city
called the “Poona Cantonment.” As the epidemic situation in the region
unfolded,
large numbers of people meanwhile were flocking into the city from the
surrounding infected areas to live there with friends […] one would
expect to find therefore, first a number of imported cases of the disease,
occurring in widely scattered localities: then later a large number of
indigenous cases, also widely scattered, arising from numerous infected
centres […] especially when we bear in mind the fact that rat fleas are
readily carried by man, almost unknown to himself, from one place to
another, and that the fleas by choice select rats for their hosts rather than
man. Centres of epizootic infection arise in this way in areas where a his-
tory of the importation of the infection could only be obtained with the
greatest difficulty. The history of the epidemic in the city will be seen to
bear out this latter hypothesis.
Next there follows a detailed description of the early appearance of
plague cases which corresponds closely to this outline of the epidemic
process and to the IPRC’s notions of plague epidemiology as developed
in their research projects in the field and in the laboratory.34
Other aspects of spread over distances were studied in especially
designed research projects. Greenwood’s major epidemiological study
“On the Spread of Epidemic Plague Through Districts with Scattered
Villages,” namely villages in three districts situated in the northern
provinces of the Punjab and the United Provinces (corresponding
roughly to the present-day province of Uttar Pradesh) is of particular
interest. He was able to conclude that the origin of the great majority of
village outbreaks was the importation of the disease from elsewhere,
and the corollary that recurrence of plague could in the main be attrib-
uted to re-importation of the disease, and that the chance that a village
would be attacked increases with increasing population.35 The implica-
tion was that plague was spread by leaps between scattered villages by
human agency by old-fashioned means of transportation. It was
also shown in another study that certain regions were conspic-
uously exposed and that the proximity to habitual lines of travel and
34
IPRC 1910d: 510–18, 527–9, 534–5.
35
IPRC 1910b: 361, 363, 368, 371, 374, 377–9, 443.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free access164 chapter four
transportation, especially railways, was a factor in this result; this study
also emphasized the importance of metastatic spread, this time by
modern means of communication.36
Shortly afterwards, Greenwood published yet another report on the
epidemiological structures of plague in the Punjab and pointed out,
among other things, that “effective centres of spread” arose in three
ways:
(1) By the introduction of an infected human being.
(2) By the human importation, on the person or in the clothing, of infec-
tive fleas.
(3) By the importation of infective fleas or infected rats in merchandise.
All the methods by which a centre is established ab extra, will still apply
for the establishment of secondary centres within the community, and in
addition we have to reckon with chances of spreading through the rat
population by the transference of infective fleas from individual to indi-
vidual. […] What is material, however, seems to be the apparent fact that
the extent of an epidemic must largely depend upon the number of “effec-
tive” centres established early in the epidemic. The diffusive power of
plague is evidently not so great that one “effective” centre is sufficient to
develop a wide-spread outbreak […] many conflagrations will be
required.37
Indisputably, the IPRC showed that bubonic plague was spread by
human agency both between communities and within communities
and that the central means were importation and spread by infective
rat fleas in clothing or merchandise, and that spread between contigu-
ous rat colonies was only of minor importance, the epidemic process
being mainly dependent on an increasing incidence of new “effective
centres” where each new “conflagration” tended to increase the dynam-
ics of spread by multiplicative powers. This shows how completely
erroneous Twigg’s, Scott and Duncan’s and Cohn’s assertions are with
respect to the IPRC’s views on the mechanisms and dynamics of the
spread of bubonic plague.
In 1924, Liston, formerly a member of the IPRC, published a short
outline of bubonic plague. In the part on epidemiology of plague he
takes as his point of departure the study in which a vaccinated person
on three successive days went inside a house in Bombay which had
been evacuated after finds of dead rats and human cases of plague,
36
IPRC 1911a: 47–61.
37
IPRC 1911b: 76, 78.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free accessthe spread of bubonic plague over distances 165
stayed in one room only briefly, and attracted 144 rat fleas that were
picked off his legs (see above). He writes: “It is quite possible that, had
we not examined this man’s legs and removed the fleas from them, he
might have carried some of these fleas to his home, and there they
might have transferred their attraction to their natural hosts,” unleash-
ing a rat epizootic among the rats that would translate into plague cases
among the persons in the house. He continues:
These observations show that infected rat fleas do not generally travel
far from the spot where they left their host; they generally lie in wait
for the passing of another host. If this new host is a rat, the fleas are
readily attracted to it and may then be carried to a distance limited by
the extent of the rat’s wanderings. If the new host is a man, then the
distance over which the fleas can be carried may be considerably greater,
and will be limited by the distance and by the speed at which the man
travels.
Plague thus progresses in two very distinct ways: first, to contiguous
areas—here infection is carried chiefly by rats; secondly, to more distant
centres per saltum—in this case infection is transported by men. Men
may carry rat fleas either on their person or in their clothing, or they may
transport rats and their fleas with merchandise, in which case railways
and ships materially assist the dissemination of the disease.38
This may be considered a final summary of the IPRC’s findings. As can
be seen, Liston underscores the fact that plague spreads in two main
ways, by contiguous spread between rats and per saltum, that is by
leaps. However, in the last case, Liston clearly distinguishes between
spread by rat fleas over short or intermediate distances by human
beings in clothing or luggage, and spread over long distances in mer-
chandise by ship, railway or lorry; in a pre-modern perspective one
could also add by horse and cart or pack horse. In this way he distin-
guishes between more or less local spread in short-range leaps estab-
lishing new centres of spread in a locality in a more or less multiplicative
way and, thus, constituting an important and even crucial element
in the dynamics of the development of plague epidemics in urban cen-
tres, localities or districts; and longer distance disseminations. This
view is clearly completely compatible with the presentation of IPRC’s
work above.
The IPRC’s findings are not unique, they do not differ from the
results of epidemiological studies of the modes and mechanisms of
38
Liston 1924: 997, 999, 1001.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free access166 chapter four
spread of plague elsewhere. Plague researchers working on plague epi-
demics in Egypt stated that they were of the opinion
that the creation of apparently independent foci of rat and human plague
in a town like Qûs is attributable to the transportation of infective rat-
fleas by persons who inhabit or visit infected areas in the town; these
persons may or may not suffer from plague in consequence.
The possibility of the carriage from one hamlet to another of infected
fleas by human beings is strengthened by the results of examining the
clothes of four plague patients from Atmour Mistigid. Only two P. irri-
tans [human fleas] were found in the four lots of clothes, and the num-
bers of X. cheopis were 4, 4, 4, and 1 […]. And, further, 185 X. cheopis
were taken from the clothes of three of our rat-catchers after they had
dug out [rat] nests in houses in Atmour Mistigid.39
The highly localized character of the work of these plague researchers
and its consequent short-range perspective is reflected in these conclu-
sions, which emphasize that in local towns or villages the spread of
plague is engendered by the movement of rat fleas in the clothing of
human beings.
In a manual for medical and health workers working in China writ-
ten by several leading plague specialists, C.Y. Wu agrees completely
with the IPRC and subsequent researchers’ findings that plague is
spread by infected rat fleas in clothing and goods, the central state-
ments here being enumerated from 1 to 5:
(1) “All the evidence available goes to prove rather […] that infected
fleas imported into a locality through human agency first cause
plague among the rats, than that human beings are directly
attacked.”
(2) It is accepted that “X. cheopis may be carried great distances on
human beings or in their baggage,”
(3) It is also accepted that in this part of Asia the “medium of trans-
portation of infected rat-fleas over long distances [was] by ship
transporting rice and grain,” and it is specifically mentioned that
Rangoon had acquired a “sinister reputation” as a source of impor-
tation of plague in these commodities.
(4) Wu notes that “Cotton is another effective vehicle for the transpor-
tation of X. cheopis,” and that Hirst had come to the conclusion that
39
Petrie, Todd, Skander et al. 1924: 128, 131.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free accessthe spread of bubonic plague over distances 167
“raw cotton, weight for weight, is an even more dangerous vehicle
of X. cheopis than grain or any other kind of merchandise.”
(5) He also underlines the “ever-present menace of dissemination by
infected rats hidden in cargo or by infected fleas secreted in grain
and cotton shipments.”40
Obviously C.Y. Wu here describes spread of plague over various
distances by transportation of rat fleas in clothing, luggage or
commodities.
The term spread per saltum is used by Wu Lien-Teh in this compos-
ite standard work on plague written by several specialists and, in his
monograph on primary pneumonic plague, he distinguishes between
contiguous spread and spread per saltum for this type of plague.41 He
also cites J.D. Long’s observation in a report on plague on the
West Coast of South America, that under favourable climatic conditions,
such as low temperatures and relatively high humidities, “fleas can act as
reservoirs of plague infection, carry it over long distances and later under
favourable conditions transmit the disease.”42
In 1914, J.J. van Loghem and N.H. Swellengrebel, two Dutch research-
ers, published a study on the epidemiology of plague in Java (Indonesia)
titled “Contiguous and Metastatic Spread of Plague.” For reasons that
will become clear below, the gist of this paper will be presented in more
detail than would be necessary under normal circumstances. At the
heart of the matter is a discussion of the pace and pattern of spread as
defining features that can distinguish between modern bubonic plague
and historical plague epidemics. Van Loghem and Swellengrebel note
that
at the start of an epidemic of human plague it seems often as if the plague
has moved by leaps: the neighbouring localities of a plague infected vil-
lage remain often untouched while villages situated farther away become
involved in the sufferings […]. There can hardly be any doubt that rat-
borne plague can be introduced from far away by human agency. The
epidemic in Java is no exception in this respect. The question is how this
rapid transfer over longer distances (metastatic spread, as we will call it)
comes about.43
40
C.Y. Wu 1936a: 285–90, and 1936b: 485.
41
Wu Lien-Teh 1936b: 204; Wu Lien-Teh 1926: 182–3.
42
C.Y. Wu 1936a: 286–7.
43
“Kontinuerliche und metastatische Pestverbreitung” = “Contiguous and Metastatic
Spread of Plague.” My translation from German. See also below: 176–7, 179, 182–3.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free access168 chapter four
They point out that the IPRC explained “metastatic spread over land by
the assumption that human beings transport rat fleas in their clothing
or luggage,” but also that the conditions in Java were fundamentally
different for two main reasons:
(1) because there were far fewer cases of human plague in Java than
in India, “which indicates a lower intensity of plague among
rats” and
(2) because the number of fleas on the rats was far smaller.
For these reasons “the likelihood that rat fleas would ride with human
beings and that human beings would assume a significant role in the
transfer of plague, seemed therefore at the outset smaller than in British
India,” although the occurrence of metastatic spread of plague between
distant villages had been noted and discussed with interest.
In this context, they point out that a leading Dutch medical scholar
(Dr de Vogel) had shown that in the city of Malang the centres of infec-
tion were in places where large quantities of rice were stored and then
distributed to the surrounding districts. He had also shown that the
beginning of plague epizootics in Java occurred shortly after the impor-
tation of large quantities of rice from plague infected areas abroad
(Rangoon and Singapore). Plague was first introduced into the harbour
of the city of Surabaya, and subsequently several smaller urban centres
connected with Surabaya by railway were attacked almost simultane-
ously and large numbers of dead rats were found in goods wagons of
the railway:
This indicates the predominant significance of transportation of com-
modities for the metastatic spread of plague.
This view was corroborated by a study of the clothing and luggage of
inhabitants of Malang that wished to leave the city because of the epi-
demic there. Only very few rat fleas were found, and the case was much
the same in the plague hospitals, evidence that supported
the view that it was transportation of commodities and not transporta-
tion by human agency that played a predominant role in the metastatic
spread of the epizootic. [….]
Furthermore, it has been shown that the epizootic was spread along
the lines of communication and that, in this connection, we have also
been able to ascertain the transportation of house rats and their fleas
in public means of transportation (railways, ships). We would also like
to mention that the different conditions in British India are at least
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free accessthe spread of bubonic plague over distances 169
partly due to the fact that there the flea numbers are much higher than
in Java.44
As the reader will have noted, the concept of metastatic spread appears
not only in the paper’s title, but also six times in the passages just cited
from van Loghem and Swellengrebel’s paper (other instances could
have been quoted). This suffices to show and to prove that this concept
is real and represents the same notion of the mode of spread by plague
over considerable distances that the IPRC and other plague researchers
call spread per saltum45 or spread by leaps.
This clarification is necessary to a discussion of Cohn’s misleading
assertions in his “review” of my monograph on the Black Death in The
New England Journal of Medicine:
Benedictow casts aside any rate of disease spread that was faster than he
likes: at these junctures, the Black Death made “metastatic leaps.” But
even with his various stratagems, his results still show the medieval
plague travelling 30 times as fast as the modern one—a discrepancy he
does not explain or even admit to.46
The meaning of the word ‘stratagem’ should be explained, in this case
according to Oxford Advanced Learner’s Dictionary of Current English:
“trick, plan or scheme to deceive sby [= somebody] (esp an enemy).”
This means that Cohn accuses me of doing my best to deceive col-
leagues, students and other readers who are interested in the Black
Death, and that at the heart of this alleged stratagem is the concept of
metastatic leaps. According to Cohn, then, this concept is not based on
scholarly studies and facts but is fabricated,47 in order to preserve by
dishonest means my conventional theory of the microbiological nature
of the Black Death as bubonic plague. Clearly, Cohn’s assertion is com-
pletely unfounded and untrue in every respect.
Cohn maintains that medieval plague spread thirty times faster than
modern bubonic plague. This assertion is obviously based on his false
assertion that bubonic plague does not spread by metastatic leaps. He
also asserts that
44
Van Loghem and Swellengrebel 1914: 460–1, 468, 470–80. My translation from
German.
45
See also Wu Lien-Teh 1926: 182, and 1936b: 204.
46
Cohn 2005: 1354–5.
47
See also above: 151 and fn. 3.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free access170 chapter four
Historians have realized since the work of Graham Twigg, in 1984, that
the Black Death and the subtropical Y. pestis traveled at vastly different
speeds. Even with the railway and the steamship, the 20th-century plague,
because of its dependence on the homebound rat, spread overland at
about 8 miles per year, whereas the contagious Black Death almost
equaled that speed per day.48
The only correct assertion in this citation is that Twigg also claims that
bubonic plague spreads only contiguously between adjacent rat colo-
nies. However, Twigg does not assert that the modern spread rate
according to this mechanism of dissemination was “about 8 miles per
year,” since he correctly refers to IPRC’s observation on the spread rate
of contiguous rat plague (see below),49 so the spread rate of 8 miles per
year is falsely attributed to Twigg by Cohn. The assertion that the yearly
spread rate of 20th-century plague was about eight miles or 12.9 km
per year is at variance with the studies of the IPRC and subsequent
researchers on plague epidemiology broadly presented above and also
with all standard works on plague: it is simply a fictitious figure. One
could also wonder how it might be possible that so many excellent
plague researchers, many of them with strong interests in the history of
plague, should collectively have overlooked the fact that the Black
Death and modern plague “travelled at vastly different speeds.”
Accurate information on the pace of contiguous spread of plague
among colonies of black rats was obtained in one instance by the IPRC,
namely 300 feet or 91.4 m in six weeks, corresponding to 792 m in a
year. The IPRC organized some large-scale projects in order to study
the mechanisms and process of the spread of plague over considerable
distances, what they called spread per saltum50 (which Dutch scholars
called metastatic spread). The observation of the territorial spread rate
between conterminous rat colonies was made in one of several quite
remote and isolated villages on the outskirts of “Bombay Presidency”
selected by the IPRC in order to study how plague arrived.51 It was
observed in Sion Koliwada, Sion village’s northerly part (see above)
after it had been evacuated by inhabitants following the discovery of a
dead rat and the occurrence of a case of human plague.52 Thus in
48
Cohn 2005: 1354–5.
49
Twigg 1984: 131–2.
50
IPRC 1907h, 1907i, 1907j.
51
IPRC 1907h: 799–873.
52
IPRC 1907h: 805–28.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free accessthe spread of bubonic plague over distances 171
Sion Koliwada the plague could not be spread further by the dynamics
of human agency,53 by persons picking up rat fleas in their clothing at
home or elsewhere or in locally purchased grain and redistributing
them among neighbours in a process that would establish new effective
centres of spread at a multiplicative rate, as would be the case in histori-
cal plague epidemics. The observation of spread between rat colonies
was an unplanned result of the experiment and occurred under highly
artificial circumstances.
The study corroborated (again) the by then well-established obser-
vation that plague is spread over short and intermediate distances by
persons carrying rat fleas in their clothing or luggage or to a distance
especially by transportation of certain types of merchandise by ship or
railway. The crucial part of the study of this specific case in Sion
Koliwada is that the IPRC investigated the origin of the infection there,
identified the person who had arrived with it and whence she had
arrived: “the infection of Koliwada was not of indigenous origin but
was imported, as it were by mere chance, from “Bombay City,” having
been brought by a woman who resided in an infected quarter of the
city.” In other words, plague arrived in Koliwada by a metastatic leap
trough the agency of a person coming from an infected locality. One
could also take interest in the relationship between the first find of a
dead rat in Koliwada and the subsequent first human death there in
plague, which shows that the woman had brought with her an infected
rat flea which had first sought out its natural host, a black rat, unleash-
ing an epizootic and soon the release of a swarm of increasingly hungry
rat fleas.54
Conspicuously, Cohn does not mention that the central objective of
the IPRC’s studies on the mechanisms of spread of plague around
Mumbai (and in the Punjab) was specifically to study metastatic spread
of plague, the type of plague spread which Cohn denies that the
Commission observed or found to be crucial in the epidemiology of
bubonic plague. Cohn can only make such a highly misleading and
unfair assertion as a basis for his attack on my monograph exactly
because he denies established fact, that bubonic plague spreads meta-
statically and that the usual vehicle of this mode of spread is infected
rat fleas.
53
IPRC 1907h: 827; Lamb 1908: 19.
54
IPRC 1907h: 805–7, 835.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free access172 chapter four
To my knowledge, none of the scholars who has studied plague epi-
demiology in the last hundred years has suggested that bubonic plague
spreads over land by movement of black rats between rat colonies, as
maintained by Twigg, Scott and Duncan and Cohn. The fact is that the
black “homebound” rat is exactly that, and the technical names for the
animal are the Latin-derived terms “commensal rats,” meaning rats
which are associated with the table, or “domestic rats,” meaning rats
which are attached to or associated with human housing. Rats do not,
or are loath, to move “over land” but are sedentary animals seeking
shelter in burrows and nests near their source of food. If possible, they
avoid the dangers from birds and beasts of prey inherent in crossing
open land, and for the same reason they are nocturnal animals, leaving
their nests or burrows for foraging at night.55 Scholars generally deny
that bubonic plague is spread by black rats by movement over land. The
IPRC never found infected black rats at any great distance from human
habitation. Only in Java has it been reported that black rats “may be
trapped almost a kilometre along the side of roads between villages.”56
Rats are inadvertently spread in goods by human agency, by ship,
railway, cart, or packhorse, or other means of transportation.
Cohn maintains that
He [Benedictow: my insertion] speeds up the 20th-century plague by
reporting infection times only for California, where the disease is carried
by the prairie dog, not the homebound rat, and has been known to move
as fast as 15 miles per year.57
The fact is that the geographical term California is never mentioned in
my monograph which can be easily ascertained in the “Index of
Geographical Names and Peoples,” and I have never stated anything to
this effect. In order to produce this spurious assertion, Cohn must con-
fuse the pace of expansion of plague foci by wild rodents in nature58
with the spread rates of rat-plague epizootics and plague epidemics
among human beings in their human social contexts, which makes his
disparaging comment obviously fallacious and unfounded. To my
55
Pollitzer 1954: 285–6; Wu Lien-The 1936b: 213–5. In the countryside, black rats
may move from their nests and burrows associated with farm houses and into the fields
when crops are harvested, later to return to their home base. This is, then, a highly
localized cyclical occurrence.
56
Hirst 1953: 305–6.
57
Cohn 2005: 1355.
58
See Benedictow 2004: 46–8.
O. Benedictow - 9789004193918
Downloaded from Brill.com04/13/2021 05:02:04AM
via free accessYou can also read
