大西洋奴隶贸易中奴隶和船员死亡率的决定因素（1）

CHATGPT翻译，因为图片扫描，可能部分漏译，不过应该不影响大致意思

## 来自你的消息:

NBER WORKING PAPER SERIES

DETERMINANTS OF SLAVE AND CREW

MORTALITY IN THE ATLANTIC SLAVE TRADE

Richard H. Steckel

Richard A. Jensen

Working Paper No. 15140

NATIONAL BUREAU OF ECONOMIC RESEARCH

1050 Massachusetts Avenue

Cambridge, MA 02138

January 1985

We have benefitted from the comments or help of Raymond Cohn,

Philip Curtin, David Eltis, Stanley Engerman, Roderick Floud,

Donald Haurin, Kenneth Kiple, Robert Margo, Jerry Thursby, and

seminar participants at Brigham Young University. Judy Collingwood

and Cheryl Zinker provided valuable research assistance The

research reported here is part of the NBER's research program in

Development of the American Economy. Any opinions expressed are

those of the authors and not those of the National Bureau of

Economic Research.

NBER Working Paper #1540

January 1985

Determinants of Slave and Crew

Mortality in the Atlantic Slave Trade

ABSTRACT

This paper measures and. analyzes death rates that prevailed in the

Atlantic slave trade during the late 1700s. Crew members died primarily

from fevers (probably malaria) and slaves died primarily from gastrointestinal

diseases. Annual death rates in this activity were 230 per thousand

among the crew and 83 per thousand among slaves. The lack of immunities

to the African disease environment contributed to the high death rates

among the crew. The spread of dysentery among slaves during the voyage

was probably exacerbated by congestion and. poor nutrition. Death rates

differed systematically by region of origin in Africa and. season of the

year. There was little interaction between the incidence of slave and.

crew deaths. The high death rates make the slave trade a demographic

laboratory for study of health and mortality ad an economic laboratory

for study of markets for free labor.

Richard H. Steckel

Economics Department

Ohio State University

Columbtis, Ohio 1132l0

(6lI) 422-50Q8

and

Richard A. Jensen

Economics Department

Ohio State University

Columbus, Ohio 132lO

(61)-i.) Li.229339

## 来自ChatAI的消息:

工作文件系列NBER

大西洋奴隶贸易中奴隶和船员死亡率的决定因素

Richard H. Steckel

Richard A. Jensen

工作文件编号15140

美国国家经济研究局

马萨诸塞州剑桥市马萨诸塞大道1050号

1985年1月

我们受益于Raymond Cohn、Philip Curtin、David Eltis、Stanley Engerman、Roderick Floud、Donald Haurin、Kenneth Kiple、Robert Margo、Jerry Thursby以及布里格汉姆青年大学的研讨会与研究人员的评论和帮助。Judy Collingwood和Cheryl Zinker提供了有价值的研究帮助。此处报告的研究是NBER发展美国经济研究计划的一部分。任何观点均属作者个人观点，而非美国国家经济研究局的观点。

NBER工作文件编号1540

1985年1月

大西洋奴隶贸易中奴隶和船员死亡率的决定因素

摘要

本文测量和分析了18世纪后期大西洋奴隶贸易中的死亡率。船员主要死于热病（可能是疟疾），而奴隶主要死于胃肠道疾病。在这项活动中，每年船员的死亡率为千分之230，奴隶的死亡率为千分之83。在船员中，由于缺乏对非洲疾病环境的免疫力，死亡率较高。奴隶在航行期间蔓延的痢疾可能受到了拥挤和营养不良的加重影响。死亡率在非洲起源地区和季节间有明显差异。奴隶和船员死亡率之间的交互作用很少。高死亡率使得奴隶贸易成为一个研究健康和死亡率的人口学实验室，也是研究自由劳动市场的经济实验室。

Richard H. Steckel

俄亥俄州立大学经济系

哥伦布, 俄亥俄州11321

(6lI) 422-50Q8

和

Richard A. Jensen

俄亥俄州立大学经济系

哥伦布, 俄亥俄州13210

(61)-i.) Li.229339

## 来自你的消息:

I NTRODUCTI ON

One of the early political expressions of opposition to slavery occurred

in the late 1700s when Parliament considered proposals to ameliorate condi tions in the slave trade. The treatment and mortality of slaves emerged as

central questions in deliberations and the debate led to collection and study

of evidence on conditions in Africa, the middle passage, and the New World.

Legislation in 1788 took the form of limits on the slave—carrying capacity of

vessels; the requirement that slave vessels carry a surgeon who cared for

slaves and who monitored conditions through a journal that noted deaths and

causes of death: and a system of bonuses to ship captains and surgeons.

Controversy over the health and mortality of persons involved in the

slave trade has persisted from the era of the campaign against slavery to the

present. The debate continues in part because the rhetoric, the charges and

counter-charges, and the tactics of the supporters and foes of the institution

often obscured the underlying realities. More fundamental to modern research,

however, are the facts that substantially limited information on these ques tions survives to the present and the levels and determinants of health and

mortality are central to larger questions such as the economic costs of the

slave trade to Western Europe, the profitability of slavery and the slave

trade, and the impact of legislation on the slave trade.

Studies of mortality in the Atlantic slave trade have relied extensively

on data from company records, newspapers, and government sources such as port

records, papers required by legislation monitoring or regulating the trade,

and the results of special inquiries.' The measure of mortality commonly con structed from these sources is the ratio of slave deaths to the number of

slaves embarked, although some authors have used the ratio divided by voyage

2

length multiplied by 1000.2 The central questions of research have included

mortality differences by region of departure from Africa, the effect of crowding on mortality, voyage length and the probability of death, the secular

trend in mortality rates, and the importance of vessels that had an unusually

high number of deaths.

## 来自ChatAI的消息:

引言

在18世纪末期，议会考虑了改善奴隶贸易条件的提案，这是对奴隶制度反对的早期政治表达之一。在讨论中，奴隶的待遇和死亡率成为核心问题，这场辩论促使人们收集和研究了有关非洲、中途海程和新世界的情况的证据。1788年的立法限制了船只承载奴隶的能力，要求奴隶船只携带一名医生负责照顾奴隶，并通过一份记录死亡和死亡原因的日志来监测情况，还设立了一套对船长和医生的奖励制度。

从反对奴隶制度的运动时期到现在，关于奴隶贸易中涉及的人员健康和死亡率的争论一直存在。这场辩论持续存在的原因部分是因为支持者和反对者之间的修辞、指责和反指责以及策略经常掩盖了潜在的现实。然而，对于现代研究来说，更重要的是，迄今为止，这些问题上存在着极为有限的信息，并且健康和死亡的水平和决定因素对于诸如西欧奴隶贸易的经济成本、奴隶制度和奴隶贸易的盈利能力以及立法对奴隶贸易的影响等更大的问题具有重要意义。

对大西洋奴隶贸易死亡率的研究在很大程度上依赖于公司记录、报纸和政府来源（如港口记录、监督或规范贸易所需的文件，以及特别调查的结果）的数据3。通常从这些来源构建的死亡率衡量标准是奴隶死亡人数与上船奴隶数的比率，尽管一些作者使用了这个比率除以航行时间乘以1000的结果2。研究的核心问题包括根据非洲离港地区的死亡率差异、拥挤对死亡率的影响、航程长度和死亡的概率、死亡率的长期趋势以及具有异常高死亡人数的船只的重要性。

## 来自你的消息:

This paper extends work in the area by analyzing 92 surgeons' logs

required by Dolben's Act of 178R. The ships in the sample originated from the

major supply regions in Africa and span the years 1792 to 1796. The excep tional value of the logs for measuring levels and studying determinants of

mortality stems from the combination of information on dates of death, causes

of death, and records for slaves as well as crew. Because dates of death and

the number of persons at risk are available, true death rates can be calcu lated. The death rates are relevant for comparisons with the mortality expe riences of other populations. The chronological record also permits study of

death rates by day during loading and during the voyage as opposed to study of

the journey as a whole that is common in previous work. Although the diagnos tic capabilities of ship surgeons were primitive by modern standards, the

symptoms of several diseases or disease categories common in the slave trade

were sufficiently distinctive to be useful. This combined with the systematic

enumeration of causes enables a statistical examination of causes of mortality

that is more detailed then previously undertaken. The mortality records of

the crew furnish a backdrop against which slave mortality can be compared and

cont rasted.

The analysis begins with death rates by cause of death, The results of

this inquiry suggest a plan of action for study of major causes of death

during loading and during the voyage. The next portion compares the findings

of this paper with previous work, and the last section considers the general ity of the results. The analysis generally accepts the data at face value;

3

the hazards of doing this and features of the data are discussed in the appendix.

DEATH RATES BY CAUSE OF DEATH

## 来自ChatAI的消息:

本文通过分析178R年Dolben法案要求的92位外科医生的日志扩展了该领域的工作。样本中的船只来自非洲主要供应地区，时间跨度为1792年至1796年。这些日志通过提供死亡日期、死因和奴隶以及船员的记录的信息的组合，为测量水平和研究死亡率的决定因素提供了异常的价值。由于有死亡日期和风险人数的数据，可以计算出真实的死亡率。死亡率与其他人群的死亡经历进行比较是相关的。时间记录还允许研究装载期间和航行期间的每天死亡率，而不是像以前的研究中那样研究整个航程。虽然与现代标准相比，船医的诊断能力是原始的，但在奴隶贸易中常见的几种疾病或疾病类别的症状足够独特以被使用。这与对死因的系统计数结合起来，使得对死亡原因的统计研究更加详细。船员的死亡记录为对比和对照奴隶死亡提供了背景。

分析从死因的死亡率开始。该研究的结果为研究装载期间和航行期间的主要死因提供了行动计划。接下来的部分将本文的研究结果与以前的研究进行比较，最后一节考虑了结果的普遍性。这项分析通常接受数据的表面价值；对此存在的风险以及数据的特点在附录中进行了讨论。

按死因分组的死亡率

## 来自你的消息:

Construction

The surgeon maintained a chronological journal of loadings and unload—

ings, usually in categories of men, women, boys, and girls, although sometimes

only males and females or men and women were enumerated. Slaves were frequently held on board until departure for the West Indies. Slaves unloaded

include those sold to other vessels and pawns (captives held as security

before transactions were completed) and sick slaves returned to shore.3 Slave

sales or transfers to other ships were extensive on a few vessels. A computer

program written for these data tabulates the number at risk each day from

first loading to the final destination in the West Indies.

Muster rolls establish the crew size at the time of departure from

England for 45 out of 92 vessels in the sample. Crew size for the remaining

vessels was estimated from a regression of crew size on tons.4 The available

muster rolls show that relatively few crew members left the ship or were taken

on board before the ship reached the West Indies. As an approximation it was

assumed that crew size changed only through deaths. Thus the estimated number

of crew at risk may be inaccurate for particular vessels.

Death rates were calculate from the chronological record of deaths and

data on the average number of person-years at risk during a particular time

period.5 The tables express results in annual rates per 1000 to facilitate

comparisons with other populations.

## 来自ChatAI的消息:

施工

该外科医生保持了对装载和卸载过程的时间顺序记录，通常按照男性、女性、男孩和女孩的类别进行分类，尽管有时只列举男性和女性或男人和女人。奴隶通常被留在船上，直到前往西印度群岛。卸下的奴隶包括出售给其他船只和抵押品（在交易完成之前作为抵押品拘留的俘虏），以及返回岸上的患病奴隶。一些船只进行了大量的奴隶销售或转移。为了这些数据编写了一个计算机程序，以确定从首次装载到西印度群岛的最终目的地之间每天处于风险状态的人数。

在样本中，通过船舶登记簿可以确定92艘船中有45艘在离开英国时的船员规模。其余船只的船员规模是根据吨位和船员规模的回归估计出来的。可用的船舶登记簿显示，在船只抵达西印度群岛之前，只有很少数量的船员离开船只或上船。作为近似，假设船员规模仅通过死亡而改变。因此，对于特定船只，估计的处于风险状态的船员人数可能不准确。

死亡率是根据死亡的时间记录和特定时间段内处于风险状态的人数的平均数计算出来的。表格以每1000人年的年度比率表示结果，以便与其他人群进行比较。

## 来自你的消息:

Medical Knowledge

The information available on casues of death must be approached cau tiously given the rudimentary medical knowledge of the late 1700s.6 The

4

difficulties stem from the lack of a germ theory of disease to guide inquiry

and from the poor state of medical instruments and procedures necessary for

precise diagnosis. In making a diagnosis that would guide treatment the surgeons drew upon visible symptoms of the patient. Unfortunately not all important symptoms of a particular disease were visible, the symptoms may have

changed over time as the disease ran its course, and some diseases with distinctly different causes have symptoms in common. The problems of identification were compounded if the patient had more than one disease.

Medical diagnosis of cause or causes of death is important for investigation of socioeconomic and epidemiological factors in mortality. Knowledge of

the disease or diseases involved conveys information about the net influence

of the prevalence or pathogens, mechanisms of transmission, and the susceptibility of individuals in the population. Unfortunately the surgeons recorded

few visible detals and their observations were often grouped under broad headings such as "fevers," "dysentery," and "diarrhea." The fevers no doubt

included malaria and probably yellow fever and the gastrointestinal diseases

no doubt included shigellosis (bacillary dysentery)and possibly amoebic

dysentery.7 However, fever sometimes acconpanies gastrointestinal diseases

and vomiting and diarrhea may acconpany malaria and yellow fever.8 Thus it is

not safe to conclude that "fevers" were confined entirely to diseases spread

by mosquitoes or that gastrointestinal diseases such as "diarrhea" were confined entirely to diseases spread by ingestion of contaminated matter.

## 来自ChatAI的消息:

医学知识

考虑到18世纪末的基础医学知识，对死因的信息必须谨慎处理。困难源于缺乏引导调查的病原体理论以及医疗仪器和诊断程序的不完善。在进行诊断以指导治疗时，外科医生主要依靠患者的可见症状。然而，特定疾病的一些重要症状可能不可见，随着疾病进展，症状可能会发生变化，而一些不同原因引起的疾病可能有共同的症状。如果患者有多种疾病，辨识问题就更加复杂。

了解死因或死因的医学诊断对于研究社会经济和流行病学因素对死亡率的影响至关重要。有关所涉及疾病的知识可以传达有关流行病原的净影响、传播机制以及人群中个体的易感性的信息。然而，外科医生只记录了少量可见细节，他们的观察经常被归为广泛的范畴，如“发热”、“痢疾”和“腹泻”。毫无疑问，“发热”可能包括疟疾，很可能还有黄热病，“胃肠道疾病”很可能包括志贺氏菌痢疾和可能的阿米巴痢疾。然而，发热有时会伴随胃肠道疾病，呕吐和腹泻可能会伴随疟疾和黄热病。因此，出于安全考虑，不能得出“发热”完全限于通过蚊虫传播的疾病或“腹泻”完全限于通过摄入污染物造成的疾病的结论。

## 来自你的消息:

Patterns

Tables 1—4 show the death rates calculated for slaves and crew during

loading and the voyage. Comparisons involving the overall rates given in the

last row of the third column of each table demonstrate that loading and the

5

voyage were extraordinarily hazardous to life. The rates are exceptional when

placed in the proper perspective of the probable age range of the slaves and

crew. Death rates are ordinarily highest during infancy, then decline during

childhood and reach a minimum just before or during adolescence, The rates

usually rise gradually for many years beyond the minimum and then accelerate

upwards in the 60s and beyond. Under ordinary circumstances most deaths were

confined to the very young and the old. Typical eighteenth—century Western

European populations in the age range 10—34, for example, probably had average

annual death rates not exceeding roughly 1 percent.9 Furthermore, the upper

limit of recorded experience for this age group when not living under severe

circumstances such as war or famine is less than 2 percent.10 The slaves and

crew faced death rates several times higher than their contemporaries engaged

in ordinary activities. The slave trade therefore provides a laboratory for

study of health and mortality under extreme circumstances.

The slave trade was relatively hazardous for the crew, particularly

during loading. Compared to slaves the death rates of the crew were more than

5 times higher during loading and nearly twice as high during the voyage. The

absolute levels of crew death rates were on the low to middle range of magnitudes reported by others for European populations residing in western Africa

during the eighteenth and nineteenth centuries.1' The surgeons logs support

the view that western Africa was the "white man's grave." The extraordinary

death rates for the crew raise interesting questions about the operation of

markets for free labor that engaged in the slave trade, Comparisons of the

slave trade with other types of voyages would add a dimension of perspective

to the slave trade and may provide insights into causes of mortality.

## 来自ChatAI的消息:

图表1至4显示了装载和航行期间计算出的奴隶和船员的死亡率。通过比较每个表的第三列最后一行给出的总体死亡率，可以看出装载和航行对生命极为危险。考虑到奴隶和船员可能的年龄范围，这些死亡率是异常的。死亡率通常在婴儿期最高，然后在童年期下降，在青春期之前或期间达到最低点。在最低点之后，死亡率通常在接下来的许多年逐渐上升，然后在60多岁及以上急剧加速。在普通情况下，大多数死亡都局限在幼儿和老年人中。例如，典型的十八世纪西欧10-34岁年龄范围的人口，平均每年的死亡率可能不超过1%。此外，当这个年龄组的人没有生活在战争或饥荒等严重环境下时，记录经验的上限不到2%。奴隶和船员面临的死亡率是从事普通活动的同时代人数倍之高。因此，奴隶贸易为研究极端环境下的健康和死亡提供了一个实验室。

相对而言，奴隶贸易对船员而言相对危险，特别是在装载期间。与奴隶相比，在装载期间船员的死亡率要高出5倍以上，在航行期间几乎是其两倍。船员死亡率的绝对水平处于报告的18世纪和19世纪西非居民的水平中等偏低范围内。外科医生的记录支持西非是“白人的坟墓”的观点。船员的异常死亡率对从事奴隶贸易的自由劳动力市场的运作提出了有趣的问题。将奴隶贸易与其他类型的航行进行比较，可以从不同角度观察奴隶贸易，并可以提供有关死亡原因的见解。

## 来自你的消息:

The chances of death changed considerably during the course of loading

and the voyage. mong slaves the safest period occurred about 3 to 6 months

6

before departure; the death rate at this time was 22.8 per thousand, or in the

neighborhood of twice the rate expected for settled populations not facing

severe circumstances. From this low the rate nearly tripled to 64.5 per thou sand during the four weeks before departure, continued to rise during the

early part of the voyage, and reached a peak of 145.9 per thousand during days

29—42 of the voyage. Slave death rates followed an inverted—U shape during

the voyage and differed by a factor of roughly 2 to 1. Compared to slaves,

crew death rates were less volatile during loading and reached a plateau of

nearly 350 per thousand in weeks 5-12. Over three-quarters of all crew deaths

during the voyage occurred within the first four weeks.

During the operations of loading and the voyage over 60 percent of the

crew deaths were from fevers and over 40 percent of the slave deaths were frOm

gastrointestinal diseases. The share of crew deaths from fevers was slightly

higher during loading (68%) compared to the voyage (54%). Among slaves thQ

shares of deaths from gastrointestinal diseases were approximately constant

during loading (38%) and the voyage (41%). Gastrointestinal diseases claimed

16 percent of crew deaths and fevers amounted to 8 percent of slave losses.

The minor causes of slave deaths included respiratory diseases (5%), sui cide (3%), scurvy (2%), insurrection (2%), and accidents. Among crew deaths 8

percent involved accidents (primarily drowning), 2 percent scurvy, 1 percent

respiratory, and 1 percent insurrection. The fotheru category includes well—

defined but relatively infrequent ailments such as yaws and dropsy, several

poorly—described diseases (e.g. sudden death," "sulks," "fits," "delirium,"

and "relaxed"), and many illegible entries. Most of the slave deaths from

respiratory diseases (89%) and scurvy (88%) occurred on the voyage, particu larly near the middle and the end of the trip, which is consistent with claims

7

and evidence that the voyage was debilitating. The low death rates from

scurvy suggest that measures sometimes taken as early as the late eighteenth

century to include vitamin C in the diet through foods such as lime juice may

have been used.12 Information on exact dates of death available in the logs

indicates that slave suicides were concentrated during or shortly after

slaves were first loaded and at the time the ship left Africa.

## 来自ChatAI的消息:

在装载和航行过程中，死亡机率发生了显著变化。在奴隶群体中，最安全的时期是在出发前的3到6个月；此时的死亡率为每千人中22.8人，大约是非受严峻条件影响的定居人口的两倍。从这个低点开始，死亡率在出发前的四周内几乎增加了两倍，继续在航行初期上升，到达航行的第29至42天期间达到每千人中145.9人的峰值。奴隶的死亡率在航行过程中呈倒U型变化，差异大致为2比1。与奴隶相比，船员的死亡率在装载期间波动较小，并在第5至12周达到每千人近350人的稳定水平。航行期间，四分之三以上的船员死亡事件发生在头四周内。

在装载和航行过程中，超过60%的船员死于发热病，超过40%的奴隶死于胃肠疾病。与航行相比，装载期间船员死于发热病的比例略高（68%对54%）。奴隶死于胃肠疾病的比例在装载（38%）和航行（41%）期间几乎保持不变。胃肠疾病占船员死亡的16%，而发热病占奴隶死亡的8%。

奴隶死亡的其他原因包括呼吸道疾病（5%）、自杀（3%）、坏血病（2%）、起义（2%）和事故。船员死亡中8%是事故（主要是溺水）、2%是坏血病，1%是呼吸道疾病，1%是起义。其他原因包括定义明确但相对较少见的疾病，例如鸡麻风和水肿，以及许多无法辨认的记录。大多数呼吸道疾病（89%）和坏血病（88%）导致的奴隶死亡发生在航行期间，特别是在航行中期和结束时，这与航行具有消耗性的特点相符。由于坏血病的死亡率较低，这表明可能在18世纪末期就开始采取措施，如在饮食中添加维生素C（例如酸橙汁）。记录中关于死亡确切日期的信息表明，奴隶的自杀行为主要集中在奴隶首次被装载和船只离开非洲的时期。

## 来自你的消息:

It is important to note that smallpox and measles were virtually absent

as causes of death in these data. Inoculation against smallpox was practiced

in England as early as the mid 1700s, and the availability of the technology

suggests that it may have been used on slaves.13 It is also possible that

native Africans practiced inoculation.14 Because the logs pertain to only to

a few years (1792-1796), it is also possible that epidemic conditions for this

disease were temporarily favorable.

Implications

Because the underlying causes and methods of transmission of many

diseases are quite different, the investigation of determinants of mortality

will be undertaken by disease category. The data in Tables 1-4make it clear

that fevers and gastrointestinal diseases should be the focus of attention.

Other causes of death are relevant to the entire picture, but their frequency

of occurrence——even for the most important minor causes such as accidents

among the crew and respiratory diseases among slaves-—is so low that syste—

matic study is not feasible with these data.

The configurations of death rates by day of loading, day before depar ture, and day of voyage suggest that determinants of mortality substantially

involved a time—dependent process. The importance of time is easily seen for

gastrointestinal diseases involving slaves; death rates more than doubled from

8

the beginning to the end of loading and continued to climb, reaching a peak

during the third to the fifth week of the voyage. The pattern of death rates

resembles the beginning, height, and end of an epidemic.

The concepts of agent, host, environment, and incubation period as

employed by epidemiologists are useful for understanding mortality from dysentery.'5 The spread of an infectious disease depends upon the probability of

contact between an infectious agent and individuals, and the susceptibility of

individuals. The amount of infectious material in the environment is a function of the number of infected individuals or hosts of the agent, the duration

and extent to which infectious material is expelled by hosts3 the rate of survival of the agent in the environment, the route of entry into the host, and

the existence of alternative reservoirs or hosts of the agent. The availability of hosts varies with the mobility and interpersonal contact within the

population group, and the immunityor other means of resistence to the agent

by individuals. The incubation period is the interval of time between contact

with an infectious agent and the onset of illness. Each disease has a characteristic incubation period, which may vary among individuals. The discussion

so far has assumed a situation with case—to-case transmission of disease, but

the concepts can easily be extended to include the spread of infection from a

central source such as contaminated food or liquid.

## 来自ChatAI的消息:

需要注意的是，在这些数据中，天花和麻疹几乎没有作为死因出现。英国在18世纪中叶就开始实行天花接种，技术的可用性表明它可能被用于奴隶身上。同时，土著非洲人也有可能实行接种。由于这些日志只涉及到几年（1792年至1796年），因此这种疾病的流行状况在这段时间内可能是有利的。

影响

由于许多疾病的根本原因和传播方法相当不同，对死亡率决定因素的研究将按疾病类别进行。从表格1-4可以看出，发热和胃肠道疾病应该是关注的焦点。其他死因对整个情况也有影响，但是它们的发生频率，即使对于最重要的次要原因，如机组事故和奴隶呼吸道疾病，也是如此低，以至于这些数据不适合进行系统研究。

装载日、出发前一天和航程中的死亡率配置表明死亡率的决定因素在很大程度上涉及时间相关的过程。从奴隶在装载开始到结束期间，胃肠道疾病的死亡率增加了一倍以上，并继续上升，在航行的第三到第五周达到了峰值。死亡率的模式类似于一个流行病的开始、高峰和结束。

流行病学家使用的代理、宿主、环境和潜伏期等概念对于理解痢疾死亡率很有用。传染病的传播取决于传染剂与个体之间的接触概率和个体的易感性。环境中传染物质的数量取决于感染个体或传染剂的寄主及其排泄传染物质的持续时间和范围、传染剂在环境中的存活率、进入宿主的途径以及传染剂的替代储存库或宿主的存在。寄主的可用性因人群内部的移动性和人际接触而有所不同，个体的免疫或其他抵抗传染剂的手段也如此。潜伏期是接触传染剂和发病之间的时间间隔。每种疾病都有特定的潜伏期，可能在个体之间有所不同。到目前为止的讨论假定了一种情况，即病例间的传播，但是这些概念很容易扩展到包括从一个中心源（如被污染的食物或液体）传播感染的情况。

## 来自你的消息:

The epidemiology of malaria and yellow fever involve intermediate hosts.

The parasites responsible for these infections spend part of their life cycle

inhabiting mosquitoes and so the path of transmission is from man (or other

host) in his infectious stage, to mosquito, and subsequently from mosquito to

man. These diseases therefore involve populations of mosquitoes and humans

(or other hosts), each with its own set of susceptible and infective hosts.

9

African natives were ordinarily infected with malaria as children, and those

who survived had a limited immunity to further symptoms, but could act as

hosts.'6 Thus malaria was hyperendemic in coastal areas of western Africa,

and the spread of the disease among European newcomers resembled a stochastic

version of contamination from a central source. Unlike malaria, yellow fever

kills the victim within 5 to 7 days or follows by rapid recàvery and life—long

immunity that prevents the individual from acting again as a host. The yellow

fever parasite occasionally died out in parts of .Africa, particularly in iso

lated areas, but the chances of an epidemic increased as the nonimmune popula tion grew and as contacts with the outside carried the disease. Europeans may

have contributed to periodic yellow fever epidemics as noninimune hosts.

The spread of infectious disease is a complex process that can be summa rized using mathematical models.'7 In a simple deterministic model of an epi demic, for example, disease spreads by contact between infectives and suscept ibles, there is no removal from circulation by death, recovery, or isolation,

and infections remain contagious throughout the epidemic. The epidemic curve

shows the rate at which new cases accrue over time under the conditions or

parameters specified by the model. The time pattern of deaths is derivable

from the epidemic curve by incorporating magnitudes for the share of those

infected who fail to survive and the delay from the onset of symptoms to

death. More complex models can involve infection and removal, intermediate

hosts, geographical spread, stochastic phenomena, latent and infectious peri ods, and lost immunity. The a priori formulation of a model appropriate for

slaves or crew would be difficult in part because many conditions relevant to

the model are unknown. Indeed, the relevant conditions are an important

object of investigation. Nevertheless, an epidemic model is a useful frame work for thought. Some of the parameters in such a model were given and

10

others were under the control or influence of slave traders, ship captains,

and surgeons. The extent and nature of control are topics for the next section of the paper.

## 来自ChatAI的消息:

疟疾和黄热病的流行病学涉及中间宿主。引起这些感染的寄生虫在其生命周期的一部分中寄生在蚊子身上，因此传播路径是从人类（或其他宿主）的传染阶段到蚊子，然后再从蚊子传给人类。因此，这些疾病涉及蚊子和人类（或其他宿主）的人群，每个人群都有自己一组易感和感染性宿主。非洲原住民通常在儿童时期感染疟疾，幸存下来的人对进一步出现症状具有有限的免疫力，但可以作为宿主。因此，疟疾在非洲西部沿海地区是高度流行的，而在欧洲新来者中的疾病传播类似于从中心源头进行污染的随机版本。与疟疾不同，黄热病会在5到7天内杀死患者，或者出现快速康复和终生免疫，从而防止个体再次充当宿主。黄热病病原体有时会在非洲的某些地区消失，尤其是在孤立的地区，但随着免疫不全人口的增长和与外界的接触，流行病的机会增加。欧洲人可能是黄热病周期性流行的非免疫宿主的原因之一。传染疾病的传播是一个复杂的过程，可以用数学模型来概括。例如，在一个简单的确定性流行病模型中，疾病通过感染者和易感者之间的接触传播，没有通过死亡、康复或隔离从环境中消除，感染在整个流行病期间都具有传染性。流行病曲线显示了在模型指定的条件或参数下随着时间积累的新病例率。死亡的时间模式可以通过将被感染者中未能幸存的比例和从症状发作到死亡的延迟纳入流行病曲线来推导。更复杂的模型可以涉及感染和消除、中间宿主、地理传播、随机现象、潜伏和传染期，以及失去免疫力。针对奴隶或船员的先验模型制定在一定程度上会存在困难，因为对于模型的许多相关条件是未知的。事实上，相关条件是研究的重要对象。然而，流行病模型是一种有用的思维框架。这样一个模型中的一些参数已经给出，而另一些参数则受到奴隶贩子、船长和外科医生的控制或影响。控制的程度和性质是本文的下一节的主题。