Barbara J. Shapiro *
Testimony in seventeenth-century English natural philosophy: legal origins and early development
Studies In History and Philosophy of Science Part A
Volume 33, Issue 2, June 2002, Pages 243-263
This essay argues that techniques for assessing testimonial credibility were well established in English legal contexts before they appeared in English natural philosophy. ‘Matters of fact’ supported by testimony referred to human actions and events before the concept was applied to natural phenomena. The article surveys English legal views about testimony and argues that the criteria for credible testimony in both legal and scientific venues were not limited to those of gentle status. Natural philosophers became concerned with testimony when they shifted their attention from universal statements about nature to particular natural and experimental events. Testimony thus became important in the construction of natural and experimental histories constructed by English naturalists. The shift to a more Baconian approach to natural investigation, itself shaped in part by legal concepts and practice, made it possible for members of the Royal Society to adopt an already familiar and societally approved approach to testimony. However, the essay also suggests how the use of scientific instruments and the desire to avoid the adversarial processes of the law modified legal conditions for fact determination, and made it possible for later generations to associate the concept of fact, supported by credible testimony, with the natural rather than the human sciences.
The subject of testimony and the related topic of credibility have recently become a central topic in the history and philosophy of science. From at least the mid seventeenth century, English Baconians linked observation and experiment to testimonial credibility. Reliance on witnessing, criteria for scientific testimony and a discourse of ‘fact’ emerge in English natural philosophy simultaneously in a new emphasis on the careful reporting, verifying and recording of particular natural and experimental events. Steven Shapin and Simon Schaffer (1985) have suggested that the credit attributed to the testimony of gentlemen was essential to the construction of Restoration-era scientific ‘matters of fact’. They argue that Robert Boyle epitomized the knowledge-making capacities of high social status, and that Robert Hooke’s credibility was impaired by his relatively low social status (Shapin & Schaffer, 1985). Shapin also suggests that this Restoration gentlemanly credibility had its origins in the norms of Renaissance courtly society ( Shapin and Shapin).
Intellectual historians and historians of science are inclined to examine distinct disciplinary traditions, such as science, law or art. There have, however, been a series of interdisciplinary developments. It now seems obvious that one cannot understand many aspects of early modern natural philosophy without giving serious attention to religion and theology. Most historians of science recognize that other social aspects of scientific development also require some attention. Earlier studies emphasized connections between capitalism, bourgeois values and the scientific revolution (Zilsel; Hill; Merton; Jacob and Jacob). More recent ones have concerned themselves with courtly norms and their influence on intellectual life. However fruitful these developments have been, they have resulted in a tendency to underestimate the role of other societal institutions and norms.  If earlier historians were too inclined to associate the development of early modern science with the norms of Puritanism and a growing capitalistic economy, current historians overemphasize the role of aristocratic and gentlemanly norms.
I argue here that the issues of credible testimony now seen as critical to seventeenth-century English natural philosophy first became relevant in the context of law and that the criteria for evaluating testimony were well established in English legal contexts long before they appeared in English science. To understand how and why credible testimony became such an important part of the empirical natural philosophy of the seventeenth century it is necessary to focus on the way in which legal notions of witnessing and credibility were adapted to the investigation of natural phenomena.
While by no means all societies develop clearly defined norms for determining the ‘facts’ of nature, most, if not all , have developed some means of resolving disputes arising out of human transactions. Those means necessarily include ways of handling disputes about who said what and did what to whom.
Credibility of courtroom testimony was not an issue in the
early Middle Ages when ordeals of various kinds were employed to determine
legal issues. Credibility became
central in the later medieval era when ecclesiastics and lay rulers sought a
more rational foundation for legal decision-making.
At this point law drew on the Roman
rhetorical tradition transmitted to medieval scholars through the writings of
Canon law quickly developed criteria to evaluate ‘matters of fact’, based on the testimony of witnesses who had themselves seen or heard the act in question.  From the time of Tancred onward, canon lawyers adapted classical rhetorical criteria to ecclesiastical law, and these evidentiary criteria were then transmitted to the rapidly developing Romano-canon law (Donahue, 1981, p. 31; Gilissen, 1965). The distinction between ‘matters of fact’ and ‘matters of law’ came to play a central role in both continental and English law.
In the Romano-canon system, ‘facts’ might be fully proved either by confession of the accused or the testimony of two credible witnesses. Credibility was determined by criteria such as gender, age, social status and education. The nature of the witness’s friends and associates and his life style were relevant. Determinations of both fact and law were in the hands of highly educated, professional judges, who employed these criteria to evaluate credibility. The testimony of one credible witness, which constituted a ‘half proof’, was sufficient to trigger judicial torture to elicit a confession constituting ‘full proof’. Considerations of testimonial evidence were thus central to the administration of Romano-canon law.
English ecclesiastical courts utilized much the same body of evidentiary principles and procedures as their continental counterparts. The common law courts, which covered most civil and criminal causes, developed differently. ‘Matters of law’ were in the hands of experienced judges, while ‘matters of fact’, that is, the truth of particular past human acts, were determined by lay juries.
Criteria to govern the jury’s role in fact-determination, however, evolved slowly, because jurors were first treated as ‘knowers’ of fact rather than evaluators of the testimony of others about facts. Initially jurors were local residents who themselves knew who had done what to whom. Only when witnesses become commonplace in common law courts in the sixteenth century did juries begin to evaluate their testimony for accuracy, truthfulness and credibility. From that time onward testimony and its problems would become familiar to all those who participated in jury trials, or heard or read about them. 
Witnesses, except those for the defense in criminal trials, testified under oath, an incentive, though never a complete guarantee, of truth-telling. Jurors also evaluated the credibility of those who did not testify under oath. Felons were excluded, their oaths being of ‘no weight’, madmen, children and idiots for ‘want of skill and discernment’. 
Determination of witness credibility became central to the criminal trial with the growing use of witnesses in the common law courts.  Francis Bacon at the beginning and Sir Mathew Hale at the end of the seventeenth century agreed that the law left ‘the discernments of credit of testimony wholly to the juries’ consciences and understanding’. Hale noted ‘It is one thing whether a witness be admissible to be heard, another thing, whether they are to be believed when heard’. Although jurors were to consider witnesses as ‘honeste, good and indifferent, till the contrary be shown’, they were ‘judges of the fact, and likewise of the probability or improbability, credibility or incredibility of the witnesses and the testimony’.  Early in the sixteenth century juries were already considering how ‘lyghtness of mynde, hatred or corruption of money’ affected credibility. Somewhat later, Hale, the most respected jurist of his day, indicated jurors were to judge the ‘Quality, Carriage, Age, Condition, Education and Place of Commorance of Witnesses’ in giving ‘more or less Credit to their Testimony’. Judge Francis North also emphasized considerations of reputation, ‘inclinations’ and ‘education’. Thomas Hobbes noted that the evaluation of testimony involved both ‘the saying of the man’ and ‘his virtue’. Moral status and reputation were related to, but not identical to, social and economic status. Assumptions about reliable testimony corresponded roughly to the existing social hierarchy. Independent persons counted for more than dependents, servants, wives or the poor.
Societal inclination to trust those of high status, however, was undermined in a variety of ways. Testimony by anyone with a pecuniary interest, however small, was excluded from civil cases. Gentlemen engaged in countless lawsuits that pitted the word of one gentleman against that of another. So juries frequently had to evaluate the testimony of disagreeing gentlemen. Partiality, interest, a poor reputation or a dissolute life style might undermine gentle status. Gentleman defendants accused of felony, no more than lower-class defendants, were permitted to testify under oath. Most of the jurors themselves were of ‘middling’ status. Their own life experience surely had taught them that gentlemen did not always tell the truth. Social status was one, but only one, factor in assessing witness credibility, and the judges of that credibility typically were not gentlemen.
Expert witnesses were fairly rare, but skill and experience were relevant to credibility because ‘Discernment must arise from the skill, and will appear from the reasons and accounts they give of their knowledge’. Those giving ‘plain and evident marks of their knowledge’ were more credible. Although common law courts were not bound by the two-witness rule, Hale and later Locke stated a preference for multiple witnesses. Numbers, however, were not always decisive. Refusals to indict or convict might occasionally occur in the face of the testimony of twenty or more witnesses. According to Hale, juries are entitled to ‘disbelieve what a Witness swears’ and ‘may sometimes give Credit to one Witness, tho’ opposed by more than One’. They might give a ‘verdict contrary to such Testimonies, the Truth whereof they have just Cause to suspect, and may and do often pronounce their verdict upon one single Testimony’. 
Prevailing common-sense belief also bore on the credibility of testimony. The ‘Probability of [th]e Matter’ sometimes ‘overthrows’ testimony. ‘For if the fact be contrary to all manner of experience and observation’ it was ‘too much’ to receive on the oath of a single witness. Inconsistent testimony also removed the witness ‘from all credit, for things totally opposite cannot be believed from the attestation of any man’. 
Impartiality and disinterestedness were important norms of both the Romano-canon and common law systems. The common law attempted to reduce or eliminate partiality and bias in jurors, judges and witnesses. If we consider the courtroom - as others have considered the scientific experiment - as a site of knowledge-making, that is, an environment where participants engaged in creating or determining the ‘truth’ of something by a set of site-specific rules, we can see that juries, judges, witnesses and counsel participated in a process designed to produce ‘morally certain’ determinations of ‘matters of fact’. As ‘judges of the fact’, jurors were thought in most instances to be capable of producing a just and true knowledge of such facts (Shapiro, 2000, pp. 30-3). These assumptions were best spelled out by Sir Geoffrey Gilbert, author of The law of evidence (1754) who argued that although first-hand sense data was the best source of knowledge, it was of limited usefulness to courts because the law dealt with ‘transient things’ of no ‘constant being’. When we ‘cannot see or hear anything ourselves, and yet are obliged to make a judgement of it’, it is necessary to ‘see and hear by report from others’. It was reasonable to give ‘faith and credibility… to the honesty and integrity of credible and uninterested witnesses, attesting any fact under the solemnities and obligations of religion, and the dangers and penalties of perjury’. When these conditions are met, the mind ‘equally acquiesces as on a knowledge by demonstration, for it can not have any more reason to be doubted than if we ourselves had heard and seen it’ (Gilbert, 1754, pp. 2, 4).  Some alleged matters of fact could be believed; others were ‘doubtful’ or false. The legal system could not operate with an extreme skepticism. It assumed that many ephemeral facts of human action could be established with a high degree of certitude and that ordinary persons of ‘middling’ status could evaluate testimony for credibility in order to arrive at impartial, truthful verdicts.
Substantial numbers of the middling classes participated in or were spectators of trials at common law (Cockburn & Green, 1988). Assize and quarter sessions attracted large audiences. A substantial portion of the English population was familiar with the practice of witnessing, the assessments of witness credibility, and the role of ordinary people in evaluating testimony.
Many leaders of opinion in English society became familiar with the treatment of testimony through the office of justices of the peace, whose many duties included pre-trial examination of criminal suspects and witnesses. From the sixteenth century onward, justices were instructed in these tasks by widely circulated manuals containing criteria for the evaluation of testimony including the condition, age, sex, discretion, fortune, reputation, life style and fidelity of those giving testimony. Edmund Bohun’s Justice of the Peace (1694) reminded readers that all ‘matters of fact’ required testimony. Belief in testimony was ‘stronger or weaker’ according to the ‘Credibility of the Person that relates it’. 
The distinction between first-hand or eye and ear witness testimony and second-hand or hearsay testimony was a familiar one, though hearsay testimony was not always excluded. Oaths to tell the truth taken by witnesses and jurors were part of the fact-establishing process, although it was recognized that witnesses might lie or make errors and that jurors might give false judgements. Jury verdicts could never be proved with the certitude that early modern persons associated with the terms ‘science’ or ‘mathematical certainty’. The language employed to indicate the highest kind of probability available in ‘matters of fact’ was ‘a satisfied conscience’, ‘moral certainty’ and later ‘beyond reasonable doubt’ (Shapiro and Shapiro). The ability to reach such certainty on the basis of credible testimony formed a fundamental and familiar part of English culture.
Legal concepts and norms were adopted in several venues prior to, and simultaneously with, their adoption by empirically oriented naturalists. Like jurists, historians frequently invoked the value of first-hand observation and credible testimony, discounted hearsay accounts and stressed the need for impartial witnesses and disinterested judgment (Shapiro, 2000, pp. 34-63).
Similar evidentiary conventions characterized the rapidly developing genre of travel writing. Here again we find numerous references to careful, first-hand observations recorded by credible observers claiming impartiality. Because travelers often lacked witnesses to substantiate their claims, and frequently reported strange and previously unknown phenomena, travel reporters were sometimes suspected of exaggeration and even lying. Soon after its formation, the Royal Society began to sponsor and support the gathering of information from travelers in order to provide new knowledge or correct the errors of former accounts. The Society insisted on reports from reliable sea captains or other credible travelers with first-hand knowledge. Second-hand reports, sometimes the only ones available, were to be scrutinized with special care. During the Restoration era methods for reporting from exotic locales came to be treated as one aspect of general scientific methodology. 
Recording and disseminating ‘news’ also employed notions of
credible testimony. Printed news,
which began in
Legal practices, historical writing, travel reporting and news were discourses of fact. Adoption of the concept of ‘matter of fact’, with its emphasis on first-hand observation and credible testimony, by seventeenth-century empirically oriented naturalists meant that the scientific community also came to participate in the ‘discourse of fact’. That development was facilitated by the long-standing employment of the discourse in law and history and its introduction in travel and news reporting. Proof by credible witnesses’ testimony was familiar to the ‘new philosophers’ as it was to other Englishmen and obviously applicable to their endeavors.
It was Francis Bacon whose works, beginning with the Advancement of learning in 1605, promoted and popularized the new empirical and experimental natural philosophy. Bacon’s natural philosophy would be based on natural histories collected by credible observers. The Baconian vision rejected traditional natural history, with its fictional as well as real animals, plants and other natural phenomena (Ashworth, 1990). It also rejected much of the Aristotelian approach to natural philosophy, which neither emphasized the role of closely observed particulars nor concerned itself with artificial, that is, man-made, effects. Baconian natural history, which comprehended both the natural and the artificial, was both broader in scope than its predecessors and more precise about its observational procedure.
Given Bacon’s preeminent legal status and his vast legal knowledge, it is not surprising that he applied legal concepts to the observation of natural phenomena.  It was an easy step to apply to natural phenomena the criteria so long used in establishing past human acts with a reasonable degree of certainty. For Bacon, natural history ‘treats of the deeds and works of Nature, civil history those of men’. The ‘deeds and works’ of both concerned particulars ‘circumscribed by time and place’. Bacon condemned the then current natural history as relying on ‘rumour and vague fames’ or the ‘gossip of the streets’ instead of ‘the weight of lawful evidence’. Nothing was being ‘duly investigated, nothing verified, nothing counted, weighted or measured’. Naturalists gave only a ‘glance or two upon facts and examples and experience’ before invoking ‘their own spirits to give them oracles’. The new natural history would dwell ‘purely and constantly on the facts of nature’. He rejected the ‘emblematic world view’, in which a melange of symbols, correspondences, observations, proverbs and fables formed a complex web of verbal and natural associations (Ashworth, 1990). His natural history would be compiled ‘as if every particular were stated upon an oath’.  Credible eyewitness testimony was essential, but experiments, which were more subtle than immediate sense observation, as well as sense-enhancing instruments, were especially encouraged. The collected observations of virtually all natural phenomena observed or experimentally created would then be sifted and examined to eliminate error and establish the facts of nature. Although Bacon rarely used the Latin word ‘factum’, he readily employed legal approaches to establishing belief in particular human deeds and the ‘deeds and works’ of nature, as well as to man-made experiments. Although somewhat suspicious of ‘wonders’, Bacon, like the news reporter, wished to take note of the anomalous and the accidental. Baconian natural history also included well certified histories of trade.
According to Bacon the credit of all things observed must be evaluated as ‘certainly true, doubtful whether true or not, certainly not true’. The doubtful must be reported in language suggestive of its tentative nature, such as ‘it is reported’, or ‘I have heard from a person of credit’. In important matters
the name of the author should be given… whether he took it from report, oral or written,… or rather affirmed it of his own knowledge; also whether it was a thing which happened in his own time or earlier; and again whether it was a thing of which, if it really happened, there must needs have been many witnesses; and finally whether the author was a vain-speaking and light person or sober and severe; and the like points, which bear on the weight of the evidence. 
Bacon’s criteria did not include social status, and he assumed artisans would be consulted in compiling histories of trade, suggesting that experience and expertise were appropriate bases of credibility at least for some observations.
Bacon’s application of the historico-legal methods of determining human action to natural ‘facts’ opened the way for the next generation of empirically oriented naturalists. Thereafter in natural history, as in law, matters of fact were properly proven through testimony by first-hand witnesses of suitable credibility.  This approach would become central to the program of the Royal Society.
Several points must be underlined. The first is that credible witness testimony became important to natural philosophers when their attention shifted from universal statements about nature to statements about particular natural and experimental events. Observation was not absent from Aristotelian natural philosophy but was not its central feature. Cartesian natural philosophy also employed experiment and observation, but typically to provide examples of natural principles rather than as the basis for constructing those principles (Dear, 1995).
The second point concerns natural history and natural philosophy. Natural histories were collections of well established matters of fact. Natural history is a term currently far more circumscribed than in the seventeenth century, when it covered a wide range of matters of fact, some observed personally, some credibly reported and some created experimentally. Natural history covered most of the natural sciences and included aspects of technology, agriculture and facets of anthropology. Knowledge of principles and causes, that is natural philosophy, not the collection of facts itself, however, remained the final goal. Legal approaches to testimony were appropriated by the new empirically grounded natural philosopher in constructing natural histories. Once facts were established, the construction of principles, explanations and causes proceeded by means having little to do with courtroom techniques necessarily adapted to the discovery of the particular not the general.  The impact of the law was important but limited.
We now turn to the research program of the Royal Society, founded early in the Restoration era. First we examine the programmatic statements relating to testimony of three men: Thomas Sprat, author of the Society-commissioned History of the Royal Society, Joseph Glanvill, another Royal Society publicist, and Henry Oldenburg, Secretary of the Royal Society and founder of the Philosophical Transactions.
Observed and experimentally produced ‘matters of fact’ became central to the research agenda of the Royal Society (Shapin & Schaffer, 1985). Scientific ‘facts’, like legal facts, were established primarily by witnesses whose testimony would be evaluated on the basis of a set of legally derived criteria of credibility, such as opportunity, ability, probity, skill, fidelity, social status, experience and reputation.
Sprat, Glanvill and Oldenburg, spokesmen and propagandists for the Royal Society, all insisted that observation and experiment were the core of the Society’s agenda. For Sprat the new philosophy was grounded on matter of fact, that is, on natural history based on observation and experiment. With some exaggeration he suggested that the Society’s members ‘only deal in matters of Fact’. Although the Society privileged first-hand reports over hearsay or second-hand experience, it was often ‘forced to trust the reports of others’. Sprat described the Society’s primary work as ‘Directing, Judging, Conjecturing, Improving and Discoursing upon experiments’. Witnessing by substantial numbers of persons would give credence to the observed experiments. The Society’s judgment on matters of fact, that is, on what had been observed, was based on the ‘concurring testimony of its members’. He characterized members as impartial judges before whom reports were given and judgments made, and described members as accepting or rejecting ‘matters of fact’.  Seventeenth-century usage of ‘matter of fact’ was often closer to what in modern usage would be called an ‘assertion of fact’. Thus for Sprat it would not have seemed self-contradictory to speak of a ‘false matter of fact’. Some matters of fact thus might be true and others doubted or rejected. 
Joseph Glanvill also employed the terminology of ‘matter of fact’ to characterize the virtuosi’s focus on ‘plain Objects of Sense’, which provided the greatest certainty. This certainty often came from testimony, testimony that, under the best circumstances, resulted in ‘indubitable assent’. For such assent to occur, reporters must be disinterested, their reports full and acceptance general. Fortunately, the Royal Society contained many such members, men of ‘wit and Fortune… where fondness of preconceived opinions, sordid Interests, or affectation of strange relations, are not like to render… reports suspect or partial, nor want of Sagacity, Fortune, or care, defective’. Reports of their ‘Tryals’ could therefore be received as ‘undoubted Records of certain events’. Glanvill pointed out the need to proportion assent to the ‘degree of the evidence’ and to be confident ‘only in those observations distinctly and clearly apprehended’. Though error was always a possibility, matters of fact well proved ought not to be denied. 
Oldenburg, who provided statements of the Society’s goals and methods in his vast foreign and domestic correspondence and in the Philosophical Transactions, similarly wrote of building a natural philosophy from ‘a large and truthful natural history’ derived from careful observation and the faithful recording of what had been observed.
In recent years two Restoration virtuosi, Robert Boyle and Robert Hooke, have become central to discussions of English science. Boyle, as Shapin and Schaffer point out, played an important role in publicizing the concept of matter of fact (Shapin & Schaffer, 1985). Like the Royal Society publicists and popularizers, Boyle believed that natural history was the ‘only sure Foundation of Natural Philosophy’. That foundation consisted of matters of fact, that is, experiments and specific observations ‘faithfully made and deliver’d’. Boyle frequently invoked legal terminology, referring at one point to ‘testimony of nature’. He referred to ‘judicious and illustrious witnesses’ and at another point noted that ‘matters of fact ought to be brought to trial’. Knowledge of fact was derived from sense and was communicated to others by means of testimony. Reported experience might be based on one’s own sensation, the testimony of others, or revelation. Boyle insisted that he used testimony not as authority to support his views but as ‘Certificates to attest Matters of fact’. 
Testimony therefore became an essential component of establishing the facts of nature, and the natural philosophers were well aware that witnesses might vary in number and be more or less credible. The Royal Society emphasized the importance of credibility despite the fact that many reporters, especially those participating in its worldwide fact-gathering endeavors, were not personally known to the membership (Carey, 1977). Some kinds of fact-establishing required little more than honesty, sharp eyes, and an ability to describe or illustrate what had been viewed. In others, skill in anatomy, physics, chemistry or the use of various scientific instruments might be required. Credibility was assessed on the basis of a range of considerations that included social status but also the experience, skill, fidelity, impartiality and number of observers.
Recently Serjeantson has suggested that the Royal Society’s approach to testimony involved a modification of the traditional modes of argument taught by rhetoric and dialectic, with Boyle as the crucial figure in altering the use of testimony in argument from one of ‘authority’ to that of ‘truth’ (Serjeantson, 1999). Boyle may indeed have been a key figure in introducing this more modern concept of testimony into the Royal Society, but he was operating in the context - as he was himself aware - of a legal practice long grounded in precisely this approach to testimony. Sound belief in both law and natural history was couched in the language of moral certainty, or truth beyond a reasonable doubt. Such belief might later be proved erroneous, but it was as firm as was possible when a sufficient number of credible witnesses reported observing something that was not thought to be impossible. 
This shared context of law and natural history suggests a reexamination of the Shapin and Schaffer gentleman thesis noted at the beginning of this study (Shapin & Schaffer, 1985; Shapin, 1991). For, as we have seen, in law gentle status was relevant but far from decisive in assessing credibility, and Bacon, who did so much to advance natural history and experimental philosophy, drew attention to whether witnesses were ‘vain speaking’, ‘light’, ‘sober and severe’, but not to their social status.
If most of the members of the Society were gentlemen,  many
were only in the process of becoming so as they gained preferment in church
and state. Gentlemen and aristocrats
typically treated English clergymen as being of middling status.
It is unclear whether, or when, John
Wilkins, the son of an
Both legal and scientific knowledge were dependent on testimony. In the law courts the word of a gentleman was respected, but the legal threshold of independence and thus credibility was more likely to be drawn at the line of the yeoman rather than that of the gentleman. The Society certainly preferred gentlemen members, but it did not reject the testimony of those of middling status, especially if the testifiers possessed relevant skills and expertise.  The programmatic needs of the Society often required the testimony of those with less than gentlemanly status. The Society’s Georgical Committee, for example, sent out questionnaires ‘to experienced Husbandmen’ who would ‘from their owne and their knowing friends observations and experience’, report on agricultural practices. ‘Those skillful in husbandry’ would ‘impart their knowledge’ of ‘the chief particulars observable’.  The Society’s numerous investigations of trades, crafts and technological processes also relied on knowledgeable but ungentlemanly informants.
The Society often relied on the testimony of individuals whose social status and expertise were not known, particularly when the information came from foreign sources. English diplomats serving abroad were frequently asked to acquire and transmit factual data from knowledgeable local informants. Roughly the same situation prevailed in reports commissioned from travelers and ship captains. Indeed few if any ship captains would have qualified as gentlemen. Although eyewitness testimony was considered best, second-hand testimony was often all that was available. In such situations the social status of the original source on which the second-hand testimony was based might well be unknown or less than elevated.
Certainly the virtuosi were concerned about problems of credibility, but they hardly viewed the issue as one resolvable solely by consulting the social register or the customs of Italian courtiers. On one occasion Boyle noted that the skillful ‘should be Credited in their own Art; Especially when those things whose Nature they so Confidently take upon them to teach others are not only Productions of their own Skill, but such as other Know not else what to make of’.  On another occasion he noted that many pieces of testimony were ‘insufficient for want of moral Qualification in him that gives them’. Even if witnesses were ‘honest and sincere’ they ‘might be insufficient if the matters of fact require Skill’. 
Hooke was sensitive to a variety of problems associated with testimony. Not every observer was ‘fitt to be a collector’. ‘Observations not rightly and accurately made’ were ‘pernicious and Destructive’. The natural historian must be taught ‘what he ought to observe, how to examine it, how to preserve & register it. How to range and order it’. Although this work was best done by ‘a society of men… both of Parts and Fortune, & of such as have a will as well as abilities’, gentlemanly status alone was not sufficient. The naturalist must be ‘knowingly versed in all the various ways of examining & trying of matter of making experiments,… on various kinds of substances’. Under these conditions experiments might ‘stand like so many witnesses to give testimony of the truth or against that error,… a most severe examination of these witnesses must be made before a jury can… give their verdict or a judge pronounce sentence for branding one proposition or hypothesis as erroneous… or for establishing another for truth or axiom’. Hooke referred to his own experiments as ‘History and matter of Fact’ and noted, ‘How neer the nature of Axioms must all those Propositions be which are examin’d before so many Witnesses’. 
The Royal Society members were indeed for the most part gentlemen, but many of them were closer to the middling sort than to the aristocracy. They esteemed gentle testimony, but heard, and indeed sought out, testimony from all those favorably situated by skill or experience credibly to provide it. This was not a world in which the mere word of a gentleman was either necessary or sufficient. 
As we noted in connection with law, ultimately the word of a gentleman certainly cannot be decisive when the words of two gentlemen conflict as they often did in the course of lawsuits. Conflicting testimony was a common problem for all the discourses of fact. If spokesmen and members frequently analogized the Royal Society to a law court, in practice they avoided making collective judgments about matters of fact. Unlike the courts, they could usually avoid deciding between conflicting reports of fact, and they had the further advantage over jurors and jurists that it was possible, in at least some instances, to repeat experiments and observations.
Conflicting testimony, however, could become disruptive, and
did so when respected astronomers Adrien Auzout and Johannes Hevelius reported
conflicting observations of the 1664-5 comet and referred their dispute to the
Society for judgment (Shapin & Schaffer, 1985).
In this instance Royal Society
President Viscount Brouncker and the
Philosophical Transactions used similar legal language, referring to the
‘Controversie being about matter of fact, wherein Authority, Number, and
reputation must cast the Balance’. Although
initially the weight of Society opinion appeared to run against him, Hevelius
characterized the Society as ‘skilled and impartial judges’, and continued to
offer his ‘just proofs and defense’. He
insisted he had reported his ‘observations with great care’, indicating that
his observations had been viewed ‘with other notable persons’.
He asked the Society to ‘rigorously
investigate, examine, compare… [and] then declare their judgement’.
At the opposite end of the spectrum concurring testimony from a number of witnesses was welcomed. Hooke emphasized examination before ‘many witnesses’, and Sprat, with considerable exaggeration, indicated that while two or three witnesses were decisive in law, the Royal Society might have the ‘concurring Testimonies of threescore or a hundred’. 
Though the new natural history was to be based on first-hand observation, Baconian naturalists recognized the weakness of the senses and promoted the use of sense-enhancing instruments such as the microscope, telescope and air pump. Instruments would modify the nature of testimony because previously unobservable phenomena could now be observed or measured. Astronomy was transformed by new and improved telescopes, as were anatomy, embryology and pathology by the microscope. Instruments sometimes resulted in distortion or error but on the whole enhanced the credibility of those employing them. Those with telescopes, for example, were deemed better observers than those without. The social status of some observers might be counterbalanced by skillful use of the microscope or telescope by others. Instruments also added a new dimension to witnessing when naturalists manufactured and manipulated the phenomena they were to observe and witness.
Although deeply indebted to legal and historical notions, scientific witnessing also diverged from them. The former depended on the sensual acuity of ordinary persons observing unique events. The latter increasingly relied on instruments and replicability. The Royal Society began with reports of distant facts and the testimony of multiple witnesses to a particular experiment conducted in the presence of the membership or some of it. Eventually science moved in the direction of replication rather than relying on multiple witnesses of a particular event as the key mode of verification. In time these differences would generate a sense that the testimony of the scientist was more credible than that of ordinary persons.
The scientific norm of the impartial investigator, like that of the credible witness, was indebted to the legal tradition, where the very idea of ‘justice’ implied impartiality and absence of bias. Because it was believed that biased judges, juries and witnesses contaminated legal fact-finding, institutional safeguards, however imperfect, were developed to prevent interested testimony and judgment. The norm of impartiality was also adopted by the Royal Society, and members often proclaimed their lack of interest and bias. Boyle claimed to cultivate ‘chemistry with a disinterested mind’, without caring for ‘personal advantages’. John Wallis stressed the need for ‘disinterested, and unbiased persons to judge’ the dispute between astronomers Hevelius and Auzout.  Statements of the need for impartial observers, impartial witnesses and impartial judgments were commonplace in the publications of the virtuosi and became part of the scientific ideology, an ideology frequently contrasted to that of the disputatious scholastics, who did not rely on empirical data.
No doubt scientific impartiality owes something to gentlemanly norms as well as to the judge, jury and witness. Yet the image of the gentleman, particularly when infused with aristocrat or courtier colors, could also elicit visions of pride, refusal to admit error, disputatiousness, deference to superior rank and overly ornamented language that were far from the norms, practices and language of natural philosophy proposed by the Royal Society.
‘Testimony’, a familiar concept in a number of disciplines, had by the last decades of the seventeenth century achieved a secure position in the making and assessment of empirical knowledge. Its status was made abundantly clear in John Locke’s immensely influential Essay concerning human understanding (1690). A member of the Royal Society and a close friend of Robert Boyle, Locke, like the jurists and the virtuosi, viewed testimony as the basis for ‘matter of fact’. In judging testimony one considered
1. The Number. 2. The Integrity. 3. The Skill of the Witnesses. 4. The Design of the author, where it is a testimony out of a book cited. 5. The Consistency of the Parts and Circumstances of the Relations. 6. Contrary testimonies… [As] the Relaters are more in number, and of more Credit, and have no Interest to speak contrary to the Truth; so that matter of Fact is like to find more or less belief.
Significantly, the social status of those giving testimony is unmentioned by Locke. Knowledge was based on ‘the certainty of Observations,… the frequency and constancy of Experience, and the number and credibility of Testimonies’. Locke adopted the legal rule of rejecting copies of records and copies of copies, finding that rule equally suited to ‘inquiry after material Truths’. ‘The farther off’ testimony was ‘from the original… the less force and proof it has’. Thus ‘A credible Man vouching his Knowledge of it, is a good proof; But if another equally credible do witness it from his Report, the Testimony is weaker; and a third that attests the Hearsay of an Hearsay, is yet less considerable’. Locke’s jury-like fact evaluator then ‘casts up’ or ‘summs up’ ‘upon the whole evidence’. Human testimony and matter of fact, for Locke, covered all observable phenomena. When a particular matter of fact is ‘consonant to the Observation of our selves and others, in the like case, comes attested by the concurrent Reports of all that mention it, we receive it as easily, and build as firmly upon it, as if it were knowledge; and we reason and act thereupon with as little doubt, as if it were perfect demonstration’. 
This paper has suggested that legal methods for establishing
the believability of ‘matters of fact’ had important consequences for
seventeenth-century natural history and thus for seventeenth-century English
science more generally. It argues that
when naturalists turned their attention to the particulars of natural
phenomena and to experiment, they adopted a method of proof based on legal
notions of credible testimony. Legal
practices were familiar to a very large segment of the English population,
many of whom participated directly in the process of legal fact-finding.
To deny the ability of rather ordinary
individuals to report on and evaluate matters of fact would have amounted to a
rejection of the legal system in which the English took such pride.
The conventions associated with legal
witnessing were thus easily incorporated into the scientific enterprise and
required little justification by the naturalists.
Over time, the testimony of the
naturalist gained increased status, in part because of the use of
sense-enhancing instruments, in part because of the possibility of
replication, and in part because of public recognition and admiration for the
scientific accomplishments of such men as Boyle, Hooke and
The legal tradition of establishing facts was well established
and widely known in
1 Recently, however, Serjeantson (1999) has emphasized the role of rhetoric and dialectic in the development of scientific testimony.
2 Although Serjeantson (1999) contrasts the law with rhetoric and dialectic, many concepts and modes of argument in law from the medieval period onward were derived from rhetoric and dialectic. The disciplines therefore should not be sharply contrasted. For discussions of rhetoric and law see Shapiro (2001), Schoeck; Schoeck; Giuliani; Bland; Knafla; Prest and Hohmann. See also Kargon and Shapiro. If methods of treating legal testimony were indebted to Cicero, Quintilian and the Ad Herennium, questions of ‘fact’ in the legal arena were not questions of authority in the sense of citing respected authors. The law deployed criteria of rhetorical origin to believe or disbelieve the testimony of those claiming to be eyewitnesses and thus to establish the truth of ‘matter of fact’. Medieval and early modern lawyers, both English and continental, also deployed the principles of classical rhetoric to develop modes of argumentation and interpretation of legal texts.
3. Circumstantial evidence might also be used. Matters of fact in most situations were proved by testimony of witnesses or by ‘authentic’ documents.
4. For a thorough discussion of the interrelationships between the common law, civil law and equity’s law of proof, see Macnair (1999).
5. Gilbert (1754), pp. 3, 139-140, 144, 148.
6. It was also possible to convict on the basis of circumstantial evidence.
7. Bacon (1857-74), Vol. I, p. 513; More (1987), p. 157; Hale (1736), Vol. I, pp. 277–280, 635; Macnair (1993), p. 143; Hale (1971), pp. 154, 164, 165. See also Coke (1644), p. 163; Hale (1677), p. 128; Thomas Hobbes (1962), p. 157.
8. Hale (1971), p. 164; Hale (1677), p. 128; Thomas (1979), p. 473; Gilbert (1754), pp. 3, 139-140, 144, 148.
9. Macnair (1993), p. 143; Gilbert (1754), p. 147. See also More (1987), Vol. X, p. 157; Hale (1971), p. 164.
10. Gilbert’s Law of evidence, which summarized current views, was written during the reign of Queen Anne and was planned to form a portion of a general work on English law (Macnair, 1999, p. 20). William Nelson’s The law of evidence appeared in 1717. Macnair suggests that there were no legal rules governing the evidence presented to a jury before about 1550, that some Elizabethan and Jacobean judges thought there were rules, and that there was a definite law of evidence by the 1660s (Macnair, 1999, p. 22).
11. See Fitzherbert (1583); Lambarde (1581);
12. The Society prepared a list of queries for travelers.
13. Bacon was familiar with the modified civil law procedure practised in Chancery as well as with the procedure of the common law. For discussions of the relationship between Bacon’s legal thought and his natural philosophy, see Kocher (1957), pp. 3-26; Wheeler; Wheeler; Cardwell; Sargeant; Neustadt and Martin.
14. Bacon (1857-74) Vol. IV, pp. 19, 94-95, 192-193, 292; Bacon (1995), pp. 11, 107-108, 224-227.
15. Bacon (1857-74) Vol. IV, pp. 260-261; Bacon (1995), pp. 209-311.
16. Bacon (1857-74), Vol. IV, pp. 26, 28-29, 30-31, 252, 260-261; Bacon (1995), pp. 24-26, 29, 227, 297, 309-311.
17. Ancient and Renaissance astronomers were also
concerned with accurate observations but did not refer to ‘fact’ and
testimony. The ‘transmutation
histories’ of the alchemists utilized the concept of matter of fact and the
language of witnessing. See
18. For a different view see Sargeant (1989).
19. Sprat (1958), pp. 70, 83, 99, 100.
20. Usage in
21. Glanvill (1671), pp. 143-144; Glanvill (1665), p. 118; Glanvill (1676), Essay II, pp. 46, 49-50.
22. Boyle and Boyle, Preface; Boyle (1690), p. 52; Boyle (1772), Vol. I, p. 34; Vol. II, pp. 741, 742. See also Kargon (1971), pp. 72-81. Boyle recognized the distinction between testimony as authority and testimony as the grounding for belief in ‘matter of fact’. See Serjeantson (1999).
23. An important exception was proof of miracles. These were proved to a moral certainty by multiple concurrent witnesses, although not possible by natural means. See Shapiro (2000), pp. 168-188; Van Leeuwen (1963); Reedy (1985); Burns (1981). Continued belief in the possibility of witchcraft remained a problem since for some witchcraft was possible and for others not. The laws against witchcraft remained in force, but prosecutions and convictions sharply declined (Thomas; Shapiro and Bostridge).
24. Hunter (1981).
25. Frank (1980).
26. For a critique of Shapin’s emphasis on honor, and his inclination to contrast honor with ‘lie’ rather than error, see Lipton (1998), pp. 8-14. For a similar inclination to treat ‘lie’ rather than error as the opposite of truth in the legal context, see Fisher (1997).
27. The ‘Heads of Enquiry’ were printed in the
28. Boyle (1661), p. 167.
29. Westfall (1956), pp. 115, 116.
30. Oldroyd (1987), pp. 146-47, 153, 157; Oldenburg (1965-86), Vol. II, p. 383; Hooke (1665), Preface.
31. For the norms of civility and the status of urban freemen see Barry (2000).
32. Hooke (1665), Preface; Sprat (1958), p. 100.
33. Royal Society, Boyle Letters I, 108r, 131r; Boyle (1772), Vol. II, p. 626; Oldenburg (1965-86), Vol. III, p. 342. See also Vol. IX, p. 658; Vol. II, p. 27; Hooke (1665), Preface.
34. Locke (1975), Bk. IV, Ch. XV, Sect. 2, 5, 6;
Ch. XVI, Sect. 1, 6, 9, 10.
Hypotheses unsupported by matter of fact were unacceptable; those based on
‘sensible experience’ were to be treated as ‘artificial Helps’ but not
‘Philosophical Truth’ (Bk. II,
Ashworth, 1990. W.B. Ashworth , Natural history
and the emblematic world view. In: D. Lindberg and R. Westman, Editors,
Reappraisals of the Scientific
Bacon, 1857-74. Bacon, F. (1857-74). Works, 14
vols. (J. Spedding, R. L., Ellis & D. D. Heath, Eds.).
Bacon, 1995. Bacon, F. (1995).
Novum Organum with other Parts of the
Great Instauration (P. Urbach & J. Gibson, Trans. & Eds.).
Barry, 2000. J. Barry , Civility and civic
culture in early modern
Bland, 1957. D.S. Bland , Rhetoric and the law
student in sixteenth-century
Bohun, 1684. E. Bohun ,
Justice of the peace. (1684)
Bostridge, 1997. I. Bostridge ,
Witchcraft and its transformations c.
1650-1750. , Clarendon Press,
Boyle, 1661. R. Boyle ,
Certain Physiological Essays.
Boyle, 1662. R. Boyle ,
New Experiments Touching the Spring of
Boyle, 1690. R. Boyle ,
The Christian Virtuoso. (1690)
Boyle, 1772. Boyle, R. (1772).
Works, 6 vols. (T. Birch, Ed.).
Burns, 1981. R.M. Burns ,
The great debate on miracles: From
Joseph Glanvill to David Hume., Bucknell University Press,
Cardwell, 1990. K.W. Cardwell , Francis Bacon Inquisitor. In: W.A. Sessions, Editor, Francis Bacon’s legacy of texts, AMS, New York (1990), pp. 260–289.
Carey, 1977. D. Carey , Compiling nature’s history: Travelers and travel narratives in the early Royal Society. Annals of Science 54 (1977), pp. 269–289.
Cockburn and Green, 1988. J.S. Cockburn and T.A.
Green , Twelve good men and true: The
criminal trial jury in
Coke, 1644. E. Coke , Third part of the
Crompton, 1606. R. Crompton , L’office et
auctoritie de justices de peace. (1606)
Daston, 1991. L. Daston , Marvelous facts and
miraculous evidence in early modern
Daston, 1996. L. Daston , Strange facts, plain facts, and the texture of scientific experience in the Enlightenment. In: S. Marchand and L. Lunbeck, Editors, Proof and persuasion: Essays on authority, objectivity and ideology, Turnhout, Brepols (1996), pp. 22-49.
Dear, 1995. P. Dear , Discipline and experience:
The mathematical way in the scientific revolution. ,
Donahue, 1981. C. Donahue , Proof by witnesses in
the church courts of medieval
Drake, 1996. E.T. Drake ,
Restless genius: Robert Hooke and his
earthly thoughts. ,
Espinasse, 1956. ‘Espinasse, M. (1956).
Fisher, 1997. G. Fisher , The juries rise as lie detector. Yale Law Journal 107 (1997), p. 475.
Fitzherbert, 1583. Fitzherbert, A., Sir (1583).
L’office et auchthoritie de Justices de
Fletcher, 1940. J.M.J. Fletcher , Seth Ward,
Frank, 1980. R.G. Frank ,
Gilbert, 1754. Gilbert, G. (1754).
The law of evidence.
Gilissen, 1965. J. Gilissen , La Preuve en Europe du XVIe au debut du XIXe siècle, Rapport de Synthèses. In: La Preuve, Deuxieme Partie: Moyen Age et Temps Moderne, Recueils de la Société Jean Bodin pour l’histoire comparative des institutions, 17, Editions de la libraire encylopedique, Brussels (1965), pp. 755–833.
Giuliani, 1969. A. Giuliani , The influence of rhetoric on the law of evidence and pleading. Juridical Review 62 (1969), pp. 216–251.
Glanvill, 1665. J. Glanvill ,
Scepsis scientifica. (1665)
Glanvill, 1671. J. Glanvill ,
A Praefatory Anser to Mr Henry Stubbe.
Glanvill, 1676. J. Glanvill ,
Essays on Several Important Subjects in
Philosophy and Religion. (1676)
Hale, 1677. M. Hale ,
The primitive origination of mankind.
Hale, 1736. M. Hale ,
History of the pleas of the Crown.
Hale, 1971. Hale, M. (1971).
History of the Common Law of
Hill, 1965. C. Hill ,
Intellectual origins of the English
Revolution. , Clarendon Press,
Hobbes, 1962. T. Hobbes , Leviathan. , Simon & Schuster (1962).
Hohmann, 1989. H. Hohmann , The dynamics of stasis: Classical rhetorical theory and modern legal argumentation. American Journal of Jurisprudence 34 (1989), pp. 171-197.
Hooke, 1665. R. Hooke ,
Hunter, 1981. M. Hunter , The Royal Society and its Fellows, 1660-1770. , British Society of the History of Science, Chalfont St. Giles, Buckinghamshire (1981).
Hunter and Schaffer, 1989. M. Hunter and S.
Schaffer , Robert Hooke: New studies.
, Boydell Press,
Jacob, 1980. J.R. Jacob , Restoration ideologies and the Royal Society. History of Science 18 (1980), pp. 25-38.
Jacob, 1976. M.C. Jacob ,
The Newtonians and the English
Kargon, 1971. R. Kargon , The testimony of
nature: Boyle, Hooke and experimental philosophy.
Kocher, 1957. P. Kocher , Francis Bacon on the science of jurisprudence. Journal of the History of Ideas 18 (1957), pp. 3-26.
Lambarde, 1581. W. Lambarde ,
Eirenarcha, or Of the Office of the
Justice of the Peace. (1581)
Lennard, 1932–34. R. Lennard , English agriculture under Charles II: The evidence of the Royal Society ‘enquiries’. Economic History Review 4 (1932–34), pp. 23-45.
Lipton, 1998. P. Lipton , The epistemology of testimony. Studies in History and Philosophy of Science 29 (1998), pp. 1-31.
Locke, 1975. Locke, J. (1975).
Essay concerning human understanding
(P. H. Nidditch, Ed.).
Macnair, 1993. M. Macnair , A Fragment on Proof by Francis North, Lord Guildford. Seventeenth Century 8 (1993), pp. 143–148.
Macnair, 1999. M. Macnair ,
The law of proof in early modern equity.,
Martin, 1992. J. Martin ,
Francis Bacon, the state and the reform
of natural philosophy., Cambridge University Press,
Merton, 1970. R.K. Merton
, Science, technology and society in
seventeenth-century England., Harper
More, 1987. T. More , The Debellacyon of
Neustadt, 1987. Neustadt, M. (1987).
Making of the Instauration: Science,
politics, and law in the career of Francis Bacon. PhD thesis,
Oldroyd, 1987. D.R. Oldroyd , Some writings of Robert Hooke on procedures for the prosecution of scientific inquiry, including his ‘Lectures of Things Requisite to a Natural History’. Notes and Records of the Royal Society 41 (1987), pp. 145–167.
Philosophical Transactions of the Royal Society,
1665. Philosophical Transactions of the Royal Society (1665-6).
Prest, 1977. W. Prest , The dialectical origins
of Finch’s Law.
Raven, 1959. C.E. Raven ,
John Ray, naturalist. (2nd ed.),,
Reedy, 1985. G. Reedy ,
The Bible and reason: Anglicans and
Scripture in late seventeenth-century England.,
Schoeck, 1953. R. Schoeck , Rhetoric and law in
Sargeant, 1989. R.M. Sargeant , Scientific
experiment and legal expertise: The way of experience in seventeenth-century
Schoeck, 1983. R. Schoeck , Lawyers and rhetoric
Scriba, 1972. C.J. Scriba , The autobiography of John Wallis, FRS. Notes and Records of the Royal Society 25 (1972), pp. 17-47.
Serjeantson, 1999. R.W. Serjeantson , Testimony
and proof in early-modern
Shapin, 1994. S. Shapin ,
A social history of truth: Civility and
science in seventeenth-century England.,
Shapin, 1991. S. Shapin , A scholar and a
gentleman: The problematic identity of the scientific practitioner in
Shapin and Schaffer, 1985. S. Shapin and
Shapiro, 1968. B.J. Shapiro ,
John Wilkins 1614-1672: An intellectual
Shapiro, 1983. B.J. Shapiro ,
Probability and certainty in
seventeenth-century England., Princeton University Press,
Shapiro, 1986. B.J. Shapiro , ‘To a moral
certainty’: Theories of knowledge and Anglo-American juries, 1600–1850.
Shapiro, 1991. B.J. Shapiro ,
‘Beyond reasonable doubt’ and ‘probable
cause’: Historical perspectives on the Anglo-American law of evidence.,
Shapiro, 2000. B.J. Shapiro ,
A culture of fact:
Shapiro, 2001. B.J. Shapiro , Classical rhetoric and the English law of evidence. In: V. Kahn and L. Hutson, Editors, Rhetoric and law in early modern Europe, Yale University Press, New Haven (2001), pp. 54-72.
Sprat, 1958. Sprat, T. (1958).
The history of the Royal Society of
Thomas, 1979. K. Thomas ,
Religion and the decline of magic.,
Van Leeuwen, 1963. H.G. Van Leeuwen,
The problem of certainty in English
thought 1630-90., Nijhoff,
Westfall, 1956. R.S. Westfall , Unpublished Boyle papers relating to scientific method. Annals of Science 12 63-73 (1956), pp. 103-117.
Wheeler, 1983a. H. Wheeler , The invention of modern empiricism: Juridical foundations of Francis Bacon’s philosophy of science. Law Library Journal 76 (1983), pp. 78-120.
Wheeler, 1983b. H. Wheeler , Science out of law: Francis Bacon’s invention of scientific empiricism. In: D.H. Nelson and R.I. Sklar, Editors, Towards a humanistic science of politics, University Press of America, Lanham (1983).
Zilsel, 1945. E. Zilsel , The genesis of the concept of scientific progress. Journal of the History of Ideas 6 (1945), pp. 325-349.