SOURCE: “John Henry Newman and the Grammar of Assent,” in Reason and the Heart: A Prolegomenon to a Critique of Passional Reason, Cornell University Press, 1995, pp. 55-83.

[In the following essay, Wainwright observes Newman's process of informal reasoning—his “illative sense”—as it is demonstrated in An Essay in Aid of a Grammar of Assent.]

Consider these chains of reasoning. (1) Our conviction that Great Britain is an island is well-founded. We have no doubt that it is true. But if asked to give our evidence for it, we can only respond that “first, we have been so taught in our childhood, and it is so on all the maps; next, we have never heard it contradicted or questioned; on the contrary; everyone whom we have heard speak on the subject of Great Britain, every book we have read, invariably took it for granted; our whole national history, the routine transactions and current events of the country, our social and commercial system, our political relations with foreigners, imply it in one way or another. Numberless facts, or what we consider facts, rest on the truth of it; no received fact rests on its being otherwise” (GA [An Essay in Aid of a Grammar of Assent] 234-35).¹ Our belief that Britain is an island is not based on a rigorous deductive or inductive argument. But it is reasonable; a variety of independent considerations support it and nothing supports its denial.

(2) A historian of the Middle Ages asserts that the Aeneid could not be a thirteenth-century forgery. Her conviction is partly based on her knowledge of the capacities of the medieval mind. This knowledge depends on a lifetime of reading and study. Although many of the considerations that contributed to it have been forgotten and others have merged into a general impression of what the medieval mind could and could not do, her knowledge is real. “We do not pretend to be able to draw the line between what the medieval intellect could and could not do; but we feel sure that at least it could not write the classics. An instinctive sense of this [as well as a faith in testimony] are the sufficient, but the undeveloped argument on which to ground our certitude” (GA 237).

(3) Just as we “instinctively” infer “the fact of a multiform and vast world, material and mental” from the “phenomena of sense,” and just as a child instinctively recognizes in the “smiles or the frowns” of a face “not only a being external to himself but one whose looks elicit in him confidence or fear,” so we spontaneously infer God's existence from “particular acts of conscience” (GA 67-68; see also 97, 102). Our experience of guilt and moral inadequacy “instinctively” suggests the presence of a “moral governor” and “judge.”

(4) Someone argues: “The Catholic religion is true, because its objects, as present to my mind, control and influence my conduct as nothing else does.” Or, it is true “because it has about it an odour of truth and sanctity … as perceptible to my moral nature as flowers to my sense, such as can only come from heaven.” Or it is true “because it has never been to me anything but peace, joy, consolation and strength, all through my troubled life” (GA 174).

(5) “‘I think,’ says the poor dying factory girl in the tale [Elizabeth Gaskell's North and South], ‘if this should be the end of all, and if all I have been born for is just to work my heart and life away, and to sicken in this dree place, with those millstones in my ears forever, until I could scream out for them to stop and let me have a little piece of quiet, and with the fluff filling my lungs, until I thirst to death for one long deep breath of the clear air, and my mother gone, and I never able to tell her again how I loved her, and of all my troubles—I think, if this life is the end, and that there is no God to wipe away all tears from all eyes, I could go mad!’ Here,” says Newman, “is an argument for the immortality of the soul” (GA 247).

None of these arguments meet rigorous deductive or inductive standards. Newman nonetheless believes that each is a good argument and that its conclusion is reasonable. This assessment rests on three convictions. (1) Many good arguments are neither deductively valid nor inductively sound. Their conclusions are not entailed by their premises. Nor can they be derived from them by inductive extrapolation (by generalizing from the character of a fair sample, for example, or by inferring that an event will occur because similar events have occurred under similar conditions in the past). They are, instead, inferences to the best explanation. A hypothesis is adopted because it provides a more plausible explanation of a range of facts than its competitors. The most plausible explanation of the facts cited in the first argument, for instance, is that Great Britain is an island.² (2) There is no “common measure between mind and mind” that can be used to conclusively settle disagreements (GA 82); on the contrary, reason is “personal” and reflects the experience and cast of mind of the person who employs it. The evidence is often too complicated and “delicate” to be fully articulated. (The second argument is a case in point.) Some evidence is only accessible to men and women who have immersed themselves in the subject matter or are extensively acquainted with the phenomenon under investigation. (A person whose conscience is deadened will not appreciate the third argument.) The evidence on which we rightly rely, then, is not always fully stateable or universally accessible and, so, cannot be used to secure general agreement. Even when the evidence is stateable and accessible, differences may remain. Although there are rules and guidelines for assessing informal reasoning, their application requires judgment. But judgments reflect the characters and biographies of the people who make them, and these vary from one person to another. (3) Our hopes and fears, needs and desires, longings, “instincts,” and “divinations” sometimes rightly affect our assessment of a body of evidence. The fourth and fifth arguments are examples.

Newman's pioneering recognition of the importance and ubiquity of inferences to the best explanation is no longer controversial. His other two contentions are. We will examine them in the following sections.

THE ILLATIVE SENSE

Newman calls the faculty of informal reasoning “the illative sense.” “Illative” was “defined in contemporary dictionaries [from 1864 and 1901] in terms of ‘inference; deduction; conclusion.’ Newman may have noticed the word in Locke who writes of ‘illation’ as the intellectual faculty which ‘consists in nothing but the perception of the connexion there is between the ideas, in each step of the deduction.’ … The phrase ‘illative conjuctions’ is found in Whately's Logic in a passage which Newman helped to compose.”³ Newman's own usage is ambiguous. M. Jamie Ferreira points out that “the illative sense” sometimes refers to the power of “judging and concluding” in all its forms including those which are “biased and degraded” and that it sometimes refers to the power “in its perfection.”⁴

The faculty is principally employed in three ways: (1) in conducting an argument, (2) in assessing prior probabilities, and (3) in evaluating an argument's overall force.

In conducting an argument, the illative sense is used to “scrutinize, sort, and combine” premises and to “correctly [employ] principles of whatever kind, facts or doctrines, experiences or testimonies, true or probable” (GA 282). Consider a dispute between two historians. Each has first to decide which “opinions … to put aside as nugatory,” what evidence is relevant, what has “prima facie authority,” and the relative weights to be placed on various kinds of evidence” (“tradition, analogy [with familiar historical phenomena?], isolated monuments and records, ruins, vague reports, legends … language,” etc.). “Then arguments have to be balanced against each other.” Finally, each must decide whether a conclusion can be drawn and how certain it is. “It is plain how incessant will be the call … for the exercise of judgment” (GA 284).

The illative sense, then, decides which considerations are relevant, assigns weights to different kinds of considerations, marshals the evidence in some sort of order, applies appropriate principles (those used in assessing testimony, for example), and balances the positive and negative considerations against each other. Although these things can be done well or badly, “it is plain … how little that judgment will be helped on by logic and how intimately it will be dependent upon the intellectual complexion of the” reasoner (GA 284). In causal inquiries, for example, judgments of relevance will be affected by whether “we view a subject” as a system of efficient causes or as a system of final causes (GA 290). Again, Gibbon's account of “the rise of Christianity is the mere subjective view of one who could not enter into its depth and power” (GA 291); by restricting his attention to surface phenomena, Gibbon failed to consider relevant facts and experiences.

Our illative sense is also responsible for judgments of antecedent probability. We legitimately dismiss some hypotheses and opinions without argument.⁵ Those we cannot dismiss as irrelevant or absurd are assigned a certain probability. But these assignments “will vary … according to the particular intellect” that makes the assessments (GA 233). And we not only assign conflicting probabilities to the premises. Each of us also has “his own view concerning” the likelihood of the conclusion “prior to the evidence; this view will result from the character of his mind. … If he is indisposed to believe he will explain away very strong evidence; if he is disposed, he will accept very weak evidence” (“LSF” [“Love the Safeguard of Faith against Superstition,” in Fifteen Sermons Preached before the University of Oxford] 226).

For example, one historian finds the proposition “No testimony should be received, except such as comes from competent witnesses” more plausible than “Tradition, though unauthenticated, being … in possession, has a prescription in its favor.” Another does not. As a result, the historians assign different initial probabilities to the same events (GA 294). Again, Pascal's cumulative case argument for Christianity depends “upon the assumption that the facts of Christianity are beyond human nature” and that it is thus antecedently unlikely that human beings would have invented them. Hence, “as the powers of nature are placed at a high or low standard,” the force of his argument “will be greater or less; and that standard will vary according to the respective dispositions, opinions, and experiences, of those to whom the argument is addressed” (GA 245). In short, before drawing a conclusion we assign degrees of probability (plausibility) to the argument's “antecedents” (principles, facts, testimony, and so on).⁶ We also approach the argument with views of the conclusion's antecedent probability. These judgments reflect our experience, knowledge, and temperament and affect our evaluation of the argument's strength.

One of the illative sense's most important functions is to assess an argument's overall force. “The mind itself is more versatile and vigorous than any of its works, of which language [i.e., logic] is one, and it is only under its penetrating and subtle action that the margin disappears … between verbal argumentation and conclusions in the concrete. It determines what science cannot determine, the limit of converging probabilities and the reasons sufficient for a proof” (GA 281-82). It determines, that is, how strongly the argument's “antecedents” support its conclusion.⁷ Assessments of an argument's force, too, reflect the histories and tempers of the persons who make them. For when it comes to “the question, what is to become of the evidence, being what it is,” each must decide “according to (what is called) the state of his heart” (“LSF” 227).

Just as there are no formal rules for producing or recognizing good poetry, so there are no formal rules for determining the truth in concrete matters. In “concrete reasonings,” the “ultimate test of truth or error in our inferences” is “the trustworthiness of the Illative Sense that gives them its sanction” (GA 281). The ultimate test, in other words, is our own best judgment. My judgment, however, is irredeemably personal, for I can only view the various pieces of evidence “in the medium of my primary mental experiences, under the aspects which they spontaneously present to me, and with the aid of my best illative sense” (GA 318, my emphasis).

Formal reasoning (whether deductive or inductive) is not a real alternative to illative inference. The mind's acts of concrete reasoning are too subtle, varied, and intricate to be fully verbalized. Try, for example, to articulate the mental acts involved in arriving at the conclusion that the emendations of an early commentator on Shakespeare's Henry V are reliable (GA 217-22), or to adequately verbalize the grounds of one's impression that whatever the medieval mind could do, it could not produce the Aeneid (GA 235-37), or the “delicate and at first invisible touches” in an anonymous publication which point toward a certain author (GA 259). Because our reasoning cannot be fully symbolized, formal principles cannot get a firm grip on it.⁸

But (and this is the more important point) even if our concrete reasonings could be adequately symbolized, our illative sense would remain indispensable. Formal inference and concrete reasoning are not real alternatives since the former is an abstraction from the latter. Concrete reasoning is formal reasoning “carried out into the realities of life, its premises⁹ being instinct with the substance and the momentum of the mass of probabilities, which acting upon each other in correction and confirmation, carry it home to the individual case” (GA 232-33). One employs illation in deploying formal arguments, for one's assessment of the premises' plausibility and of the relevance of the argument's conclusion to the case at hand rest on assumptions and tacit understandings that cannot be reduced to formal principles.

“NO COMMON MEASURE”

Their illative senses often lead people to opposed conclusions. Why is this the case? Partly because the “first principles … with which we start in reasoning on any given subject matter” are “very numerous and vary … with the persons who reason … only a few of them [being] received universally” (GA 66). Some historians, as we have seen, assume that “no testimony should be received, except such as comes from competent witnesses.” Others assume that “tradition, though unauthenticated, being … in possession, has a prescription in its favour” (GA 294). Their assessments of evidence will correspondingly differ. Bishop Butler denied that “a revelation, which is to be received as true, ought to be written on the sun.” Newman, on the other hand, thinks that “something may be said in [the claim's] favour” (GA 295). One's attitude toward it will obviously affect one's assessment of the evidence for an alleged revelation. Again, judgments of relevance are affected by the aspects under which a person views a subject—whether or not she looks on nature as a system of efficient rather than final causes, for example, or views thought as a language, or deliberation as a system of weights and balances. “‘Tacit understandings’” and “vague and impalpable notions of ‘reasonableness’ … make conclusions possible.” But they are also “the pledge of their being contradictory. The conclusions vary with the particular writer, for each writes from his own point of view and with his own principles, and these admit of no common measure” (GA 287).¹⁰

Why do our first principles and points of view differ? The “intuitions, first principles, axioms, dictates of common sense, presumptions, presentiments, prepossessions, or prejudices” with which we approach a body of evidence are reflections of our experiences (TP [The Theological Papers of John Henry Newman on Faith and Certainty] 108). Because the latter vary, so will the former.

Our impression of an argument's overall force is also affected by “personal” factors. People sometimes withhold assent from an argument through “a vague feeling that a fault lay at” its “ultimate basis” or because of “some misgiving that the subject matter … was beyond the reach of the human mind.” Again “moral causes, arising out of our condition, age, fortune,” and so on, “prejudice,” or lack of sustained attention can cause us to withhold assent. Or we may remain unpersuaded because “we throw the full onus probandi on the side of the conclusion,” refusing to assent until the arguments are not merely good but conclusive (GA 142-44). These attitudes and vague convictions are rooted in our characters and personal histories. Because these differ, so too will our estimates of an argument's force.¹¹

Assessments of an argument's force are also a function of our real or notional assents.¹² Newman's discussion of real assent is somewhat confusing. Real assents are assents to propositions that are really apprehended. Newman's distinction between notional and real apprehension, however, conflates the distinction between the apprehension of general and of singular propositions, and the distinction between “two modes of apprehending propositions,” “as counters to be combined and calculated inter se” and “as signs to be ‘cashed’ in terms of their appropriate mental images” and associations.¹³ The latter, though, is most important. Real apprehensions are “thingish and imaginative.”¹⁴

One's assent to the force of an argument may depend on whether one's assent to its premises is real or not. For example, the argument from evil will not unduly trouble those whose assent to “Evil exists” is largely notional for they don't appreciate evil's horror. Cosmological arguments are not likely to impress someone who is not struck by the fact that things exist when nothing might have or lacks a vivid sense of how odd it would be to discover a contingent being that had no causes whatever, and whose apprehension of “Something exists” or “Every event has a cause” is therefore notional.

But real assents depend on images (or acts of imagination), and these depend on personal experience. Because “the experience of one man is not the experience of another,” real apprehension and assent “is proper to the individual and, as such, thwarts, rather than promotes, the intercourse of man with man” (GA 82-83). If assessment of an argument's force partly depends on real assent to its premises, consensus will be difficult to achieve.

The illative sense does not, then, provide “a common measure between mind and mind” (GA 82). “Explicit argumentation” concerning the merits of competing principles, assumptions, points of view, and other “starting points” and “collateral aids” to argument, is “sometimes possible to a certain extent.” But “it is too unwieldy an expedient for a constantly recurring need” (GA 290). Even when we do resort to it, agreement is unlikely. For our supporting arguments, too, must be assessed by the illative sense, and the latter reflects “personal characteristics, in which men are in fact in essential and irremediable variance with one another.” The most we can do is “point out where the difference … lies, how far it is immaterial, when it is worthwhile continuing an argument …, and when not” (GA 283).

Our disagreements are so intractable in fact that we can only conclude “that there is something deeper in our differences than the accident of external circumstances [experience, training, and the like]; and that we need the interposition of a Power, greater than human teaching and human argument, to make our beliefs true and our minds one” (GA 293).

Some disagreements, of course, are less intractable than others. The first two of our five arguments involve ordinary cumulative case reasoning. No sensible person would reject the first and no reasonable historian would cavil at the second. Our spontaneous inferences from sensory phenomena to a world of independent physical and mental objects or (according to Newman) from the phenomena of conscience to a moral governor involve natural dispositions to interpret the data in a certain way. But the arguments to the truth of the Catholic religion and the factory girl's argument involve what William James calls “divinations” and “instincts,” and these vary from mind to mind. In cases such as these, it seems doubtful whether anything less than divine intervention can secure agreement.

Yet does not this give the game away by revealing the illative sense's subjectivity? Truth may not be relative (Newman clearly thinks it is not),¹⁵ but illative reasoning surely is. And because all reasoning involves illation, even the most rigorous thinking appears tainted by subjectivity.

This clearly is not Newman's intention. Newman believes, for example, that good cumulative case arguments for the faith are “valid proofs.” Although they cannot be “forced on the mind[s] of anyone whatever,” they are capable of convincing anyone who “fairly studies” their premises (TP 27, my emphasis). And, in general, if one's argument is good, one will find “that, allowing for the difference of minds and of modes of speech, what convinces him, does convince others also. … There will be very many exceptions but those will admit of explanation” (GA 300, my emphasis). Some opposed “intuitions,” for example, can be discounted because they have been created by “artificial and corrupt” social codes and practices. Others can be dismissed as expressions of raw and uncultivated human nature (TP 70-79). The important point is that our illative sense can be well or badly employed. If everyone were to use it rightly, most major disagreements would disappear.

THE ILLATIVE SENSE AND PROPER FUNCTIONING

The test of an argument's validity is “the judgment of those who have a right to judge” (GA 248, my emphasis). “Other beings are complete from their first existence, in that line of excellence which is allotted to them.” Human beings, on the other hand, must acquire the excellence proper to them “by the exercise of those faculties which are” their “natural inheritance.” Each person must complete “his inchoate and rudimental nature … out of the living elements with which his mind began to be” (GA 274). The standard of proper functioning is not furnished by how most people use their ratiocinative powers. It is determined by how those who have perfected them do so. What does proper functioning involve, then, and what are the tests of good illative reasoning?

Experience and practice are necessary. Those are qualified to judge “who by long acquaintance with their subject have a right to judge. And if we wish ourselves to share in their convictions and the grounds of them, we must follow their history, and learn as they have learned. We must take up their particular subject … give ourselves to it, depend on practice and experience more than on [formal or explicit] reasoning” (GA 269). For example, the prerequisite for assessing “We shall have a European War, for Greece is audaciously defying Turkey” is the “experience [“of diplomatists, statesmen, capitalists, and the like”] strengthened by practical and historical knowledge” (GA 241).

Moral qualifications may also be necessary. Indeed, the only “department of inquiry” that does not require “a special preparation of mind” is “abstract science” (GA 321). The reasoner's “moral state” is not important “in a subject-matter so [comparatively] clear and simple as astronomical science,” for instance (although the type of thinking is the same as that employed in “soft” disciplines and in ordinary life, viz., cumulative case reasoning) (GA 253). In other areas, though, the moral state of the inquirer is crucial. For example, in assessing the prospects of war in Greece, experience and knowledge should be “controlled by self-interest” (GA 241).¹⁶

Moral qualifications in the narrower sense are sometimes required. The rays of truth “stream in upon us through the medium of our moral as well as our intellectual being.” The “perception of its first principles which is natural to us is enfeebled, obstructed, perverted, by allurements of sense and the supremacy of self, and, on the other hand, quickened by aspirations after the supernatural” (GA 247). The moral and spiritual qualities needed for successful religious inquiry, for example, are those needed for admission into God's company;¹⁷ the ratiocinative process by which we come to know God is thus a “discipline inflicted on our minds” that molds “them into due devotion to Him when He is found” (GA 276). Again, the mind's “safeguard [against error and superstition] … is a right state of heart … holiness, or dutifulness, or the new creation, or the spiritual mind, however we word it. … It is love” (“LSF” 234).¹⁸ And although a sound and sensitive conscience is needed to properly assess the evidence for natural religion, an investigation of the claims of revelation requires more. One must not only be imbued with the “opinions and sentiments” of natural religion (GA 323); one also must long for revelation and be firmly convinced of God's goodness. We should approach the evidence for revelation as “suppliants,” not “judges.” Those who “resolve to treat the Almighty” with “lawyerlike qualifications”¹⁹ will not discern the evidence's force (GA 330-31).²⁰

Robert Holyer has called our attention to the fact that Newman's views on the imagination's epistemic effects are ambivalent.²¹ He asserts, for example, that the “brilliancy of the image” is not a sufficient reason for assent. “The natural and rightful effect of acts of imagination upon us … is not to create assent, but to intensify it,” bringing propositions home to us so that they stir our emotions and activate our wills (GA 81). But he also suggests that imagination has a proper role to play in eliciting assent. “Real ratiocination and present imagination” should infuse our inferential processes “in order to their due exercise” (GA 250). The factory girl argument is a case in point. The “image” in the argument is the narrative that includes it. The function of the image is “to show very clearly the relation of the emotions experienced by the girl, the beliefs implicit in them, and the conclusion about God.” Most of us have experienced these emotions to some degree but without seeing their relation to belief in God. “The narrative makes this relationship clear by giving us a very obvious example of it.”²² I have argued that real assents to an argument's premises are sometimes needed to appreciate its force. But real assents depend on imagination, and the latter can be lively or weak, cultivated or undisciplined. Newman believed that “one important effect of living a religious life is that it schools the imagination” and emotions.²³ By doing so, it enhances our capacity to appreciate the force of religious arguments.

How does one know when one is using one's illative powers properly? There are (at least) three indications that one is doing so. The first is “the agreement of many private judgments in one and the same view” (GA 248). Newman says, for example, that his argument from conscience would not be “worthwhile my offering it unless what I felt myself agreed with what is felt by hundreds and thousands besides me” (GA 318). In matters of religion, ethics, metaphysics, and the like, “each of us can [ultimately] speak only for himself. … He brings together his reasons and relies on them, because they are his own, and this is his primary” and indeed his “best evidence.” Nevertheless, if it “satisfies him, it is likely to satisfy others” provided that his reasoning is sound and his conclusion true. “And doubtless he does find … that allowing for the difference of minds and modes of speech, what convinces him, does convince others also.” Their agreement is “a second ground of evidence” (GA 300-31). (The first is one's reasons.) Universal agreement should not be expected because people's illative senses are often undeveloped or misemployed. A failure to secure substantial agreement, on the other hand, indicates that one's illative powers are being used idiosyncratically.

Other signs that one has drawn the right conclusion are “objections overcome, … adverse theories neutralized, … difficulties gradually clearing up,” consistency with other things known or believed, and the fact that “when the conclusion is assumed as an hypothesis, it throws light upon a multitude of collateral facts, accounting for them, and uniting them together in one whole” (GA 254, 255-56). A sign that one has reasoned rightly, in other words, is that one's argument satisfies the criteria for inferences to the best explanation.

Successful practice is a further indication that our reasoning is sound. In the absence of other indicators sufficient to warrant certainty, “our only test is the event or experience. Hence the proverbs ‘The proof of the pudding’ etc., etc.” (TP 92).²⁴

THREE OBJECTIONS

This section discusses three difficulties. The first two are not serious. The third is.

1. Newman believes that illative reasoning cannot be formally represented. Jay Newman claims that this is obfuscating. The factory girl argument, for example, can be represented as follows:

Either my life is meaningless, or there is a God to wipe
away all tears from all eyes.
Life is not meaningless.
Ergo.(25)

Again, “in arriving at or confirming the conclusion that Britain is an island, we may make complex deductions as well as complex inductions. … We can produce dozens of [formal] arguments” for the conclusion. For example, “‘I have examined over a thousand’ maps and they have ‘misled me on less than 1 per cent of the occasions that I have relied on them.’ This map of the world indicates that Britain is an island. Ergo.” Or, “Almost everything Uncle Paul has told me is true. Uncle Paul told me that Britain is an island. Therefore, etc.”²⁶ Or again, in commenting on the example of the person who ascertains the authorship of an anonymous publication on the basis of its style, manner, and so on, Jay Newman claims that “after some introspection and reflection,” he could “start to enumerate the main considerations that led him to his judgment.”²⁷

Several comments are in order. First, that our reasoning can be cast in the form of an inductive or deductive argument (or a complex set of them),²⁸ does not imply that our reasoning had that form. (Just as the fact that I can provide reasons for my believing Mary does not imply that my justified belief in what Mary told me was based on those reasons.) Second, the three arguments we are now discussing (or at least the last two of them) are best represented as inferences to the best explanation and not as deductive or inductive arguments.²⁹ Third, the validity of these arguments, and of other inferences to the best explanation, does not depend on our ability to recast them in deductive or inductive form.

In any case, Jay Newman's objection misses the point. Even if all the considerations in cases like the third can be recovered and represented (which I doubt), and even if all three arguments can be recast in a deductive or inductive mode, the effective deployment of the formal representations depends on illation. As Newman argues, an exercise of the illative sense is needed to grasp a formal argument's force. (A word of caution by William G. Ward is also in order. Replacing informal arguments with formal equivalents can sometimes be dangerous. For a person may “calamitously misapprehend the balance of the reasons pro and con from the very probable circumstance, that those facts which tell on one side may be far more easily put into shape, or are far more precisely located in his memory, than those (legitimately preponderating) facts which tell on the other.”)³⁰

2. In an otherwise sympathetic article, Robert Holyer suggests that an argument's imaginative force can be easily confused with its logical force. When a proof appeals to the imagination, “the degree of conviction” it “inspire[s] is in excess of the logical force of the argument.”³¹ All that the factory girl argument strictly establishes, for example, is a need to believe in God.³²

This suggestion, while plausible, fails to do justice to one of Newman's most important contentions—that our imagination and needs can be epistemically relevant.

Is there a gap between (e.g.) the factory girl argument's imaginative and logical force? There is if we identify the latter with the average force the argument has for sane and informed reasoners, or with the force it has for those who refuse to let themselves be influenced by the needs and emotions that the argument expresses. And it is true that imagination and emotion can mislead us. People wrongly “believe what they wish to be true … readily believe reports unfavorable to persons they dislike, or confirmations of theories of their own.” These “inducements to belief … prevail with all of us.” Even “faith degenerates into weakness, extravagance … [and] prejudice” when “our wishes are inordinate, or our opinions are wrong” (“FRC” [“Faith and Reason, Contrasted as Habits of Mind,” in Fifteen Sermons Preached before the University of Oxford] 189-90). It is also true that imagination and emotion can add force to arguments on both sides of an issue—atheological arguments, for instance, as well as theological ones. Their effects are not invariably benign.

Imagination and emotion, then, seem to be one thing, reason and intuition (the grasp of first principles), another. Newman himself says as much. “The mind without any doubt is made for truth. Still, it does not therefore follow that truth is its object in all its powers. The imagination is a wonderful faculty in the cause of truth, but it often subserves the purposes of error—so do our most innocent affections” (TP 152).

The Grammar, on the other hand, contends that the illative reasoning that is utilized in the deployment and appraisal of all arguments, even formal ones, is unavoidably suffused with imagination, emotion, and need. Newman's final opinion appears to be that this is as it should be. “Real ratiocination and present imagination” should infuse the processes of (informal) inference “in order to their due exercise” (GA 250).

Note that we often use epistemic terms to appraise acts of imagination. They can be profound, insightful, or perceptive. Their products can have the ring of truth, place reality in a new and revealing light, and so on, or, on the contrary, be shallow, ring false, or distort reality. That the imagination and emotions can be misused does not imply that, when properly used, they cannot have positive epistemic value. What is needed is a critique of the imagination and emotions—an account of when they are epistemically beneficial and when harmful. We are not totally in the dark. Appreciation of evil's horror, for instance, is needed to grasp the force of the argument from evil. If Christian theism is correct, however, it can be excessive or (perhaps more accurately) can unduly occupy the imagination to the exclusion of other appropriate emotions and images—trust, for example, or Christ as the image of God's love. In other cases, images and emotions that make atheological arguments persuasive to some may be inappropriate. Possible examples are narratives like Butler's The Way of All Flesh or the pictures of Christianity painted by Thomas Paine and other deists.³³

Is there a gap between an argument's imaginative and its logical force? There need not be if (1) an argument's logical force is the force it should have for sane and informed reasoners, and (2) a properly disciplined imagination and emotion should infuse the processes of inference “in order to their due exercise.” To simply assume that (2) is false begs the question.

Nor can one just assume that needs are epistemically irrelevant. Holyer argues that what the factory girl argument strictly proves is not the conclusion but that we need to believe it. The alleged gap would be illusory, however, if this need were epistemically relevant. It is reasonably clear that Newman thought it was. In a letter of 1884 to Wilfred Ward, he wrote, “it seems to me you mean to say that the same considerations which make you wish to believe are among the reasons which, when you actually do inquire, lead you prudently to believe, thus serving a double purpose. Do you bring this out anywhere? On the contrary, are you not shy of calling these considerations reasons? Why?”³⁴ The implication is that Newman thought they were. To simply assume that our wishes and needs are not “reasons” begs the question at issue, “Do wishes, needs, emotions, acts of imagination, and the like sometimes have positive epistemic value?”

3. The most serious charge is irrationalism. Jay Newman believes that Newman has “made the reasonableness of religious belief [though not its truth] a subjective matter.” People's illative senses lead them to opposed conclusions, and there is not any principled way of resolving their differences. He concedes that the problem could be defused by showing that the illative senses of (e.g.) Montaigne and others who would be unimpressed by the factory girl's argument “were somehow defective.” For the judgments that count are those of people “who have a right to judge” (GA 248). He doubts, however, that Montaigne is less qualified to judge her argument than the uneducated factory girl.³⁵ Newman of course thought that “holiness or dutifulness, or the new creation, or the spiritual mind … [or] love” is the qualification needed (“LSF” 234). But Muslims, Buddhists, and others make similar claims. They too insist that they possess a proper moral disposition, “a serious, sober, thoughtful, pure, affectionate, and devout mind” (“LSF” 250).³⁶

The trouble with Newman's thesis is that illative reasoners cannot show that those who disagree with them are in an epistemically inferior position because they do not have any common standards to appeal to. The consequences for his claim are disastrous. If the factory girl's inference “is so personal that it has nothing to do with any common measure, standard, or rules,” it can be neither valid nor invalid.³⁷ Nor is Newman entitled to “assume that his insight is ‘deeper’ and ‘more powerful’ than [say] Gibbon's” if he is not “prepared to produce the explicit, verbal, publicly assessable arguments that will resolve the dispute.”³⁸

While a full reply to this objection must await Chapter 4, three observations are in order.

(1) Whether all inference is rule governed or not depends on how “rules” is understood. It is not, if by “rules” we mean “algorithms.” There are no algorithms for testing inferences to the best explanation. And though the selection of scientific paradigms is governed by what Thomas Kuhn calls “scientific values” (simplicity, comprehensiveness, precision, etc.), there are no mechanical decision procedures that a scientist can use to demonstrate the superiority of her choice. In the absence of algorithms, however, rational men and women can arrive at opposed conclusions without obvious error.

(2) Jay Newman appears to believe that one has no right to trust one's judgment unless one can convince (all or most) informed and intelligent people of its soundness. Newman, for example, has no right to trust the factory girl's argument if he cannot convince people like Montaigne of its correctness. This (as I shall argue in Chapter 4) sets an impossibly high standard. It implies, for example, that we have no right to beliefs on philosophically disputed matters. It also assumes that the special qualifications that Newman believes we need to discern the argument's force are epistemically irrelevant. Newman thinks that people like Montaigne fail to appreciate the argument because their conscience and religious sensibilities have been blunted. To assume that the argument must convince every informed and intelligent reasoner implies that this qualification is not epistemically required. The assumption is therefore question-begging.

(3) Jay Newman correctly points out that the Christian's opponents, too, can appeal to special qualifications. But what follows? That he will not be able to convince everyone? That he cannot provide a non-question-begging defense of his epistemic position? Probably. Whether these consequences are disastrous, though, depends on whether it is reasonable to require universal agreement or non-question-begging justifications of basic epistemic practices. I have suggested that the first sets an impossibly high standard. William Alston has shown that the second does so as well.³⁹ Sense perceptual practices, for instance, cannot be justified without epistemic circularity.

Even if these points are granted, however, disagreement among illative reasoners—and the role passion, need, and emotion play in reasoning—are troublesome. Why, when all is said and done, should we trust our illative powers? Why think they are reliable? We will address this question in the next section.

THE RELIABILITY OF THE ILLATIVE SENSE

It is essential to Newman's position that illative reasoning be natural. (This is why he wants to demonstrate its pervasiveness.) Epistemic positions that ignore the way in which the mind actually works are empty. An evidentialist such as Locke, for example, insists on stateable evidence, demands that arguments be based on universally received premises, and so on. But in doing so, Locke is governed by “his own ideal of how the mind ought to act, instead of interrogating human nature, as an existing thing, as it is found in the world.” This is an error, for an examination of the mind as an existing fact is needed not only to discover what “our constitutive faculties” are actually like but also to determine “our proper condition” (GA 139, my emphasis). Newman assumes that our “own living personal reason” can be in either a “healthy” or unhealthy condition (GA 239). He says, for example, that the “supra-logical judgment, which is the warrant for our certitude about” conclusions “throughout the range of concrete matter” is an expression of “the true healthy action of our ratiocinative powers” (GA 251). But he also notes that our epistemic powers can be “enfeebled, obstructed, perverted, by allurements of sense and the supremacy of self” (GA 247). Indeed, in “ordinary minds,” they are often “biased and degraded by prejudice, passion and self-interest” (GA 261). The following example illustrates these points.

Because God's hand in nature and human affairs is hidden, it is “possible without absurdity to deny His will, His attributes, His existence.” But the reason we only “glean … faint and fragmentary glimpses of Him” is that we are “alienated from Him” (GA 309-10). The minds of those who believe “in God and in a future judgment,” on the other hand, “are in the normal condition of human nature” (GA 379). Conscience is part of our natural noetic equipment and (when it has not been blunted) clearly witnesses to God. A person whose conscience is sensitive will also assess the evidence for revelation differently from one whose conscience is not. Those who feel no need of a revelation will “come to the examination … as calmly and dispassionately as if they were judges in a court of law, or inquiring into points of science” (“FWS” [“Faith Without Sight,” in Parochial Sermons] 21). A person who desires and needs a revelation, on the other hand, “will show his caution” and prudence in believing and obeying, even though the revelation “might be more clearly attested. If it is but slightly probable that rejection of the Gospel will involve his eternal ruin,” he judges it “safest and wisest to act as if it were certain” (“FWS” 23).

Which attitude is proper? The need of and desire for a revelation are natural effects of conscience. We crave to know more clearly the one who speaks to us in conscience, and guilt cries out for atonement. Because a sensitive conscience is natural and proper, so too are this need and desire.

But how do we know that a sensitive conscience is natural and proper? And more generally, how do we know when the mind is functioning as it ought? Not just by learning how people reason, for they often reason badly. “Reason actually and historically,” “reason in the educated intellect of England, France, and Germany,” and indeed in “every civilization through the world which is under the influence of the European mind,” “reason in fact and concretely in fallen man,” is perverted by false principles (TP 143). How, then, discover the nature of proper functioning? Newman's answer is not clear but seems to be: “By determining what uses of our powers contribute to human flourishing.” A thing's natural powers are “suitable to it, and subserve its existence.” Each species find its “good in the use” of its “particular nature” (GA 273). If it does, then our faculties are functioning properly when they contribute to our well-being. Because a developed conscience is essential to human flourishing, a belief in God and an openness to revelation (which are natural effects of conscience) are expressions of properly functioning epistemic capacities.

An examination of the way in which we actually think shows that illative reasoning is natural, for “that is to be accounted a normal operation of our nature, which men in general do actually instance” (GA 270). Our illative capacities, then, are part of our natural noetic equipment. Yet why should we trust them? For three reasons. The first is “necessity.” “Our being, with its faculties … is a fact not admitting of question, all things being of necessity referred to it, not it to other things.” Indeed, “there is no medium between using my faculties, as I have them, and flinging myself upon the external world according to the random impulse of the moment” (GA 272).

The second is “interest.” “It is a general law that, whatever is found as a function or attribute of any class of beings, or is natural to it, is in its substance suitable to it and subserves its existence.” Each species thus finds its “good” in the use of its “particular nature” (GA 273). Because illative reasoning is natural to us, its proper use subserves our existence and contributes to our good.

The third reason is providence. “The laws of the mind are the expression, not of a mere constituted order, but of His will” (GA 275). “A Good Providence watches over us” and “blesses such means of argument as it has pleased Him to give us … if we use them duly for those ends for which He has given them” (GA 320-21). Confident in the divine providence, “we may securely take them as they are, and use them as we find them” (GA 275).⁴⁰

How convincing is this? David Pailin argues that even though “illation is natural, it is still liable to err. This liability declines as our experience and practice increase but it never completely disappears. The practice of illation, therefore, does not confirm Newman's general justification of its fidelity by reference to its naturalness, and to the blessings of divine Providence.”⁴¹ As it stands, Pailin's contention is false. Some liability to error is compatible with a faculty's general reliability. As Newman says, it is natural (and reasonable) to trust the senses, even though we know they sometimes deceive us. “Again, we [rightly] rely implicitly on our memory, and that, too, in spite of its being obviously unstable and treacherous. … The same remarks apply to our assumption of the fidelity of our reasoning powers” (“FR” [“The Nature of Faith in Relation to Reason,” in Fifteen Sermons Preached before the University of Oxford] 213-14).

It is true, though, that a noetic faculty should be distrusted if the trained capacity is prone to error. Wide disagreement among those with cultivated noetic capacities would be a clear indication of unreliability. Newman implicitly agrees; properly functioning noetic equipment should lead to widespread agreement. “If religion,” for example, “is consequent upon reason and at the same time for all men, there must be reasons producible sufficient for the rational conviction of every individual” (TP 86). Furthermore, “since the grounds are to be such as apply to all classes of men, they must lie deep in the constitution of our nature.” They must also “be obvious and not abstract: of a natural persuasiveness, of a nature to be intelligible to and to arrest the attention of all, and to touch them and come home to them, and work upon them” (TP 87). (Although he does not say so, Newman is clearly thinking of conscience.)

Properly used noetic faculties attain truth for the most part. “Assents to things as true, made under a sense of duty, and the guidance of the judgment, generally succeed in having TP 121-22). A widespread failure in attaining “objects in substance true” thus calls for explanation. But Newman of course has one. Religious disagreement, for instance, is explained by blunted consciences and, ultimately, original sin. Whether these explanations can be defended without circularity is another matter.

The problem of circularity also arises in another connection. If illative reasoning is necessary, we are pragmatically justified in employing it. Ought implies can. If illative reasoning is unavoidable, I cannot be faulted for engaging in it. That it is in my interest is another reason for thinking that illative reasoning is practically or pragmatically rational. But is it epistemically reasonable? Does it track the truth? It is, and does, if it is an “expression … of His will.” For God is not a deceiver.

Yet how do I know that my noetic capacities are a gift of providence? By deploying my illative capacities. “Since one of their very functions is to tell me of Him, they throw a reflex light upon themselves” (GA 275). By employing them, I learn of God's providence and thus acquire a reason for trusting them. The force of this reason, however, is not clear. For there is an obvious circularity; my justification employs the capacities whose credentials are in question. …

NEWMAN AND EVIDENTIALISM

Evidentialists such as Locke think that religious beliefs are not rationally held unless they are based on good evidence. Newman's position is closer to evidentialism than his criticism of Locke suggests.

He clearly thinks that faith is rational. “Faith must rest on reason, nay even in the case of children and of the most ignorant and dull peasant, wherever faith is living and loving” (TP 86). Yet is faith an expression of reasoning? Is it, in other words, based on evidence? Some passages suggest not. Faith is swayed “by its own hopes, fears, and existing opinions” as well as by “the actual evidence producible in its favour” (“FRC” 187-88). And in commenting on the definition of faith in Hebrews, Newman says, “its desire [i.e., hope] is its main evidence; or … makes its own evidence … not that it has no grounds in Reason, that is, in evidence; but because it is satisfied with so much less than would be necessary, were it not for the bias of the mind, that to the world its evidence seems like nothing” (“FRC” 190-91). Or again, “faith … is an … exercise of Reason” which has “grounds besides” “the actual evidence” (“FR” 207-08). Although faith is rational, it seems to go beyond the evidence.

But the impression these passages create is misleading, for by “actual evidence,” Newman means evidence that is easily stateable and widely accessible; faith's reasonings may appear illogical to the skeptic but only because “the subject matter” is “more or less special and recondite” and “the premises undeveloped” (“FR” 208). “Faith … does not demand evidence so strong as is necessary for what is commonly considered a rational conviction … and why? … because it is mainly swayed by antecedent considerations … previous notices, prepossessions, and (in a good sense of the word) prejudices” rather than by “direct and definite proof” (“FRC” 187, my emphasis). These “prejudices” and “prepossessions” determine our view of prior probabilities and the evidence's weight. For “probabilities have no definite ascertained value, and are reducible to no scientific standard, what are such to each individual,” therefore, “depends on his moral temperament. … Man is [thus] responsible for his faith, because he is responsible for his likings and dislikings, his hopes and his opinions, on all of which his faith depends” (“FRC” 191-92).

Newman's position appears to be this. Faith rests on evidence. But some of it cannot be easily recovered or stated, and the evidence as a whole is likely to seem weak to those with some moral temperaments. Newman is not an evidentialist if evidentialism requires stateable and publicly accessible evidence that compels assent regardless of a person's moral temper. Evidentialists have usually insisted on evidence of this kind. Newman nonetheless believes that properly formed religious beliefs are based on (sufficient) evidence, and that those whose noetic faculties are functioning as they should will find this evidence convincing.

Newman's contribution to the tradition Edwards represents is his demonstration that the way in which the mind reasons when influenced by religious sentiments, images, and ideas is the way in which it reasons on ordinary occasions. “Moral evidence and moral certitude are all that we can attain, not only in the case of ethical and spiritual subjects, such as religion, but of terrestrial and cosmical [i.e., scientific] questions also” (GA 252). In both cases reasoners employ cumulative case arguments or inferences to the best explanation. Although it is true that the “moral state of the parties inquiring or disputing” is not relevant in subjects such as astronomy (GA 253), its relevance is not restricted to “spiritual subjects,” for passional factors play a role in historical inquiry, philosophy, and everyday reasoning. If the way in which theists assess evidence is suspect, then so too is the way in which historians, philosophers, and ordinary practical reasoners do so, for their procedures are essentially similar. The theist's critics, therefore, must either discredit the sort of thinking that goes on in cases such as these or provide special reasons for believing that the theist's sentiments, modes of imagination, and ideas are suspect, that is, that the moral state that leads to positive assessments of the theistic evidence is likely to impair reasoning rather than to enhance it.

There is one significant difference from Edwards. If I understand Newman correctly, supernatural principles are not needed to grasp (any) religious truths.

Newman calls the kind of religious belief we have been discussing “fides acquisita” or “fides humana” and distinguishes it from “fidis divina” (TP 38). The latter involves not only “an opinion or the belief about an alleged fact or truth” but also “a determination TP 7). That is, it implies the absence of fear as well as doubt. One not only has no present doubts; one has no fear that the evidence will fail one in the future. Divine faith thus involves “a transcendent adhesion of mind, intellectual and moral, and a special self-protection, beyond the operation of … ordinary laws of thought” (GA 156). It also “follows on a divine announcement” (GA 156). Human faith “does not necessarily suppose a speaker” (TP 38), but faith “in its theological sense” believes things “because God has revealed them” (GA 94). Divine faith thus involves a divinely assisted commitment firmly to maintain one's assent or certitude in the face of any doubts that may arise, and an assent to the propositions one believes as revealed, that is, one assents to them as one assents to what a deeply trusted friend tells one.⁴²

Human and divine faith are both expressions of reason, and the type of reasoning and evidence are the same in both cases. In instances of divine faith, God strengthens the will (the resolve to maintain belief in the face of pressure and doubt) and enables the mind to assent to propositions as revealed. But he does not (as far as I can see) supply the mind with new data like the simple idea of true beauty.⁴³ Furthermore, all religious dogmas are accessible to natural reason because natural reason can be firmly convinced that they have been revealed and should therefore be believed. The difference between human and divine faith is thus a difference in the manner of belief, and not in its content, its evidence, or the process of reasoning by which it is formed. In this sense religious knowing is “naturalized.” Even if charity (and thus a new supernatural principle) is needed to appreciate the full force of the evidence for revelation, divine faith does not involve a “new sense of the heart” or “spiritual perception.”⁴⁴

Notes

1. John Henry Newman, An Essay in Aid of a Grammar of Assent (London 1870; reprint, Notre Dame, Ind.: University of Notre Dame Press, 1979), hereafter GA. Relevant discussions can also be found in several of Newman's other works: The Theological Papers of John Henry Newman on Faith and Certainty, ed. Hugo M. de Achaval and J. Derek Holmes (Oxford: Clarendon, 1976), hereafter TP; “Faith and Reason, Contrasted as Habits of Mind,” “The Nature of Faith in Relation to Reason,” and “Love the Safeguard of Faith against Superstition,” in Fifteen Sermons Preached before the University of Oxford (Oxford, 1843; reprint, Westminster, Md.: Christian Classics, 1966), hereafter “FRC”, “FR,” and “LSF,” respectively; and “Faith without Sight,” in Parochial Sermons, vol. 2 (London 1834-42; reprint, London: J. G. F. & J. Rivington, 1844), hereafter “FWS.”

2. Many, if not most, of the sample arguments in the Grammar of Assent are inferences to the best explanation. Whether they all are is doubtful. The factory girl's argument, for example, does not clearly fit this pattern. (Although one could argue that the conviction that there is a God to wipe away all tears from all eyes represents her attempt to make the most sense of the factors of her unfortunate life.).

3. David A. Pailin, The Way to Faith (London: Epworth Press, 1969), p. 144.

4. M. Jamie Ferreira, Doubt and Religious Commitment (Oxford: Clarendon, 1980), p. 35.

5. “Unless we had the right … of ruling that propositions were irrelevant or absurd, I do not see how we could conduct an argument at all; our way would … be blocked up by extravagant principles and theories, gratuitous hypotheses, … unsupported statements, and incredible facts” (GA 293).

6. This is not the same as assigning weights. A fact we regard as certain and relevant may be given little weight.

7. Although the line between estimating antecedent probabilities, viewing the evidence from the perspective of certain assumptions, and so on, and assessing the argument's force is not sharp, they are in principle distinct. One's assessment of an argument's overall force is a function of one's assessment of prior probabilities, the assumptions with which one approaches the evidence, one's judgments of relevance, one's assessment of the degree to which the premises support the conclusion if true (i.e., of how probable they make it), and so on. It is not, however, just the sum of these things.

8. Or, more accurately, they can only get a firm grip on the inadequate verbal (symbolic) representations that act as surrogates for our concrete reasonings.

9. Those schematized in the formal argument that is an abstraction from it?

10. The point is not that these principles and points of view are not subject to argument, for they are. Supporting arguments, however, must themselves be assessed by the illative sense, and this assessment, too, will reflect “tacit understandings,” “vague … notions of ‘reasonableness’,” and so on.

11. Although this is not Newman's point in these passages; at this juncture, he is trying to show that assent and inference are different mental acts.

12. Newman nowhere clearly makes this point, but I feel confident that he would endorse it.

13. John Hick, Faith and Knowledge (Ithaca: Cornell University Press, 1957), pp. 87-88.

14. H. H. Price, Belief (London: Allen & Unwin, 1969), p. 330. One really apprehends a proposition by dwelling on instances. “What matters,” in these cases, “is the degree of detail or specificity which these imaginative ruminations have.” Sensory imagery is not essential, although its presence may heighten the act of imagination's psychological effects (Price, Belief, p. 343). A person does not literally form a picture of God, for example, but “imagine[s] God as being His Lord and Master, as being one with whom he has personal relations, and as being a proper object of both fear and love” (p. 348).

15. He denies, for example, that apparently irresolvable disagreements reflecting different points of view “prove that there is no objective truth … or that we are not responsible for the associations which we attach, and the relations which we assign, to the objects of the intellect” (GA 293).

16. Presumably because (rational) self-interest induces reasoners to proceed cautiously and prudently; it prevents them from basing decisions on idle speculation, emotion, fear, hope and so on. (Emotion, hope, fear, etc., can be epistemically beneficial but they are epistemically pernicious in cases like this one.)

17. Cf. C. Stephen Evans, “Kierkegaard and Plantinga on Belief in God: Subjectivity as the Ground of Properly Basic Religious Beliefs,” Faith and Philosophy 5 (1988), 25-39.

18. Newman provides only one plausible example of how love does this. Love precludes “the worship of evil spirits” by causing “the mind to recoil from cruelty, impurity, and the assumption of divine power, though coming with ever so superhuman a claim” (“LSF” 240-41).

19. “Dispassionateness, a judicial temper, clear headedness, and candour” (GA 331). Presumably dispassionateness and neutrality are the culprits.

20. One does not need divine assistance to assent to truths of natural religion. Everyone whose conscience is not blunted discerns them. Nor is assenting to revelation on the basis of cumulative case reasoning supernatural. What is supernatural (a gift of grace) is “1. the assent to the fitness of believing” (i.e., seeing that it is good to do so), “2. the wish to believe,” and “3. the act

21. Robert Holyer, “Religious Certainty and the Imagination: An Interpretation of J. H. Newman,” The Thomist 50 (July 1986), 395-416.

22. Ibid., pp. 411-12.

23. Ibid., p. 415.

24. There is an ambiguity in Newman that is worth noting although not important for our purposes. Informal reasoning, like all reasoning, can go wrong in two ways. Its “antecedent” may be false or delusive or its inferences invalid. But reasoning from improper antecedents is no more a failure of illative reason in the strict sense than reasoning from false premises is a failure of one's deductive powers. “Intuition” must be distinguished from reasoning. “There is a faculty in the mind which acts as a complement to reasoning, and as having truth for its direct object thereby secures its use for rightful purposes. This faculty, viewed in its relation to religion, is … the moral sense; but it has a wider subject matter than religion, and a more comprehensive office and scope, as being ‘the apprehension of first principles,’ and Aristotle has taught me to call it nous or the noetic faculty” (TP 152-53). Reason is used rightly when “its antecedents are chosen rightly by the divinely enlightened mind, being such as intuitions [i.e., “the apprehension of first principles”], dictates of conscience, the inspired Word, the decisions of the Church, and the like.” It is used wrongly “when its antecedents are determined by pride, self-trust, unbelief, human affection, narrow self-interest, bad education,” and so on (TP 154). If reason habitually starts from false premises, “the mind will be in a state of melancholy disorder.” Nevertheless, reason as such is not subverted because “an enemy of the truth has availed itself of it for evil purposes” (TP 142-43) It seems to me, however, that in the Grammar “intuitions” and the “dictates of conscience” are sometimes included within the scope of the illative sense.

25. Jay Newman, “Cardinal Newman's ‘Factory-Girl Argument,’” Proceedings of the American Catholic Philosophical Association 46 (1972), 71-77.

26. Jay Newman, The Mental Philosophy of John Henry Newman (Waterloo, Ontario: Wilfred Laurier University Press, 1986), pp. 150-51.

27. Ibid., p. 157.

28. Which Newman does not deny.

29. That is, as statistical extrapolations, inferences from past regularities, and so on.

30. William Ward, “The Reasonable Basis of Certitude,” in The Ethics of Belief Debate, ed. Gerald D. McCarthy (Atlanta, Ga.: Scholars Press, 1986), p. 177. The essay originally appeared in the Nineteenth Century (1878).

31. Hoyler, “Religious Certainty,” p. 411.

32. Newman's suggestion that “an uncertain or unsound TP 162) could be read as supporting Holyer's contention. I think it more likely, however, that what Newman intends to say is that imaginatively grasping “imperfect” arguments like these can provide epistemic access to God. “Imperfect” may simply mean “lacks the demonstrative force that metaphysicians require of their arguments.”

33. Of course one would have to show that these images are inappropriate, that they caricature Christian life and doctrine.

34. Wilfred Philip Ward, The Life of John Henry Cardinal Newman, vol. 2 (London: Longmans, Green, 1912), p. 489.

35. Jay Newman, “Cardinal Newman's ‘Factory-Girl Argument,’” pp. 74-75.

36. Jay Newman, The Mental Philosophy of John Henry Newman, p. 84.

37. Ibid., p. 153.

38. Ibid., p. 175.

39. See, for example, his “Epistemic Circularity,” Philosophy and Phenomenological Research 47 (1986), 1-30.

40. In Skepticism and Reasonable Doubt: The British Naturalist Tradition in Wilkens, Hume, Reid, and Newman (Oxford: The Clarendon Press, 1986), M. Jamie Ferreira argues that Newman came to believe that we could not (coherently) mistrust our faculties; talk of “trusting” them is therefore misplaced. As a result, whereas Newman had once grounded our faculties' reliability in a belief in Providence, he effectually ceased to do so in the Grammar. To support this contention Ferreira cites four passages. (a) “I should be bound by them [“the laws of the mind”] even if they were not His laws” (GA 275). (b) “First principles come either ‘from heaven, or from the nature of things, or from the nature of man’” (Ferreira, Skepticism, p. 222, her emphasis. The internal quote is from Newman, The Present Position of Catholics in England [London: Longmans, Green, 1908], p. 293). (c) “It is enough for the proof of the value and authority of any function which I possess, to be able to pronounce that it is natural” (GA 273). (d) “That is to be accounted a normal operation of our nature, which men in general do actually instance. That is a law of our minds, which is exemplified in action on a large scale, whether a priori it ought to be a law or no” (GA 270). I do not find this convincing. (1) The necessity of using our faculties may make (the practice of) skepticism pragmatically irrational; it does not make it epistemically irrational. Nor is it clear that Newman thought that it did. (2) That a practice is natural (necessary) could be a sufficient epistemic warrant even if four faculties would not have been trustworthy if God had not created them. For if God has created them, that they are operating normally is a sufficient condition for trusting them. B can be a sufficient condition of C even if A is a sufficient condition of B's being a sufficient condition of C and B would not have been a sufficient condition of C if not-A. (B r C, A r [B r C], and not-A r not-[B r C] can all be true.) (3) The most straightforward reading of the Grammar is that Newman thought there were several grounds for trusting our faculties, each of which is sufficient. (4) The passages Ferreira cites are inconclusive. (d) does not clearly say anything about the reliability of our faculties. (b) and (c) are perfectly compatible with my interpretation. (The disjunction in [b] need not be read exclusively.) (a) could be read as asserting that I would be forced to rely on my faculties (in which case doing so would be pragmatically rational), even if God had not created them.

41. Pailin, Way of Faith, p. 150.

42. This should be distinguished from assenting to a proposition because one has good evidence that one's friend or some other trustworthy person has asserted it.

43. I am assuming that assenting to a proposition as revealed does not involve a new quasi-perceptual awareness of the divine revealer; if it does, divine faith involves new data. However, as Newman describes it, assenting to a proposition as revealed seems more like a gestalt switch than a new perception.

44. Newman's account of divine faith's epistemic (as distinguished from its moral and spiritual) dimension is thus essentially the same as Edwards's account of religious beliefs which are expressions of common grace.