Maimonides: Reason and the Limits of Language about God

“unable truly to comprehend Him”

Maimonides, The Guide for the Perplexed, I.59, a short fragment from M. Friedländer's English translation; CCEL. The context concerns the limits of comprehending God's essence, not the rejection of philosophical inquiry.

About this article

Naming the unknown is sometimes easier than admitting a lack of understanding. “Infinite wisdom” sounds confident, but what exactly does it explain? Maimonides asks whether, under the guise of God, we have described only a human being enlarged. This twelfth-century Jewish thinker joins a serious commitment to Torah with philosophical reasoning. In The Guide for the Perplexed, he limits the attribution to God of a body, component parts, and human emotions. This article examines negative theology, the difference between essence and actions, the status of biblical images, and the boundary between proof and hypothesis. We will also see why knowledge can grow through removing error without becoming a positive portrait of God's essence, and how that limit changes speech in prayer. Maimonides is not the sole voice of all Jewish thought. The practical section proposes examining one strong claim of our own: what ground it has, and where its language promises more than we know. This is an independent modern application, not a way of proving or disproving God through a short exercise.

I. A thinker at the meeting of traditions

Naming the unknown is sometimes easier than admitting that we do not understand it. All we need are majestic words: infinite wisdom, perfect power, supreme will. The sentence sounds confident, yet the question remains: what exactly does it mean? Maimonides regards such confidence with suspicion. The more familiar our description of God, the more carefully he asks whether we have described merely a human being enlarged.

Maimonides lived from 1138 to 1204. Born in Córdoba, he later lived in other regions and settled in Egypt. His work included interpretation, the systematization of Jewish law, philosophy, and medicine. For an introduction to the historical context, this article uses Kenneth Seeskin's scholarly entry in the Stanford Encyclopedia of Philosophy. It helps us keep the philosopher connected with his legal and religious work.

The Guide for the Perplexed was written in Judeo-Arabic. Its title names the situation of a reader for whom scriptural authority and philosophical reasoning appear to conflict. This is not guidance for someone simply unable to choose a goal for the week. Perplexity here involves ways of understanding taken seriously yet difficult to reconcile.

One easy answer would leave reason outside the door of religious life. Another would declare Scripture obsolete in the face of philosophy. Maimonides takes a harder path: he examines whether we understand the text correctly and whether the arguments we invoke are strong enough. Tension sometimes arises not between two truths, but between a literal image and a philosophical requirement applied without distinction.

His intellectual environment included Aristotelian philosophy and Arabic-language discussions of metaphysics. But he cannot be presented as merely translating someone else's system into religious language. He disputes, selects, and changes the way of reading. Historical distance matters: his physics and cosmology are not modern science, and philosophical proof for him is not equivalent to experimental confirmation today.

That distance makes a more precise assessment possible. We can recognize a distinction's substance without accepting the entire medieval picture of the world. Conversely, the obsolescence of some natural science does not make every question about language and the limits of knowledge meaningless. The history of philosophy requires us to see which premises sustain an argument, rather than decide the fate of a whole body of thought through the single modern label “rational” or “religious.”

II. Why a familiar image can mislead

In the Guide's first chapter, Maimonides examines the language of image and likeness. He rejects the inference that human appearance conveys God's bodily structure. His explanation concerns human intellectual capacity, not the shape of a face and limbs. Readers immediately encounter the main requirement: a word must be understood in the appropriate sense.

In ordinary speech, transferred meaning is natural. We say a decision has “ripened,” though it has no skin, or a thought “rests” on a ground, though it does not stand on the floor. An image helps express a relation. But if all the image's properties are transferred to the subject, error follows. A ripening decision implies no botany of thought; God's “hand” implies no divine anatomy. This modern linguistic analogy explains the problem without replacing Maimonides' argument.

Rejecting a literal reading need not mean abandoning the text's seriousness. On the contrary, for Maimonides, a reading that attributes bodily existence to God distorts the object of faith. People may think they are defending Scripture when they preserve its most vivid image. Yet fidelity to the image can become unfaithfulness to meaning if the image is understood in the wrong relation.

This raises a question about the interpreter. Who determines which meaning is appropriate? Does the philosopher replace the text with his own system? This is a real difficulty. We cannot assume beforehand that every nonliteral reading is deeper. An interpretation must explain why its chosen meaning fits and what in the text allows it. Otherwise, another arbitrariness replaces anthropomorphism: every inconvenient passage is declared a metaphor.

Maimonides' example is therefore interesting not for the freedom to replace words at will, but for the attempt to establish a rule. He connects reading with a reasoned conception of God. We can dispute that conception, but must then dispute the ground, rather than merely the philosopher's refusal of a familiar picture. The conversation gains substance when the precise argument changing the interpretation is visible.

III. Unity means more than the number one

Maimonides' problem begins with unity. If God is one, for him this does not merely say that the gods have been counted and the total is one. It implies the absence of composition and division. God is not assembled from body and soul, a capacity and its bearer, an essence and qualities added to it. This understanding places stricter demands on language than an ordinary reading of the singular.

In Part I, chapter 52, Maimonides considers different forms of positive description. A definition, part of a definition, a quality, and a relation create difficulties when applied to God. Their shared risk is describing Him as one describes a composite object. Words intended to praise unity can thus quietly introduce multiplicity.

To understand the logic, consider a hypothetical description of a person: he is strong, knows languages, loves music. These traits can be distinguished, can change independently, and do not exhaust his essence. Listing them is appropriate for a person. Transfer the same model to God, and we get a bearer of a set of separate capacities. Maimonides rejects precisely this structure. The problem is not that the qualities listed are insufficiently good, but the very way they are attributed.

This is an argument within a particular metaphysics. It cannot be treated as an established fact simply because it is consistent. Readers can ask why absolute simplicity should be accepted and how it relates to addressing God religiously. But before objecting, the precise claim must be understood: Maimonides does not propose removing good properties while leaving bad ones. He changes the category of description.

In this reading, unity becomes a discipline of thought. Repeating “God is one” does not yet prove that other statements preserve that meaning. Sometimes people defend a concept verbally while destroying it through their explanations. This logical problem reaches beyond religion. In any theory, particular claims need to be checked for compatibility with its central principle. Application to human systems remains a separate analogy, however: a company, person, or state does not possess Maimonides' absolute simplicity.

IV. Why “greater than us” is not enough

We might try to rescue familiar language by adding a superlative. God knows incomparably more, lives infinitely longer, has infinitely greater power. But in Part I, chapter 56, Maimonides rejects precisely such a shared scale. In his view, the difference between God and human beings is not reducible to a difference in the magnitude of one capacity.

Comparison on a scale presupposes a shared meaning of the property. Two people know different numbers of words, but “knowing a word” means roughly the same for both. If one object is hotter than another, they must have a comparable kind of temperature. Maimonides thinks divine and human knowledge cannot be joined in this way. Increasing the number therefore does not remove the categorical problem.

A hypothetical modern example: a computer stores more records than a person remembers, but increasing the number of records does not by itself describe the quality of human understanding. Measurement helps with a specific function, but does not transfer every meaning of “know” between different subjects. This is a limited analogy for the problem of comparison. It proves nothing about machine consciousness, still less about God's nature.

“We are almost the same, only on different scales” is especially dangerous here. For Maimonides, it does not express humility before infinity. It assumes in advance the shared ground he disputes. Even “infinitely” can leave our imagination with merely an excessively large, familiar bearer of power.

A difficult question follows: if words have such different meanings, how do we understand a statement at all? Replacing every term with a vaguer one cannot solve it. The farther God lies beyond ordinary categories, the more rigorously we need to explain what work language still performs. Maimonides' next step — negations and descriptions of actions — answers this question. Yet it does not remove every difficulty; it shows what content might be preserved after abandoning a vivid shared scale.

V. Negation removes error without portraying essence

In Part I, chapter 58, Maimonides proposes speaking of God through negative attributes. The point is not to list shortcomings. “Not corporeal” excludes a bodily model; “not dependent on a cause” excludes derivative existence. But removing a mistaken description does not produce a complete positive portrait of God's essence.

This boundary is easily lost. In ordinary speech, negation often helps us move quickly to a positive answer. If a person is not asleep, we may assume they are awake. The range of options is already known. But for Maimonides, the question about God is not a choice among several objects of one category. Negation should limit an unsuitable attribution, not put God into the opposite compartment of our table.

Negative theology therefore differs from “we cannot say anything at all.” It lets us claim that a particular model does not fit, and explain why. It also differs from a confident catalog of secret properties. If someone says God is “not like us” and then describes His inner experiences in detail, the initial boundary has effectively disappeared.

Philosophically, the question concerns the substance of knowledge through exclusion. Can that knowledge grow? Yes, if exclusions rest on arguments and remove actual errors. But the price is giving up the familiar feeling of having an intelligible object before our inward gaze. Readers must distinguish confidence that a description is wrong from knowledge of what an exhaustive correct image must be.

Our modern application can concern only this distinction. For example, refuting an explanation of an event is insufficient to prove our own explanation. Rejecting one hypothesis narrows the space of possibilities, but does not automatically supply a single answer. This is an ordinary logical situation, not negative theology in its full sense. Similarity of structure helps; identifying the subjects would destroy the very caution Maimonides' question teaches.

VI. Actions can be described without attributing emotions

If essence is unavailable to positive description, how can words about mercy and justice be preserved? In Part I, chapter 54, Maimonides distinguishes God's actions from the human states that usually prompt similar actions. Naming an action does not require recognizing the corresponding emotion in God. Merciful treatment and the inward experience of mercy are not identical.

For human beings, the distinction is intelligible on a less difficult level. Two physicians can perform equally careful work while experiencing different feelings. An action sometimes tells us much, but we do not always know the actor's complete inner state. Similar outcomes do not guarantee identical motives. The analogy merely explains the difference between an action and a mental state; it does not turn God into a hidden human agent.

Maimonides' limit is far more radical: emotion presupposes change, susceptibility to influence, and a structure inconsistent with his understanding of God. Readers must therefore reject the inference that described care reveals God's psychology. The language of action nevertheless remains substantive: it concerns how the world is ordered and governed as the thinker understands it.

The question also invites an objection. For many believers, closeness to God is expressed precisely through personal address, sympathy, and love. Does a strict philosophical interpretation make that closeness too abstract? This article does not settle the dispute for every participant in the tradition. It shows the subject: how to preserve a relationship's religious meaning without attributing the limits of human emotional life to God.

Specificity matters in reading. We cannot simply say “Maimonides forbids speaking about God.” He forbids or reinterprets particular modes of attribution. Nor can we claim that all Jewish tradition accepts his explanation of every image. A more cautious formulation makes his project intelligible without turning a famous thinker into the sole representative of a highly varied history. Differences then become substantive questions about the language of faith, rather than a contest between more and less rational people.

VII. Learning ignorance can take different forms

In Part I, chapter 59, Maimonides asks how different people's knowledge differs if God's essence is not comprehended. He connects progress with the reasoned removal of false attributes. Someone can investigate for years and reach an important result precisely through a more accurate refusal of an unsuitable conception.

This separates disciplined ignorance from indifference. “Who knows anyway” can mean someone has not tried to understand. “This claim cannot be accepted on these grounds” expresses work already done. Neither speaker may have a final answer, yet their situations differ. One does not distinguish the options; the other knows why some are unsound.

In the same chapter, Maimonides recounts a worshipper adding terms of praise. Rabbi Hanina objects; a parable likens such praise to praising an owner of gold for silver. The error is not simply that the praise is too slight. The person chooses an inappropriate kind of perfection. Adding epithets therefore does not solve the problem: it continues the wrong description. Maimonides separately grounds the use of established prayer words in their presence in Torah and introduction into prayer. The limit on speech here concerns the very manner of honoring God, not a literary preference for brevity.

Yet silence can be romanticized too. It does not automatically prove depth. A person may be silent from respect, fatigue, lack of an argument, or a wish to preserve untestable authority. Maimonides' discipline presupposes distinction and inquiry, not turning speechlessness into a sign of superiority. The absence of a positive definition does not remove the need to explain what we do claim.

This subject concerns public language. “Absolute,” “universal,” and “inevitable” sometimes create an impression of complete understanding. Remove the intensifier, however, and an unclear link remains between the claim and its ground. A simple question can test the sentence: what new meaning did the epithet add? If it only increased emotional force, the argument has not grown stronger. This is our modern application, not a claim that Maimonides practiced media literacy. His historical subject was speech about God; readers transfer the logical caution independently.

VIII. Reason knows the difference between an argument and a proof

In Part II, chapter 25, Maimonides begins with an unexpected concession: biblical passages about creation could be interpreted differently, like bodily descriptions of God. The mere possibility of nonliteral reading therefore does not settle the question of the world's eternity. His first argument concerns proof: he considers God's incorporeality proven, the world's eternity unproven. Sufficient ground for one reinterpretation does not become ground for another simply because the reading technique is similar.

The second argument concerns a theory's consequences. In Maimonides' account, Aristotelian eternity is linked to nature's unchanging necessity and the exclusion of miracles. He distinguishes the Platonic version: it allows miracles and does not create the same contradiction with religion, though it too remains unproven. These are two tests, not one: how strong the grounds are, and what exactly a claim changes in the understanding of revelation. Compatibility with faith does not replace proof; lack of proof does not make every theory identical in content.

In Part II, chapter 24, Maimonides discusses difficulties in his time's astronomical picture. Today's readers can value his readiness to name a difficulty, but should view the cosmology's content historically. Celestial spheres and the physical premises of that period do not become modern scientific grounds through the text's philosophical significance.

Within the chapter, a reasoned objection has a precise target. It is not enough to say that one dislikes a literal reading or finds a philosophical theory coherent. One must show why its ground demands this particular textual change. This keeps philosophical inquiry from two hasty choices: forbidding interpretation altogether or permitting any interpretation that saves a preferred hypothesis. The proposed boundary is difficult because the status of proof itself can also be disputed. This article examines the chapter's stated arguments without promising to settle the entire history of interpretations of the Guide.

A modern analogy is working with a forecast. A well-built model can be useful, but its result depends on input data and assumptions. Call a forecast a proof, and we raise the conclusion's status without adding a ground. Declare every forecast useless because of uncertainty, and we lose substantive information. Maimonides' question helps preserve different statuses, though it offers no specific modern method of statistics or science.

IX. The limit of language leaves work for reason

Maimonides asks readers to give up the easy feeling of possessing their subject. A familiar image of God may be clear precisely because it is too human. A positive epithet may seem exact because it hides composition. A superlative can enlarge the picture without changing its mistaken category. Reason's work therefore includes not only creating explanations, but also removing unsuitable ways of explaining.

Negative theology here is no ornament of mystery. It is linked to specific metaphysical requirements: unity, incorporeality, the absence of composition. Readers can discuss their grounds and consequences. But reduce the teaching to “God is incomprehensible” and its main work disappears: why certain words are limited, what descriptions of actions are allowed, and what growth of knowledge means without a positive portrait of essence.

This rigor also has a human cost. A vivid image helps us address, love, pray, and tell stories. Philosophical purification of language can feel like losing closeness. This difficulty should not be removed by mocking religious feeling or promising that all forms of faith are really identical. We face a specific project within a tradition around which serious disagreement is possible.

Its substantive value for an outside reader is the possibility of respecting the boundary of knowledge without giving up thought. Ignorance can be the result of inquiry. A metaphor can be expressive without becoming a literal description. Negating an error does not always supply a ready truth. Proof differs from a persuasive argument. These distinctions preserve meaning where majestic words easily replace completed work.

We can end not in silence about everything, but with a more exact question: what allows me to say precisely this? If the answer is known, speech gains a ground. If the grounds are insufficient, we can investigate further or honestly narrow the claim. For Maimonides, this discipline serves the understanding of God. For readers, it also recalls that respect for a subject begins with refusing to make it too like a picture convenient for us to possess.

What this means in practice

The practice below is an exercise in clear speech. It does not test religion's truth or reproduce Maimonides' entire teaching. Choose one phrase of your own that you usually say confidently: about an event's cause, a relationship's meaning, work, or how the world is ordered. It is better to begin with a small question that allows calm examination.

Write the phrase in full. Then underline words that make it maximally broad: “always,” “obviously,” “absolutely,” “everyone,” “by nature.” Ask what concrete ground supports that very breadth. Perhaps you observed several cases, read a particular text, or adopted a worldview. These grounds can matter, but they differ. Several observations do not prove “always,” and a religious conviction does not become an experimental fact through a confident tone.

Next, separate what is known from what is excluded. What can you actually describe? Which hypothesis has been refuted? If one option was rejected, have you recorded the opposite as proven without an additional step? For example, lack of support for one explanation of a conflict does not establish another person's particular motive. Sometimes an accurate formulation is more modest: “this explanation is unconfirmed; these options remain.”

The next step concerns metaphor. Are you using an image as a literal explanation? “The project is growing” can mean an increase in audience, tasks, or income. Those quantities can move in different directions. The image works while it helps us see; when a decision is needed, it helps to name the concrete change. A religious or philosophical image similarly requires us to ask what it claims and in what relation.

Finally, rewrite the phrase to preserve the substantive thought while reducing unsupported breadth. Every sentence need not become cautious mumbling. If a fact is verified, it can be named directly. The aim is to match the strength of words to the strength of grounds. After some time, see whether discussing the subject with someone who disagrees has become easier. More precise speech helps not by guaranteeing reconciliation, but by allowing the disputants to see which claim actually needs an answer.

The check can continue with a hypothetical scene. Someone says: “This project will definitely fail because nobody accepts our idea.” A forecast, a universal description of others' reactions, and a possible feeling of disappointment are joined here. Separate them. What responses were actually received? How many people have not answered? Which part of the proposal was rejected? What would success mean at all? Clarification promises no optimistic outcome. It makes the negative conclusion as strong as its grounds permit.

Then consider the opposite mistake. The absence of proven failure does not prove future success. One excessive certainty cannot be replaced by another simply because the new one is more pleasant. Perhaps the accurate formulation is: certain obstacles are known, some conditions remain unknown, and a limited next step is needed. This differs from both capitulation and a promise of inevitable victory.

Separately, check whether words have become a description of a person's hidden essence. “He is unreliable by nature” may rest on one specific unfulfilled agreement, but extends far beyond it. We can name the fact, its consequences, and the necessary boundary without claiming complete knowledge of another's character. This caution does not require tolerating a repeated breach. It allows self-protection without an additional claim that cannot be grounded.

Finally, leave room for revision in the record. What new fact would change your view? If none would, perhaps you are expressing an adopted position or a personal limit, rather than a testable forecast. These are legitimate kinds of speech, but should be named honestly. We transfer a discipline of distinction from a historical philosophical question; we do not claim that an everyday decision has the same subject or metaphysical grounds as Maimonides' negative theology.

It also helps to check whether you demand stronger grounds from someone else than from yourself. Our own claim often seems obvious, another's in need of detailed proof. Try writing both positions equally calmly and naming what supports each. If the grounds differ, compare their roles; if they are alike, do not change the standard merely from sympathy for a conclusion. The exercise does not settle the medieval dispute over knowing God. It makes modern speech more honest where confidence may have quietly replaced completed work.

Key quotation

“by negations”

Maimonides, the Guide, I.59, the paragraph on humanly accessible knowledge of God, a short fragment from Friedländer, CCEL. The expression occurs in the claim that such knowledge is attained only through negations. It concerns the reasoned exclusion of an unsuitable attribute, not an arbitrary denial of everything. The epigraph and final formula convey the tension between growing knowledge and the limit of comprehension.

Connections between traditions

These comparisons are analytical. They do not prove influence and preserve differences between the subjects the authors discuss.

Positions and sources	Connection	Limit
Maimonides, I.1, and Genesis 1:27	Philosophical interpretation tests the meaning of a biblical image.	This is an interpretive relationship within a tradition, not two independent claims about God's appearance.
Maimonides, I.54, and Shneur Zalman, Tanya, 7	Action is considered in relation to the divine, but the question is posed on different levels.	Maimonides explains God's actions without attributing human emotions to Him; Tanya examines the direction of human action. These are not two identical teachings on intention.
Maimonides, I.58–59, and the Dao De Jing, 1	A familiar name or description does not exhaust its ultimate subject.	Maimonides' incorporeal God is not identified with the Dao; an identical limit of language is not established.
Maimonides, II.25, and the Rigveda, X.129.6–7	A question about origins requires clarity about the degree of knowledge.	Distinguishing proof from an unproven theory differs from a poetic question about the beginning; the authors' answers do not coincide.
Maimonides, I.59, and Confucius, the Analects, II.17	Acknowledging a limit can express the work of understanding.	The Confucian distinction between knowledge and ignorance is not negative theology and does not presuppose Maimonides' metaphysics.

Primary sources

Kenneth Seeskin. Maimonides, Stanford Encyclopedia of Philosophy: scholarly context of his life, works, negative theology, and interpretive disputes; the thematic sections were read, not every work in the bibliography.
Moses Maimonides. The Guide for the Perplexed, English translation by M. Friedländer, second edition, 1904, electronic version at CCEL. I.1: image and likeness.
The same work, I.52: ways of attributing attributes.
The same work, I.54: actions and emotional attributes.
The same work, I.56: the absence of a shared scale for divine and human attributes.
The same work, I.58: negative attributes.
The same work, I.59: progress through negations and the limit on praise.
The same work, II.24: difficulties in medieval astronomy.
The same work, II.25: proof, hypothesis, and creation. CCEL URL numbering does not always match the chapter heading; verified headings are given here. The named chapters of the Guide were read as separate chapters of one work, not as eight independent studies.
Genesis 1:26–27, English version JPS 1917 at Mechon Mamre. Primary biblical text; the fragment was read. Role: the image as a subject of interpretation.
Shneur Zalman. Tanya, chapter 7, main text with explanations by Yosef Weinberg, Kehot/Chabad.org. The section on intention and permitted material activity was read. Role: a limited comparison within the tradition.
The Dao De Jing, chapter 1, traditionally attributed to Laozi, translated by James Legge. The chapter was read. Role: the limit of naming, without identifying the Dao with God.
The Rigveda, X.129.6–7, translated by Ralph T. H. Griffith. The entire hymn was read. Role: the poetic limit of knowledge about origins.
Confucius and his students. The Analects, II.17, English version at the Internet Classics Archive; the translator is not explicitly named on the page. The saying and nearby context were read. Role: knowledge and acknowledgment of ignorance.

About the authors

Andrey Zarinsh is the author and founder of the United Wisdom project. He researches inner transformation, the unity of traditions, intercultural dialogue, and contemporary spiritual discipline. Instagram: @andrey.zarinsh.

Maria Evilkut is a co-author of United Wisdom. Alongside her research, she seeks to test these ideas in practice and observe how the principles of unity and mutual understanding work in real life between people and cultures. Instagram: @ilkutki.