How Do Concepts Form? From Teaching Adam the Names to Scientific Models

What does the verse about teaching Adam say, and what does it not say?

The scene in Surat al-Baqarah begins with God telling the angels that He will place a successor on Earth, their question about corruption and bloodshed, and His words: “I know what you do not know.” Then come “He taught Adam all the names, then presented them to the angels,” the angels' reply, “Glory be to You! We have no knowledge except what You have taught us,” and Adam's reporting of the names [al-Baqarah: 30–33]. The argument does not conclude that the person who knows the names cannot do wrong; teaching appears in the context of a creature capable of knowledge and tested in what he does with it. The verses; Ibn Ashur's commentary.

Commentators differed about the scope of “the names.” Al-Tabari reported interpretations concerning the names of things, angels and Adam's descendants; he favored the names of his descendants and the angels, arguing from the pronoun in “presented them,” while acknowledging that a broader reading was possible in Arabic. Al-Qurtubi discussed the name, naming and the named object, and reported the view that these were the names of all things, kinds and their uses. Al-Razi considered whether the teaching encompassed things' qualities and properties, since knowing a thing's reality is higher than merely memorizing its word; this is his interpretive view, not a consensus that the verse addresses modern experimental categorization. Al-Tabari; al-Qurtubi; al-Razi.

Ibn Ashur offers a distinction useful to our subject: he discusses whether the teaching involved imparting particular names or enabling Adam to give names and express what was in his mind; he then connects the capacity for expression with teaching others and the emergence of knowledge. But he does not make settling the names of every entity that might later appear a condition for learning from the verse, nor does he regard its wording as specifying how the teaching occurred. We should distinguish his interpretation of Adam's scene from our use of that scene as an entry into thinking about knowledge that accumulates afterward. Presenting the named objects to the angels and Adam's reporting their names does not itself prove that he possessed a theory of every named object's properties. Ibn Ashur, al-Tahrir wa-l-Tanwir.

We therefore cannot extract from the word “names” alone the history of language's emergence, the structure of Adam's brain, a law of cognitive science, or the periodic table before its discovery. The book reads in the verse the meaning of teaching and the limits of a creature's knowledge; it then uses contemporary people's movement from words to investigation as an example of reflection, not an exclusive interpretation of the verse's intent. This differs from the question Did God teach Adam all languages?, which examines the linguistic and exegetical disagreement itself.

Another verse prevents us from reducing every human life to the event of teaching Adam: “God brought you out of your mothers' wombs knowing nothing, and gave you hearing, sight and hearts, so that you might be grateful” [al-Nahl: 78]. Al-Tabari explains it as a reminder of the gift of what people learn after birth through hearing, seeing and reasoning. Here the Quran recalls the source of that blessing and gratitude; it does not present a laboratory classification of brain processes. The verse; al-Tabari.

What is the difference between a name, a category, a concept and a model?

A name is a word we use to refer to something; a category groups things under a particular distinction; and a concept is a mental means of using that distinction in understanding and inference. The boundaries between these terms are not a matter of complete philosophical agreement, and knowing a name does not imply possessing a precise concept. A child might memorize “bird” and think that everything that flies is a bird, including bats and excluding ostriches. When the child learns that flight alone does not define the category, they have not merely exchanged one label for another; they have changed the basis for predictions about anatomy, reproduction and biological relatedness.

Nor is every classification a complete science. We can classify objects by color, weight, usefulness or common origin; each division may help answer one question and mislead us about another. If you group the Sun and a lamp as “light sources,” you rightly classify them with respect to illumination, but that does not mean that their mechanisms of light production are the same. A good concept does not stop at a verbal definition; it states which property it measures, when the category is useful, and which cases might undermine its generalization.

Imagine a child who has seen three birds fly and connected “bird” with flight. This is an understandable generalization; then the child sees a flying bat and an ostrich that does not fly. It is not enough for someone to teach two ready-made answers. The child must ask which traits explain the cases together, and learn that everyday categories and biological categories may serve different purposes. The distinction may begin with similarity of appearance, with anatomy, reproduction and relatedness later becoming more useful to the scientific question. This is an explanatory illustration of revising a category, not a description of a particular child's experiment or a claim that every concept is built the same way. What matters is that a counterexample changes what we count as a decisive feature; otherwise, a name becomes a cage that rejects anything unlike its first examples.

A scientific model is a selective representation of an aspect of a phenomenon: it specifies elements, relationships and conditions, then enables explanation, prediction or comparison. It may be a table, a diagram, equations or a simulation. Its value lies in what it explains and predicts, and how it handles a case contrary to its prediction, rather than in the beauty of its arrangement. Models differ in purpose and deliberately simplify; we therefore do not equate a map that works within its domain with the entire Earth. A specialist overview of models in science.

To move from a name to a theory, we need to distinguish a model from a theory. A broader theory can contain principles explaining multiple phenomena and allow different models to be built for specific cases; a successful model may in turn help revise the theory. This is not an automatic sequence through which every idea passes: some models are exploratory, some are more predictive than explanatory of causes, and a theory does not become correct merely because one of its diagrams looks convincing. We ask about the phenomenon the model targets, its assumptions, measurements, possible competitor, and a prediction that can be tested with data not used to fit it. The strength of a result differs depending on whether we compare the model only with what it was designed to explain or also test it against an independent case. These questions make “we now have an explanation” open to examination. Models in science; a paper on conditions for using model outputs in decision-making.

Questions about water can thus progress as they do in the book: water is a common name; “a liquid that dissolves many salts and has thermal properties” is a measurable description; explaining these properties through the molecule's structure and interactions moves toward deeper relationships. But the symbol H₂O is not water in a glass, and knowing the formula alone does not predict everything water does in a cell. The water and life article follows this chemical and biological example in detail; our subject here is how the inference connecting observation and model is built.

Does an idea need a word before it can exist?

It is incorrect to say that a child who cannot yet speak never distinguishes a category. Infant research indicates forms of discrimination and generalization before the ability to pronounce names. But a test of an infant's looking at a new image does not by itself reveal a richly developed “concept” like an adult's; the result depends on the experimental method and whether the distinction existed beforehand or arose during the experiment. Researchers therefore distinguish the presence of categorization performance from the claim that they fully understand its internal mechanism. A review of infant categorization; a critical review of measurement methods.

Nevertheless, words affect some learning conditions. In Althaus and Mareschal's experiments with new visual categories, hearing a shared name directed children around one year old toward the feature shared by different examples faster than in some comparison conditions. This does not mean that the word poured “the essence of the category” into the mind; it may shift gaze and attention, then help the child extract a similarity that might be reached in other ways. A comparative experiment also found that visible movement was more useful for learning the category than a name in that task. The effect depends on the question, design and experience; it is not a law stating that language is always stronger than perception. The experiment on labels and attention; the experiment comparing motion and labels.

Categorization is not a privilege shared by no other creature. In an experiment, pigeons were trained to distinguish drawings of birds and mammals, then tested with familiar and new drawings and composite forms. The discrimination transferred to some new examples, but researchers also found that the orientation of the visual form might explain a large part of their performance; we cannot automatically attribute to them a zoologist's understanding of biological relatedness. We do not infer equivalence between this performance and forming a chemical theory, nor should we deny animals every ability to distinguish things in order to establish human worth. The distinction the book discusses is broader: humans' ability to connect shared naming with explanation, measurement, writing, criticism and accumulation across generations, with moral responsibility for the use of knowledge. Cook and colleagues' experiment.

The book's example on p. 39, “An animal sees water and drinks, while a human sees water and names it,” therefore needs careful reading. It illustrates the widening of human investigation from water to its chemistry; it does not serve as proof that animals do not categorize, or that naming is the first prerequisite for all perception. Correcting this boundary strengthens the book's argument: humans are distinctive not because they alone see patterns, but because they can build on them a network of explanations that can be shared and revised.

When does classification become prediction? The periodic table

Before the periodic table, many elements were known by their names and scattered properties. Information about their masses and reactions was collected, and multiple scientists searched for regularities among them. Dmitri Mendeleev was not the first to arrange the elements, but he left positions in his table for elements not yet discovered and inferred some of their properties from neighboring patterns. A famous example is “eka-aluminium,” whose predictions were compared with measurements after gallium's discovery in 1875; scandium and germanium were discovered afterward. Organization became more than a store of names because it offered a prediction before some observations. History of the periodic table, Royal Society of Chemistry.

The meaning of “prediction” can be seen in two figures rather than merely in celebrating Mendeleev's name: the Royal Society of Chemistry's historical comparison table shows an atomic weight he predicted of approximately 68 against 69.72 for gallium, and a predicted solid density of 6.0 g/cm³ against a displayed value of 5.9 g/cm³. These are specific, verifiable comparisons; they do not mean that all his predictions were right or that numerical accuracy was the same for every property. The difference between “we found a pattern after discovery” and “we said something before finding the element and then checked it” is a major methodological distinction. The historical comparison table.

But the example becomes misleading if told as though Mendeleev knew everything or his historical table was final. Early ordering relied on atomic masses and known properties, with difficult positions and uneven predictions; later work made it possible to understand the modern ordering by atomic number. The correction does not erase the successful prediction of gallium; it shows how science progresses: a model organizes some observations, succeeds in an independent prediction, and then the basis of its regularity is reinterpreted using new evidence. The Royal Society of Chemistry; an account of atomic number and the periodic law.

Page 42 of the book says that “a name within a system” becomes a position, a relationship and a prediction. This is the core of the example, with an additional qualification: a neatly placed gap in a table is not enough; the real test is finding an element with properties close to what was stated in advance. If a prediction fails, we examine the measurement, hypothesis and system's limits rather than protect the model by changing the meaning of words after every result.

How can we know something we cannot touch? Starlight

On pp. 39–42, the book gives the example of a star: it begins as a bright point to the eye, then its light becomes material for measurement. A spectrograph separates light by wavelength, revealing intensity patterns and absorption or emission lines. These patterns are compared with laboratory measurements of materials; together with the star's temperature and density conditions, the lines allow inferences about elements in its atmosphere and their approximate abundances. Researchers do not reach out and touch the star, but neither do they “guess from its name”; they move from a measured trace to an interpretation tested in light and matter. NASA's explanation of spectroscopy.

The reader can distinguish four layers here: “star” is a name; “it has light with a spectrum” is an observation; “this pattern matches an element's lines under certain conditions” is an inference; and “its atmosphere's composition, velocity and temperature” are conclusions within a physical model. Not every isolated line is sufficient evidence, and a colored image does not provide the star's composition without calibration, comparison and assumptions. Scientific inference is strong because it shows its path and the limits of its accuracy, not because it exactly resembles direct observation.

The periodic table and spectrum may seem distant from the verse about names. Their connection is with the book's idea of seeking knowledge, not with a claim that the verse predicted them: a human moves from a limited word to a measurement others can examine, then acknowledges that the knowledge reached is acquired and partial. If learning things' names in al-Baqarah appears within an account of the wisdom of Adam's succession, reading a star's lines today is an empirical activity performed by people using instruments and evidence, not a specific interpretation of God's words “He taught Adam.”

Three mistakes that undermine the path from words to knowledge

Confusing naming with explaining. Naming a disease or phenomenon makes it easier to refer to; it does not establish its cause. A name may group unlike cases that need to be separated. The more important question is: what do we measure, and what finding would make us change the category? This applies to common scientific names as much as to any attractive internet headline.

Jumping from linguistic resemblance to scientific discovery. An old expression can resemble a new term, but its wording does not automatically contain the measurement method, margin of error or history of experiments. We therefore do not attribute water's polarity or the mechanism of spectroscopy to a verse that does not mention them. Reflection on creation is legitimate; attributing a specific meaning to the Quran requires exegetical evidence independent of our desire to prove scientific priority. See the method of reading the Quran and science.

Confusing a model's success with its completeness. Many measurements may justify relying on it within a known domain, and then a case appears that requires revision. Pages 42–43 of the book say that humans read regularity in creation rather than bring it into existence from nothing; in scientific terms, their formulation of a law is a human tool for describing and testing that regularity, not the regularity itself in all its details. Respect for science requires presenting both its real successes and its real limits.

What does responsibility mean for knowledge?

In Adam's story, the angels say, “We have no knowledge except what You have taught us” [al-Baqarah: 32]. This acknowledges the limits of knowledge in its exegetical context; a believer may draw humility in research from it without turning it into a prescribed experimental protocol. Another verse says, “Do not follow what you have no knowledge of. Hearing, sight and the heart—all of these will be questioned” [al-Isra: 36]. Al-Tabari explains among its meanings that a person should not say what they do not know, or claim to have seen or heard what they have not seen or heard. The verse is not intended to explain the modern “scientific method,” but its ethical meaning concerns the integrity of someone reporting information or announcing a result. Al-Tabari's commentary.

In a scientific article we should write: this is an observation, this is its likely interpretation, this is an open hypothesis, and this is a statement from the book or a commentator rather than a laboratory experiment. The reader has the right to find a reference that lets them review each transition. Numerous terms are not evidence of depth if they conceal uncertainty; depth means knowing why a piece of information is insufficient for a conclusion larger than itself.

This places the act of naming itself under scrutiny. Giving a category a name suggestive of certainty may make different members appear to be one thing, or turn a preliminary estimate into a final judgment. When we use a model for a decision affecting people, its conditions, errors and consequences become part of the trust we bear. “Qul Siru fi al-Ard” therefore does not end at praising the mind's abilities; its second chapter connects knowledge with succession, then with the pen and preserving knowledge, and afterward with justice in civilization. Writing and civilization's memory examines transmitting and preserving knowledge, while this article examines forming and testing it.

Who holds the name, and who tests the idea?

We can now return to the water at the beginning. A name lets us ask about it together; a concept specifies which properties we want to understand; and a model proposes a relationship open to examination. If an unexpected case appears, we do not declare reality mistaken because it fails to resemble our diagram. We revisit the category's definition, measurement and relationship. “Observe” thus becomes a responsible activity: looking at the thing, and looking again at how we know it.

There is a strength worth highlighting in pp. 39–43 of the book: knowledge is not the number of words someone possesses, but the ability to move from observation to a question another person can test, then direct what has been learned. Its figurative phrases need the boundaries we have encountered: naming things does not create them; animals and children before speech are not without categorization; and the verse about teaching Adam does not compress the history of science into an equation. With these boundaries, the meaning that serves the book remains firmer: teaching that leads to revisable knowledge, and knowledge that does not exempt its holder from responsibility.

Frequently asked questions

What is concept formation?

It is building a distinction that lets us group different examples and infer or predict something about them, with the possibility of revising that distinction when a counterexample appears. A concept's depth varies with what someone knows and can test.

Does a word create an idea?

Not in general. Nonlinguistic categorization can occur, but words may direct attention and help stabilize some distinctions and teach them to others. Their effect varies with task and experience.

Does “He taught Adam all the names” describe modern cognitive science?

The verse does not explain experiments or brain mechanisms. It is part of the scene of succession, teaching and revealing the limits of a creature's knowledge; using cognitive research to reflect on knowledge's value is a later reading, not an exclusive interpretation of the wording.

How does a concept differ from a scientific model?

The concept “element” may define the kind of things we investigate; the periodic table's model organizes relationships among elements and enables predictions. A model requires its domain and method of testing to be specified.

Is a scientific model the same as a theory?

A broad theory may address a class of phenomena, while a model formulates one aspect of it under specified conditions to explain a case or predict a result. Their relationship is neither a fixed hierarchy nor complete identity.

Does Mendeleev's prediction of unknown elements prove his classification was correct in every respect?

It is important evidence for the strength of a particular pattern, especially when some predicted properties matched later measurements. But it does not make every detail of the 1869 table immune to revision; the explanation developed into ordering by atomic number.

Why did the book connect knowledge with responsibility?

Because the ability to name, measure and use things can serve people or harm them. In al-Baqarah's context, teaching does not erase the question of corruption; it makes what we do with what we have learned a question that cannot be set aside.

Sources and limits of review

1. The Quran: al-Baqarah 30–33, al-Nahl 78, al-Isra 36.
2. Al-Tabari, commentary on al-Baqarah 31; al-Nahl 78; al-Isra 36.
3. Al-Qurtubi, commentary on al-Baqarah 31.
4. Al-Razi, commentary on al-Baqarah 31.
5. Ibn Ashur, commentary on al-Baqarah 31.
6. Ahmed Alhafiz, Qul Siru fi al-Ard: Observe How Creation Began, chapter two, “The Taught Human,” pp. 39–43. The direct source of the line of inquiry, not an empirical reference or a substitute for exegesis.
7. “Categorization in infancy,” Trends in Cognitive Sciences (2001); Alessandroni & Rodríguez, review of methods for studying early categorization (2020).
8. Althaus & Mareschal, “Labels Direct Infants’ Attention to Commonalities during Novel Category Learning,” PLOS ONE (2014).
9. “Linguistic Labels, Dynamic Visual Features, and Attention in Infant Category Learning,” Journal of Experimental Child Psychology (2015).
10. Cook, Wright & Drachman, “Categorization of Birds, Mammals, and Chimeras by Pigeons,” Behavioural Processes (2013).
11. Royal Society of Chemistry, history of the periodic table's development; OpenStax, the modern periodic law.
12. NASA Science, explanation of spectroscopy and stellar spectra.
13. Frigg & Hartmann, “Models in Science,” Stanford Encyclopedia of Philosophy.
14. “Three questions to ask before using model outputs for decision support,” Nature Communications (2020).