The named shapes — row, column, square, diagonal, scalar, identity, zero

What to assume they know

• The order of a matrix, rows before columns, from the previous topic
• The double subscript a sub i j, and which index selects which line
• The diagonal of a square grid, read from top left to bottom right
• The idea of a condition on entries stated in terms of the indices — the case i equal to j against the case i unequal to j
• Surds and decimals as legitimate entries
• What it means for one class of objects to sit inside another

What they should be able to do

• Recognise each of the seven named kinds from a printed array
• State each definition as a condition on the order, on the entries, or on both
• Say which of the seven require the matrix to be square and which do not
• Read the general forms the chapter gives for a column matrix, a row matrix and a square matrix
• Identify the diagonal of a square matrix and list its entries
• Express the diagonal, scalar and identity conditions in index form, using the split between i equal to j and i unequal to j
• Order the four square kinds by containment, and give an instance separating each consecutive pair
• Use the symbol for an identity matrix of stated order, and explain when the order may be dropped
• Explain why the zero matrix needs no order restriction and why the symbol for it carries none
• Decide small counting questions about matrices of a fixed order with entries drawn from a fixed set

Where it usually goes wrong

• "A row matrix is a column matrix written sideways." They are matrices of different orders, and this chapter has no operation that turns one into the other until the transpose arrives in the third module. Until then they are simply two different kinds.
• "Diagonal means the entries on the diagonal are non-zero." It means the entries off the diagonal are zero. The chapter's own second instance has a diagonal entry equal to two and another equal to minus one; nothing forbids a zero on the diagonal, and a zero square matrix is diagonal.
• "Scalar and identity are the same thing." The identity is the one scalar matrix whose repeated value is one. Every other scalar matrix — including the chapter's own instance with minus one repeated — is not an identity.
• "Every diagonal matrix is scalar." Only when the diagonal entries all agree. The chapter's three by three diagonal instance has three different values on its diagonal and is the separator between the two names.
• "The zero matrix has to be square." It does not. Two of the four instances printed on Part I p. 41 are not square, and this is the cleanest evidence in the section that the seven names are not one family.
• "The symbol for the identity always needs its order written on it." The chapter attaches the order and then says the subscript may be dropped whenever the context fixes it — which, from §3.4 onwards, it usually does.
• "A one by one matrix is a strange edge case." It is the instance the chapter uses first for three of the four square kinds, precisely because every condition holds on it trivially. Use it to show that the definitions are conditions, not pictures.
• "There must be a name for a matrix with zeros above the diagonal." Not in this chapter. Neither triangular matrix nor any equivalent is named anywhere in Part I pp. 34–75 — checked against the extracted text of all forty-two pages and against the page image of every one of them. Do not introduce the term.

Questions to check understanding

• Classify a printed matrix under every name it satisfies, not just one
• Write down the general form of a column matrix and of a row matrix
• List the diagonal entries of a stated square matrix
• Write the diagonal, scalar and identity conditions in index form
• Produce a matrix that is diagonal and not scalar, and one that is scalar and not an identity
• Decide whether a stated matrix is square, given only its order — the form of Exercise 3.1 Q8
• Count the matrices of a stated order whose entries come from a stated small set — the form of Exercise 3.1 Q10
• Explain why a named kind does or does not require the matrix to be square

Examples worth working on the board

Values marked verified are worked out here from the chapter's own printed data; neither answers file was opened, and this chapter prints no answers to its exercises.

• The seven definitions in printed order (§3.3, Part I pp. 39–41). Column, row, square, diagonal, scalar, identity, zero — numbered (i) to (vii) with roman numerals, each with a one-sentence definition, at least one worked instance, and for the first three a general form written with a bracketed subscript. It is nearly the containment order already, and the two odd ones out sit at the front and the back.
• The column and row instances (Part I p. 39). Read off the page image: the column instance has four entries — zero, root three, minus one, one half — and the chapter labels it as being of order four by one. The row instance has four entries — minus one half, root five, two, three — and its order is printed as a subscript on the closing bracket rather than stated in the sentence. Both general forms are given, with the fixed count written as one.
• The square instance (Part I p. 39). Three by three, holding three, minus one and zero; three halves, three root two and one; four, three and minus one. The chapter reports its order as the bare number three, which is the first place one number is used where a pair had been used before. Say that aloud: once the two counts agree, one of them suffices.
• The Note that names the diagonal (Part I p. 39). It defines the diagonal as the entries whose two indices agree, and demonstrates on a three by three matrix holding one, minus three and one; two, four and minus one; three, five and six. Verified: its diagonal entries are one, four and six, which is what the Note reports.
• The diagonal instances (Part I p. 40). Three of them, of orders one, two and three: a single entry four; then minus one and two on the diagonal with zeros off it; then minus one point one, two and three. The order-one instance is the one to dwell on — every one by one matrix satisfies the diagonal condition vacuously, because it has no off-diagonal position at all.
• The scalar instances (Part I p. 40). Again three, of orders one, two and three: a single entry three; then minus one repeated; then root three repeated. Verified: the repeated value is unconstrained — it may be negative, and it may be irrational, and the chapter's own three instances exercise both.
• The identity instances and the containment sentence (Part I p. 40). The identity is defined by the index split, one when the indices agree and zero when they do not, and instanced at orders one, two and three. The section then closes with the one containment remark the chapter makes: a scalar matrix becomes an identity exactly when its repeated value is one, and every identity is a scalar matrix. That sentence is the whole of section 10 — the chapter gives the reader one link of a four-link chain and stops.
• The zero instances (Part I p. 41). Four of them, and they are not all square: a single zero, a two by two, a two by three, and a one by two. The chapter leaves the order to be read off whatever surrounds it, and gives the symbol no subscript at all. Verified against the printed list: two of the four are not square, which is the evidence that this name is cut on a different axis from the four above it.
• Exercise 3.1 Q8 (Part I p. 43). Four options for the condition making a matrix square. Verified: the two counts must agree. The item is trivial and worth including anyway, because it is the only place the exercise set asks a student to state a definition rather than use one.
• Exercise 3.1 Q10 (Part I p. 43). How many three by three matrices are there with every entry either zero or one? Verified: nine positions, two choices at each, independently, so two to the ninth, which is five hundred and twelve. The distractor twenty-seven is three cubed and the distractor eighty-one is three to the fourth; both come from confusing the order with the number of positions. This is the best item in the exercise for making the count mn concrete.
• The Summary's compressed restatement (Part I p. 73). Seven of the bullets restate these definitions in index form, tighter than the section does. Two are worth putting side by side with the section text: the scalar bullet and the identity bullet, which differ only in that the identity bullet fixes the repeated value at one. The Summary does not restate the containment sentence — checked on the page images of Part I pp. 73 and 74 — so a student revising only from it loses the chain entirely.

Figures to have open

• A three by three grid with the diagonal cells independently lightable, for sections 5, 6 and 7. Build it with the repo's DataTable component and reuse the same grid across all three sections so the conditions visibly stack.
• Four nested regions for section 10, innermost to outermost: identity, scalar, diagonal, square, each ring carrying one separating instance. The repo's Shells component carries this directly. The nesting picture is added here — the chapter draws nothing and states only the innermost link, in words, on Part I p. 40.
• A side-by-side of the row instance and the column instance from Part I p. 39, drawn to the same entry size so the difference read is order and not scale.
• Four zero matrices of orders one by one, two by two, two by three and one by two for section 9, exactly as printed on Part I p. 41.
• No redraw of a textbook figure is possible or needed: this chapter prints no figure at all. Confirmed on the page image of every one of the forty-two pages.

Where this sits in the book

• NCERT Class 12 Mathematics, Chapter 3 "Matrices", §3.3, items (i) to (vii), Part I pp. 39–41
• The Note naming the diagonal, following item (iii), Part I p. 39
• The containment sentence closing item (vi), Part I p. 40
• Exercise 3.1, questions 8 and 10, Part I p. 43
• Summary, the seven definition bullets, Part I p. 73