The first thing my teacher ever said to me about type was not about type at all. It was a Tuesday morning in October, and I had arrived at the compositing room early, eager to show that I was serious. He was standing at the stone, looking at nothing in particular. Or so I thought.

He said: you see letters. You're not looking at the spaces yet. Come back when you can see the spaces.

It took me months to understand what he meant. Then one morning I did, and everything changed. The spaces between letters are not empty. They are shaped. They have weight. They push and pull against each other in ways that are invisible when you are looking at the letters themselves, and completely visible when you stop looking at the letters and look at what is between them.

The garden teaches this lesson too, though it takes longer. A bed that has been planted too densely looks lush in early spring and exhausted by midsummer, because the plants have been competing for what is between them — air, light, water — rather than growing into the space they were given. My teacher used to say that the best gardens are the ones where every plant has room to be what it is. He said the same about type.

The space between letters

In typography, the space between letters is called the sidebearing — the invisible zone on either side of each character within which no adjacent character should intrude. The sidebearing is set by the type designer, not by the compositor. It is part of the character's definition, not a space the compositor adds. What the compositor controls is something different: the tracking, the overall loosening or tightening of the space across an entire run of text, and the kerning, the fine adjustment of spacing between specific pairs of letters whose shapes create awkward gaps when placed side by side.

Most body text should not be tracked. The type designer has already found the optimal spacing. Adding tracking to body text — a habit I had to unlearn — makes the text look airy and open in isolation, but reduces the number of characters per line, disrupts natural spacing relationships, and makes no perceptible improvement to legibility at ordinary reading distances. The correct tracking for body text is zero.

For text set entirely in capital letters — headings set in small caps, chapter labels, running headers — tracking is not optional. It is required. Capitals were designed to sit next to lowercase letters, and when set in sequences of all-capitals they crowd together in ways that look crude. Adding tracking to all-caps text, between 0.05em and 0.15em depending on the typeface and size, restores the visual spacing that the mixed-case context would have provided naturally.

The space between lines

Line-height — the distance from the baseline of one line of text to the baseline of the next — is the most consequential spacing decision in typography. Set it too tight and the text feels airless, the eye losing its place on the return from the end of one line to the beginning of the next. Set it too loose and the lines float apart, each one an island in a sea of white.

For body text in a book of this kind — a literary essay, set in a humanist serif at eleven points — the correct line-height is somewhere between 1.5 and 1.8. The value used throughout this book is 1.76. This is not a number I arrived at by calculation. It is a number I arrived at by looking. The page looks right at 1.76. At 1.6 it looks slightly compressed. At 1.9 it looks slightly airy. The difference between 1.76 and 1.8 is imperceptible to any reader. The difference between 1.6 and 1.9 is not.

My teacher would have said that the correct line-height is the one that makes the page look like a gray rectangle when you squint at it. Not a striped rectangle, where the lines are too far apart and each line reads as a visual element in its own right. Not a dense dark rectangle, where the lines are too close and the page looks heavy and airless. A gray rectangle: even, even, even, from top to bottom, with nothing to pull the eye from the reading.

The space around the text

The margin is not leftover. This is the first thing I understood about pages, and the last thing I understood about gardens. In a garden, the space between plants is ground — it is what the plant grows into, what gives it definition, what allows it to be seen as a plant rather than as part of an undifferentiated mass. In a book, the white around the text is ground for exactly the same reason. The text exists as figure against a ground of white, and the proportion and quality of that ground determines whether the figure reads as text or as a mass of marks on a surface.

My teacher's pages were always generously margined. I once asked him why he wasted so much paper. He said he wasn't wasting paper. He was growing the text room to breathe. I didn't understand what he meant for a long time. Now I think it is the most important thing he ever told me.

The garden teaches patience in the way that nothing else quite does. You prepare the ground, you plant, you wait. The outcome is never certain. The seed may not germinate. The seedling may not survive. The mature plant may produce nothing or everything, depending on variables — soil, weather, the precise angle of the October sun — that you cannot control and can barely influence. You do what you can do and then you wait to see what the garden does.

My teacher said something similar about type. You set the measure, you choose the face, you adjust the spacing until it looks right. Then you read it. What comes back to you is not always what you intended. Words that seemed clear at the stone look strange on the page. Spacing that seemed generous looks tight in print. The page has its own opinion about what belongs on it, and that opinion is often different from yours.

Both disciplines ask you to relinquish control at the critical moment, and to trust the process you have set in motion. This is not passivity. The gardener who prepares the ground well, who selects the right varieties for the conditions, who plants at the right time and depth, has done everything that can be done. What grows from that preparation is the garden's business, not the gardener's. The same is true of the compositor who has made every correct decision — the typeface, the measure, the spacing, the margins — and then committed the work to the press. What comes off the press is the type's business, not the compositor's.

The best work I ever did was work I had forgotten I was doing. The work I was most conscious of — the work I laboured over, revised, second-guessed — was rarely my best. The unconscious craftsman is more reliable than the anxious one.

— from a letter to the author, undated

On revision

There is a kind of patience that is merely waiting, and a kind that is active. The gardener who prepares the ground in autumn is not waiting; she is practising a form of engagement with the future that requires her to hold the image of the finished bed in her mind while working on ground that looks, in November, like nothing more than dirt. The compositor who spaces a paragraph and then leaves it overnight before reading it again is not waiting; he is giving his eye the distance it needs to see the paragraph as a reader would see it, rather than as its maker.

Revision is a form of patience. The willingness to look again at what you have made, to see it as if for the first time, and to change what needs changing — this is a skill, not a personality trait. It can be learned. My teacher practised it by the simple method of never reading anything he had set on the day he set it. He put the galley away, went home, grew things, and came back the next morning with fresh eyes. By his account, he almost always found something to correct.

I have tried to practise the same discipline with these essays. Whether I have succeeded is not for me to judge. What I can say is that I have looked at each of them many times, in different states of mind and at different distances of time, and that what I see now is quite different from what I saw when I first wrote them. The essays have changed, or my eyes have changed, or both. I am not sure the distinction matters.

What follows is not an essay but a list — the kind of list that a gardener keeps out of necessity and a compositor keeps out of habit. I have been maintaining this calendar for eleven years. It is not comprehensive. It is a record of what I have grown and what I have noticed, arranged in the order that the year itself arranges things. Some entries have typographic notes. Most do not. The typographic notes are not the point; they are what happens when you spend enough time in two disciplines that they begin to talk to each other.

My teacher talked very little. This is not a complaint — I learned more from watching him than from anything he said, and what he said was usually more useful for having been withheld until exactly the moment it was needed. He was not a man who explained himself. He demonstrated, and waited for you to see what he had demonstrated, and then explained it if you hadn't.

What follows is a partial record of what he said, transcribed from memory and from the notebooks I kept during my apprenticeship. I have arranged the entries in no particular order. He did not offer his observations in any particular order. They arrived when they arrived, usually in the middle of something else, and departed without ceremony.

He died in the autumn of the year before last, at the age of ninety-one, in the garden of the house where he had lived since he was forty. The garden was, by any measure, extraordinary. He had been planting it for fifty years and it showed. Everything in it had room to be what it was.

I think of him when I set type. I think of him when I plant things. I think of him, above all, when I am trying to decide whether to kern a particular pair of letters or leave them as they are — which is to say, when I am trying to decide whether the space between them is a problem that needs repairing or a space that is simply what it is. He would have known immediately. I have to think about it every time.

The word leading — pronounced to rhyme with bedding — derives from the strips of lead that hand compositors placed between lines of movable type to increase the vertical space between them. A line of type cast in metal had no built-in space above or below it beyond what the type itself occupied. To add space, you added lead. To increase the space, you added more lead. To reduce it, you used less, or none.

The term survived the transition from metal to phototypesetting, from phototypesetting to desktop publishing, and from desktop publishing to the web. We still talk about leading as though we were placing strips of metal between rows of characters, even when the characters exist only as mathematical descriptions rendered by a processor to a screen. The ghost of the lead strip persists in the vocabulary, which is one of the ways that the history of a technology outlives the technology itself.

What the lead strip did

The lead strip did something that is easy to overlook: it separated. Not by much — a point or two, or three — but enough to give the eye a clear path from the end of one line to the beginning of the next. Without it, text set solid (the technical term for text set with no additional leading) can be read, but the reading requires effort. The eye must find its own way back, tracking across the white space at the right side of the column and then down and across to the left edge, without any assistance from the spatial structure of the text itself.

Good leading is assistance. It is spatial scaffolding for the eye's return journey. The best leading is invisible — it does its work so efficiently that the reader is unaware of it, experiencing only the ease of movement from line to line, rather than the mechanism that makes that movement easy. The worst leading, in either direction, calls attention to itself: too tight, and the reader feels the effort; too loose, and the reader notices the gaps.

The garden has its equivalent. The path through a well-planted bed is assistance — it guides the eye across the composition without commanding it, creating a route through the planting that feels natural rather than imposed. The bed without a path requires the viewer to find their own way, which is exhausting and ultimately frustrating. The bed with a too-obvious path feels rigid and over-managed, the design apparent and therefore diminished.

The compositing stone is a flat slab of iron — traditionally, it was actually stone, but iron was more precise — on which the compositor assembles the type. It is where the work is done: the setting of the measure, the locking of the chase, the correction of errors, the final verification before the page goes to the press. The stone is a workspace. The page is where the work ends up.

There is an enormous distance between these two surfaces. Not a physical distance — in a working print shop, the stone and the press are a few metres apart. The distance is cognitive. At the stone, the compositor sees the type backwards and upside down, as it must be arranged for the press to print it correctly. He cannot read what he has set; he can only check that it is correctly assembled. He is thinking about structure, not content. The page is where the content reappears, in legible form, for the first time.

My teacher told me that the most important skill a compositor can develop is the ability to move between these two perspectives — the structural and the legible — without losing either. The compositor who can only see structure makes technically correct pages that no one wants to read. The compositor who can only see content makes readable pages that fall apart under scrutiny. The best compositors hold both views simultaneously, which is a form of double vision that takes years to acquire and is never quite complete.

The garden as middle distance

The garden has taught me more about this double vision than the compositing room has. A garden must be designed and planted from the stone — from the perspective of the plan, the spacing, the structure, the arrangement — and then experienced from the page — from the perspective of the visitor moving through it, seeing what is there rather than what was intended. These two perspectives are never identical, and the gap between them is where most gardening failures live.

I have made a bed that looked, on paper, beautifully proportioned, and in person cramped. I have made a bed that looked, on paper, perhaps too sparse, and in person exactly right. The paper is the stone. The garden is the page. The distance between them is the same distance that separates intention from effect, and it is never smaller than you hope.

When I set type now, I try to remember this distance. I try to look at the page as though I am walking into a garden I have never seen before, with no memory of the stone where it was assembled. It is impossible, of course — I know too much about how it was made to see it purely as a reader would. But the attempt, made honestly, produces something closer to objectivity than the alternative, which is to see only the stone and never the page at all.

My teacher's garden was, in the end, the best argument he made for his typography. Everything in it had been placed with the attention he brought to a galley of type: deliberate, proportional, considered at the distance of the page rather than only at the distance of the stone. He spent fifty years looking at both simultaneously. The garden showed what that looked like.