History of the Sash Window: 1670s London to the Window Tax

The window you are looking at is a piece of counterweighted machinery designed in the 1670s, and in its essentials it has not been improved on since. Cord, pulley, weight, box. The reason it still works after two or three centuries is that nothing in it wears out in a way a joiner cannot renew, and the reason it looks the way it does has almost nothing to do with taste. Fire regulations, the physics of hand-blown glass, and a tax collector's counting method between them dictated nearly every proportion in a London elevation.

Knowing which forces shaped your particular window is not antiquarianism. It is how you tell what is original, what is a later alteration, and what a repair has to match.

Where it came from, and the argument about who invented it

The oldest surviving sash windows were installed in England in the 1670s. Ham House in Richmond and Palace House both retain examples from that decade, and they are recognisably the same device as the window in a Victorian terrace: two glazed frames sliding vertically past one another in a cased surround.

Who invented it is genuinely unsettled. The credit is sometimes given to Robert Hooke, the surveyor and natural philosopher who worked on rebuilding London after the Great Fire, but there is no conclusive evidence. Others read it as a Dutch import, arriving with Dutch craftsmen and Dutch fashions in the decades around the Restoration. The scholar H.J. Louw, who looked at this more carefully than anyone, concluded that the exact inventor cannot be determined.

What is not in doubt is what happened next. Within thirty years the sash had displaced the leaded casement in fashionable English building, and by the middle of the eighteenth century it was the default window for anything above cottage level. The early examples were heavy: thick glazing bars, small panes, and frames sitting almost flush with the face of the wall. Every one of those things changed, and each change has a reason behind it.

Fire law, not fashion, pushed the boxes into the wall

If you compare a 1690s window with an 1820s one, the most obvious difference is depth. The early frame stands proud, its timber box visible from the street. The later one is buried behind the brickwork with only a reveal and a slim staff bead showing. That was legislation, and it was about fire.

Two Acts did most of the work.

The Act of 1707 required wooden roofs to be surrounded by a stone parapet, which is why the projecting timber eaves cornice disappears from London terraces around that date and a brick parapet with a hidden gutter appears instead.

The Act that followed in 1709 required window frames to stop sitting flush with the wall and to be set back from its face instead. A timber box standing on the face of a building is a ladder for flame. Recess it and the brick shields it.

The Building Act of 1774 finished the job. It sorted London houses into four rates by ground area and value, each with its own rules for foundations and wall thicknesses, and it required all external window joinery to be hidden behind the outer skin of masonry as a precaution against fire. It also banned most exterior timber ornament apart from door frames and shopfronts, and it created a statutory surveyor to inspect new building and enforce compliance.

That is why a London box sash is set deep in its reveal with only a narrow strip of the frame showing, while sashes in many other parts of the country sit closer to the face. It is also why fitting a modern replacement window into a Georgian opening so often looks wrong: the frame is the wrong depth in the wall, and the eye reads the shadow line before it reads anything else.

The window tax, and the myth attached to it

The window tax was introduced in England and Wales in 1696 under William III. It was a proxy for wealth: counting windows was something an assessor could do from the street, without entering the house, at a time when any direct enquiry into income was fiercely resented.

The structure moved around a great deal over its 155 years.

The threshold changes are the interesting part. Dropping the count to seven windows in 1766 brought ordinary terraced housing into scope for the first time, and that is the period from which most genuine tax-driven blocking dates.

It is worth being careful with the folklore, though. Blocked windows are not automatically evidence of tax avoidance. Architects blocked openings for symmetry long before 1696 and long after 1851, because a blind window in the right place completes an elevation where a real one would have opened into a chimney breast or a stair. The "daylight robbery" derivation is not supported by scholarship. When you find a bricked opening, look at the brickwork: tax-driven blocking is usually crude infill within the original reveal, while a designed blind window has proper arch, cill and often a painted or rendered panel imitating glass.

Glass decided the pattern

The single biggest constraint on a Georgian window was not the joiner. It was the glassmaker.

Crown glass was made by blowing a globe, transferring it to a punty, then reheating and spinning it until centrifugal force flattened it into a disc, which could reach five or six feet across. The disc was not even. It was thinnest and clearest at the rim and grew thicker and cloudier towards the centre, where the pontil left the familiar bullseye. The good glass was the outer band. The bullseye went into cheap work, outbuildings and back windows, which is the opposite of the modern habit of putting bullseyes in front doors for effect.

Cut usable rectangles out of an annulus of good glass and you get modest panes. That is the whole explanation for the Georgian pattern. Six panes over six, three across by two up in each sash, is not an aesthetic decision that happened to be practical. It is the practical limit dressed up beautifully, and the reason the proportions look so settled is that generations of joiners refined the same constraint. Our piece on Georgian six-over-six proportions goes into how those ratios were actually set out.

Glazing bars did the structural work of holding all those small panes, and they thinned steadily as joinery improved. Early eighteenth-century bars are broad and flat. By the Regency they are slender, often 18mm to 22mm across the face, with fine ovolo or lamb's tongue mouldings that catch a shadow. The development of glazing bar patterns is the most reliable dating tool on any elevation, more so than the brickwork.

Where original glass survives it is worth keeping. It has visible draw, faint ripples and a slight surface undulation that no modern float glass reproduces, and it is the reason a period elevation glitters unevenly in low sun. Matching it when panes are broken is a specialist business, covered in heritage glass for sash windows.

Plate glass and the Victorian reinvention

Through the nineteenth century, rolled plate, drawn sheet and improved cylinder processes replaced the spun disc. Suddenly a single pane could be large, and the constraint that had produced the six-over-six simply dissolved.

The Victorian response was immediate and total. Glazing bars were dropped wholesale. Windows went to two panes over two, then one over one. Openings grew taller. And the joinery had to change to cope, because the mechanical consequences were serious.

A big sheet of glass is heavy. Ordinary 4mm glass weighs around 10kg per square metre and 6mm around 15kg, so a large Victorian one-over-one sash can easily be two or three times the weight of the Georgian sash it replaced. Three things follow:

• The counterweights got much heavier, and lead started appearing where cast iron would not fit in the box.
• The bottom rails got deeper, to carry the load and resist the sash racking.
• Horns appeared. The meeting rail joint on a barless sash is the weak point, and extending the stiles down past the meeting rail as projecting horns puts timber where the joint needs it. They are a structural answer that turned into a style, and because their shape changed steadily they are the sharpest dating detail on a Victorian house. We set out how to read them in sash horns and what they tell you.

Anyone who has taken apart a mid-Victorian window recognises the shift straight away. The frames are more heavily worked, the boxes deeper, the pulleys larger, and the whole thing is built to carry weight rather than to divide it up. How to identify and date Victorian sashes covers what else changed alongside.

The Edwardian coda, and the long interruption

By the turn of the twentieth century the pendulum swung part way back. The Edwardians kept large panes but reintroduced decoration selectively: a border of small margin lights around a big central pane, coloured or textured glass in hall windows, wider and squatter sash proportions to suit the broader suburban frontage. It is a distinctive look and easy to confuse with Victorian work until you notice the width. Edwardian window details sets out the tells.

Then it stopped. Between the wars the steel casement took hold, and after 1945 the sash window nearly vanished from new building. When it came back commercially in the later twentieth century it came back without the box: spiral balances in a solid frame, and later uPVC vertical sliders. Those work, after a fashion, but they are consumable in a way the original never was. A cord and a cast-iron weight can be renewed forever by anyone with a chisel. A proprietary spring balance in a discontinued profile cannot.

What the history means when yours needs work

All of this has direct consequences on site.

Nothing is standard. There is no such thing as a stock sash window. Every one was made for its opening by a joiner working to local practice, and the bars, rails, cills and mouldings vary house by house and sometimes window by window on the same house. Anything replaced has to be measured and matched, not ordered.

The timber is better than anything you can buy easily. Georgian and Victorian sashes were commonly made from slow-grown Baltic redwood with very tight annual rings and high resin content, cut so that the heartwood faced the weather. Modern fast-grown softwood is a different material. This is the single strongest argument for splicing a decayed section rather than making a new sash: you are keeping timber you cannot readily replace.

Layers tell the story. Paint sections, glass types, cord sizes, and whether the horns are worked from the solid or planted on will usually tell you what has been done before. That evidence is worth reading before anyone starts stripping.

Repair is not conservative nostalgia, it is the cheaper engineering answer. A window built to be dismantled, that has already survived two centuries, sitting in an opening it was cut to fit, is a better starting point than any replacement. Our sash window restoration work is built on that, and it is why so much of what we do is splicing, re-cording and re-puttying rather than making new.

What to do next

Go and look at your own windows with the chronology in mind. Count the panes. Look at the width of the bars. Check whether the box is buried in the reveal or standing proud. See whether the horns are cut from the solid or glued on, and whether the glass has any draw in it when you move your head.

Ten minutes of that will place your windows to within a couple of decades, and it will usually flag which ones are original and which are later replacements pretending. That distinction changes what is worth spending money on. If you want a second opinion, our sash window repair service covers all the London boroughs and a survey starts with exactly this: working out what you have got before deciding what it needs.