Brunelleschi's Experiment

In 1425 Filippo Brunelleschi invented game engines.

What he had invented, of course, was much more fundamental, much more revolutionary: linear perspective.  Game engines, and along with them the invention of virtual reality and augmented reality, were mere by-products of his work and as such they went without comment.

According to his adoring biographer, Antonio Manetti, here's how it happened...

One day in Florence in 1425 Brunelleschi assembled the good folk of the city in the Cathedral square.  He seems to have had something of the reputation of a wizard and the gathered audience must have been pleasantly mystified with what they saw.  Manetti doesn't give the exact details, but somewhere in the square before them were two objects, possibly placed on stands or easels, and probably concealed under cloths.  Brunelleschi asked for a volunteer.  A man was put forward and Brunelleschi accordingly asked him to stand behind the nearest veiled object which was then unveiled.  It was underwhelming.  It was a small rectangular panel, perhaps a painting, but, as it was facing away from the audience, it was difficult to tell what of.  Brunelleschi asked the volunteer to go close to the panel where he would find a small hole drilled in it.  He was to look through this hole into the square.  All the gentleman would have seen was the veiled object as it stood directly in the line of his sight, limited by the hole through which he was peering.

Brunelleschi gazes at the Duomo

(Photo: Jason Pier - used with permission

from Flickr: here)

Then the second object was unveiled.

Brunelleschi asked the volunteer what he saw.

'Sir,' he said, 'I see the Baptistry.'

The audience gasped in astonishment.

What the man was looking at was a mirror.  But what he was looking through was a painting of the Baptistery.  To him, when the mirror was unveiled it was as if Brunelleschi had simply plucked the veil from where it had hung in thin air opening up his view of the actual Baptistery in the square beyond.  What he was actually seeing was a reflection of Brunelleschi's painting of the Baptistery which matched the man's perspective exactly.  This matching of perspective made it real from the observer's limited point-of-view.

As underwhelming as this may sound to us, what the man saw would do nothing less than revolutionise the very way people in the West saw their world.

Previously, people were accustomed to depictions of the world around them which were stylised in terms of their visual perspective.  It is likely that this mis-match with visual reality was not much of a cause for discussion of disappointment.  Most of the time it must have gone without notice.  It was convention, and as such, was invisible in the same way the the page or screen upon which text is written is invisible to the reader.  However, once Brunelleschi demonstrated his grasp of linear perspective all other representational art must have become ... different for the good people of Florence.  It must have become different in the way the design of the iPhone3GS became ... different ... when the iPhone4 was unveiled.  The world was changed and people did not look at what was familiar which quite such familiar eyes.

Within a generation most visual art in Italy would take into account linear perspective.  Within a century this would rule art across Europe (even in England!).  The world - or the way people expected to see it - had been changed forever.

Now, about the invention of game engines, virtual reality and augmented reality.  Brunelleschi provided a model for those in that he had created a system whereby the viewer was immersed in a representation of reality rather than the reality itself.  The effect only worked because it provided a convincing experience - which is at the heart of video games and their engines today.  The foundation for virtual reality was also laid because the rules of linear perspective eventually lead to the transform matrix which allows 2d vector graphics to mimic 3d perspective in realtime on display screens.  And finally, Brunelleschi's experiment overlaid his version of reality - a possible reality - atop his observer's view of reality itself.  It the 'magic window' effect of overlaying digital content on a mobile phone's camera stream.

It has taken a while for technology to catch up with the vision presented almost six centuries ago and there is still some controversy over what Brunelleschi actually did in that square on that day, how he did it, and what he actually meant by it.

However, to close this little circle of ideas, you will find below a demo of one way in which Brunelleschi may have set up his demonstration.  It runs in a game engine and it is entirely dependent on your experience and your own point-of-view.

How will our growing expectations about the dynamic nature of imagery produced today shape the way we see our world tomorrow?


Click the image to begin.  W,A,S,D to move, I to move mirror up, K to move mirror down, T to switch sketches

Explore the piazza (there is no dome on the Duomo because it didn't exist in 1425).  Find the panel with the drawing of the Baptistery then go round back and peer through the peephole.  Raise the mirror in front slowly to see the drawing overlaid on the 'reality' of the 'actual' Baptistery in front of you.  You can toggle between the kind of pseudo-perspective common at the time and linear perspective, just to appreciate the impact of this visual change.

Tomb Readers

I'm on the BBC!  Again!

Back at the end of October (around Halloween), my Archaeology office partner Helen Goodchild and I teamed up with battlefield archaeologist Tim Sutherland to try to record the almost invisible medieval inscription on a tomb at Saxton parish church.  This is supposed to be the final resting place of Lord Dacre, killed at the Battle of Towton in 1461 and buried in the nearby cemetery at Saxton.

The inscription has almost completely weathered away.  To bring it back, and to record what's left accurately, we used an RTI technique and the YorkToolkit, which I developed to metrically record details revealed from RTI.

Anyway, it's been quite a lot of fun and has proved useful to Tim's groundbreaking and important research on the Battle of Towton.

So, watch it!  It's on InsideOut Yorkshire & Humber on Monday, 6th February, 7:30pm BBC One.  And on iPlayer soon after.

Simple Power

The great power of video games lies in the way they directly engage the player.  Personal experience is key to the success or failure of a game.  It is also at the heart of the promise they hold as a powerful medium of artistic expression.

The other night, while glancing through Gamasutra, the game developer news site, an interview with the indie developer, Jake Elliott, caught my eye.  His recent game, Ruins, is a meditation on the blending of simple music, simple imagery and simple game play.  I decided to give it a try (especially nice because he's giving it away).  I am very glad I did.

Ruins - © Jake Elliott (used with permission - thanks!)

Ruins is much more than the sum of its parts.  While simple, its simplicity, its stylised presentation, and music that hovers just on the edge of familiarity, draws you in.  There is a story, but you the player are left to piece the fragments together as you find them - or, more accurately, catch them.  The truly remarkable thing about this polished little gem is that it is a framed experience.  It is an experience within an experience.  The character you move about the dark, mysterious landscape, is a dog.  According to game convention, the dog represents you.  It is your avatar in this lonely little world.  But the narrative seems to come from outside that world, from our own world, and told not to you, but by you.  You are both within the game and without simultaneously.

And the experience hangs together on a single strand of music by Chopin - quiet, coming and going, providing the only colour in this shadowy world.

While the experience is brief, it is delicious, sad but welcoming, like a fond dream remembered only later.  Such a strong, intimate connection with the player, the viewer, is a significant demonstration of the potential for the medium of games to have a unique, transformative impact.  The viewer and the creator engage in a relationship where meaning is built together, where both are invested in its creation and exploration.

And in Ruins the ultimate connection with the player is that this dream invites you to sleep and to dream for yourself.

Crossing Conundrum

So, I've spent the day drawing the plan of the church of Holy Trinity Priory, York as it may have been c1450 based on the remaining evidence.

And, I've run into a problem but also an excellent opportunity to demonstrate how the act of reconstruction can actually drive forward research on historic sites, or at least uncover major gaps in our current understanding.  This happens routinely, and it's convenient that it has happened here.

So, the problem:  the crossing.  Nothing joins up.

Help!

The church was cruciform in plan with four arms or wings extending east (the chancel), west (the nave), north and south (the transepts).  While only the core of the nave survives today, enough evidence remains to be able to sketch the outline of the medieval church.  At the turn of the last century a Holy Trinity parishioner and antiquarian, Walter Harvey Brook, undertook a series of excavations on the site in advance of restoration works.  The excavations were remarkably systematic for their time and Brook took copious, detailed notes and photographs, as well as produced three notebooks full of drawings.  Brook's excavations revealed the plan of the medieval church in much more detail than is visible today.  His work is the bedrock upon which later studies of the site are based.

Two later studies in particular are very important: the Royal Commisison report in 1972 and David Stocker's article on the priory in the BAA volume on Yorkshire Monasticism in 1995.  Of the two, Stocker is the most reliable regarding the building.

The result of all of this careful work is a series of reconstruction plans of the medieval church.  Using these plans, I should be able to reconstruct much of the interior arrangement of the church.

So you would think.

Brook's plan

The Royal Commission's plan

Stocker's plan

But, as is often the case, there are gaps in understanding which pose serious challenges for reconstruction but which have gone unnoticed during the composition of academic articles on the site.  Reconstruction forces the researcher to think systematically, considering practical issues of geometry and construction techniques.  When a reconstruction exists simply in the mind it is easy to miss major problems.  Add to this the fact that we must rely on work subject to all of the unrecorded little particulars of recording on site while Brook's excavations were conducted.  And we cannot directly investigate Brook's evidence (archaeology destroys its own evidence as a matter of course).  You begin to see that we can get into trouble easily.

In the plans all arms of the church join up neatly at the crossing, naturally.  But you can see that there are differences in these.  Brook's is a bit of a fantasy coloured by what he wanted to see more than what he actually saw, and this is acknowledged by Brook himself.  In the drawing he has crossed out many features.  It is very much a sketch of hypotheses rather than a finished site drawing.  Brook's more detailed notes and drawings are much more reliable.  The Royal Commission's plan carries with it all of the weight and authority of the Commission, communicated in its sure, clean lines which do not leave room for uncertainty (except in the dotted lines indicating unexcavated, assumed, features).  Stocker's plan differs from this significantly in that he does not include a western aisle to the north transept (which I believe) but also in many subtle features as well, particularly in the placing and extent of Brook's excavated features and the still standing line of the south wall of the chancel.

The real fight here is between the Commission and Stocker.  Here, I am very much in Stocker's corner.  The Commission shows much of the line of the walls of a western aisle of the transept as excavated 'fact' based on Brook.  But Stocker checked the original source and found that he did not excavation in this location because of the cemetery, which was still operational at that time.  They also appear to have misread Brook's notes in siting his excavated transept pier base in the west rather than in the east, as Stocker shows.  Also, the Commission plan only shows a small fragment of the northeastern pier of the crossing, tracing the eastern crossing piers' full extent as conjectural.  Stocker shows that both of these piers seem to have been fully excavated by Brook.  Both plans agree, at least, on the positions of piers in the chancel's southern arcade and the relative position of the south chancel aisle wall.  But they differ significantly in the extent of that wall.

So what, then?

As drawn by both Stocker and the Commission, the plan of the church reveals a building that could not have functioned structurally in the way everyone from Brook to me (until today) have assumed, i.e. that the chancel aisles were vaulted.  Vaulting must have existed somewhere on the site because a handful of vault rib voussoirs survive.  The first pier east of the crossing should line up with the north-south walls of the transepts.  If they do, then the entire length of the aisles can be divided into convenient rectangular bays which are simple to cover with a quadripartite ribbed vault.  But, in both plans this is not the case.  The lines of the walls end in the aisle mid-bay.  It is still possible to vault these bays, but the vaults would be very awkward - more awkward than the vaulting in the same bays in York Minster.  Not having a vault in this bay would leave the remaining three bays to the east oddly alone in their vaulting, incomplete.  If an aisle is vaulted this usually runs the length of the aisle.

In fact, you can see that the medieval builders encountered this problem and seem to have found a solution by extending a short stub of wall eastward from the eastern crossing piers, to push the position of the arcade respond into line with the wall of the transept.  But in the two plans, the alignment is still poor.

The problem is that neither plan is clear about their evidence, and the evidence they do have conflicts.  What is known is the position of the southern wall of the chancel as well as the line of its southern arcade, giving the width of the chancel aisle and its length.  What is not known is the width of the transept aisles.  In the Commission plan there does not appear to be any evidence for the width of the transept aisles as they show them.  In Stocker's plan the end of the southern transept's eastern wall has been assumed to be marked by the western end of the surviving chancel south wall.  This makes the awkward corner and the structurally problematic crossing and adjoining aisles.  But in the Commission plan the south wall of the chancel does not extend as far west.  In fact, it seems to end in line with the first pier east of the crossing.  This would place the eastern wall of the transepts in perfect alignment for vaulting but make very wide transept aisles with an unusually wide western arch in the chancel arcades.  But the Commission does not seem very sure about its placing of the eastern crossing piers.

So, who is right?  And how do I finish my drawing?  I cannot re-excavate, and I'm not sure Brook left anything for me to uncover even if I did.

I think I'll take my tape measure and knock on the rectory door, ask if I can measure their garden wall (which is the south wall of the chancel).

If you want something done right...

Holy Trinity Priory

At the top of Micklegate in York, just inside Micklegate Bar stood Holy Trinity Priory, a seven acre site housing the church, monastic and service buildings of a house of Benedictine monks. All that remains today is part of the nave of the priory church, saved from destruction because it served as the parish church for the neighborhood and continues in use as such to this day.

What stands today comprises barely a third of the original church and none of the monastery's other buildings remain. However, enough about the medieval priory is known to attempt visualizing what it may have looked like, which is what Christianity & Culture and Heritage Technology did in 2010. The resulting touchscreen installed in the church invites visitors to explore the buildings and grounds of the priory as they may have appeared around 1450.

Holy Trinity Priory, York, west elevation as it may have been c1450. (Copyright Heritage Technology Ltd. and Christianity and Culture, all rights reserved)

I drew the architectural plans (and wrote some text along with Louise Hampson at CnC) for our reconstruction of the site based on archaeological and art historical research (including a GPR survey I carried out to locate the cloister. Alas, like most of my geophysical work, the results showed very little!). Geoff Arnott at HTec turned my drawings into glorious 3d and Pat Gibbs at CnC stitched it all together into a lovely, intuitive interface. It was a tremendous amount of work (especially from Geoff) but well worth it.

Geoff Arnott's stunning modelling and rendering of the church and cloisters of Holy Trinity Priory, York c1450. (Copyright Heritage Technlogy Ltd. and Christianity & Culture, all rights reserved)

That was Phase I.

Now it's time for Phase II. This time, we're working on select building interiors and, ambitiously, Geoff has tackled character animation so we can present life at the priory on the Feast of the Holy Trinity c1450. Actually, we've been working on it for a little while and with my drawings well in hand.

However, even more exciting is Geoff's grand vision for these interior scenes.  For the first time, we are including character animation to populate our interiors.  Geoff has been working hard in this area for some time but he's really pushing the boat out for this phase.  It's going to be fantastic.

This series will update as we move forward with the work. Standby.