Imagine if you will (and really, you must, for I have no photos this time), that you are looking out of an Olympia window in 1850. The view may be across the bay to a ship dropping anchor a under a magnificent sky, or it may be fog clinging or rain striking the pane just a couple of millimeters away. Either way, you would be one of the first people looking outside through a window in these parts.
More committed historians than I probably know which of the original hovels had a glass window, and in what year light first entered a building here through anything other than an open door or the space in the longhouse roof planks where smoke flowed out. Thanks to Mrs. George Blankenship, writing a pioneer history in 1914, I know that one early set of windows was in the school started by the Catholic Missionaries. Not the first windows in this neck of the woods, but probably the first ones through which daydreaming students gazed at the South Sound sky.
How would I know that they were looking at the sky? Because Mrs. Blankenship noted that the windows were high on the walls, supposedly to prevent the kids from becoming distracted. I went to school 3,000 miles away and more than a century later, but this same trick of scholastic architecture was in force: put the windows high enough that kids at desks can't see much.
How long before the miracle of never-melting ice or solidified air (or whatever the Salish mind made of the first pane) became the underwhelming reality of broken glass? My guess is: during construction. Not that that matters. Even if the school-makers and then the students managed never to break a single pane, eventually the school and all the other Mission structures fell and their windows broke (though probably not in that order, there being both boys and rocks in Olympia from early on).
If you are an archaeologist interested in the birth of a building more than its demise, those broken windows come in handy. Because it turns out that in some lucky parts of the globe, the typical thickness of a pane of glass changed systematically over time. In fact, one of the best known chronologies comes from Fort Vancouver (in WA, not BC), a Hudsons Bay Company outpost that got going before Olympia, and for a while the source of nearly all non-Native goods purchased through a secondary HBC outpost at Nisqually, almost certainly including any window glass purchased for the Mission.
What David and Jennifer Chance documented during their Vancouver study was that over time, glass thickness increased. A couple of years later, Karl Roenke made the case for a broader Pacific Northwest window glass chronology. These two schemes for turning precise measurements of pane thickness into a calendar agree for the most part, increasing confidence in this method, at least within the region.
Ideally, the archaeologist working with window glass (OK, the lab tech stuck making measurements for days) will have thousands of pieces. At Mission Spit, we got just 69 pieces. A small sample size is vulnerable to attenuation and probably other stat-talk weaknesses that I do not understand, but which can throw off results or make it hard to see a pattern.
In this case, though, the limited sample still sets the clock to right about where we would expect. It turns out that 2/3 (n=46) of the sample falls entirely within the date ranges established by the Chances and Roenke that correspond to the Mission occupation (AD 1848-1860).* If you include the chronologies' AD 1855-1885 period, which overlaps with but is not entirely within the Mission period, then over 85% of the window glass is in the expected age range. The mean, median, and modal thickness of the Mission Spit collection all fall in the 1840-1860 range of the chronologies.Three pieces are too thin to appear on the chronologies, and 4 are too thick; these may be flat pieces of bottles, rather than window glass.
So what? We already knew that the Mission was active from 1848-1860, and the glass merely agrees. But stand on the other side of the pane and take a look. First of all, there's the distribution. The chart above gives counts per period for the Chance & Chance chronology. Most of the glass is at the early end of the spread, as would be expected of a new Mission. But there is also a dropoff at the late end of the scale as well--there's nothing to suggest that occupation continued after the Missionaries left, and after their caretaker gave up interest a couple of years later.
The archaeological implication is that the salvaged Mission Spit deposit, scooped up with heavy machinery and virtually no vertical control over excavation, does not appear to be mixed with post-Mission cultural materials. Along with a few other observations, the window glass chronology for the site helps increase our confidence that we are dealing with a Mission period deposit, and not generations of subsequent settlement.
* Strictly speaking, some of the glass appears to be older than the Mission, but none of the tribal oral history or other written histories indicate that a building with windows would have been present anywhere nearby prior to the Mission. Maybe HBC was still selling off inventory from a shipment a few years prior?
More committed historians than I probably know which of the original hovels had a glass window, and in what year light first entered a building here through anything other than an open door or the space in the longhouse roof planks where smoke flowed out. Thanks to Mrs. George Blankenship, writing a pioneer history in 1914, I know that one early set of windows was in the school started by the Catholic Missionaries. Not the first windows in this neck of the woods, but probably the first ones through which daydreaming students gazed at the South Sound sky.
How would I know that they were looking at the sky? Because Mrs. Blankenship noted that the windows were high on the walls, supposedly to prevent the kids from becoming distracted. I went to school 3,000 miles away and more than a century later, but this same trick of scholastic architecture was in force: put the windows high enough that kids at desks can't see much.
How long before the miracle of never-melting ice or solidified air (or whatever the Salish mind made of the first pane) became the underwhelming reality of broken glass? My guess is: during construction. Not that that matters. Even if the school-makers and then the students managed never to break a single pane, eventually the school and all the other Mission structures fell and their windows broke (though probably not in that order, there being both boys and rocks in Olympia from early on).
If you are an archaeologist interested in the birth of a building more than its demise, those broken windows come in handy. Because it turns out that in some lucky parts of the globe, the typical thickness of a pane of glass changed systematically over time. In fact, one of the best known chronologies comes from Fort Vancouver (in WA, not BC), a Hudsons Bay Company outpost that got going before Olympia, and for a while the source of nearly all non-Native goods purchased through a secondary HBC outpost at Nisqually, almost certainly including any window glass purchased for the Mission.
What David and Jennifer Chance documented during their Vancouver study was that over time, glass thickness increased. A couple of years later, Karl Roenke made the case for a broader Pacific Northwest window glass chronology. These two schemes for turning precise measurements of pane thickness into a calendar agree for the most part, increasing confidence in this method, at least within the region.
Ideally, the archaeologist working with window glass (OK, the lab tech stuck making measurements for days) will have thousands of pieces. At Mission Spit, we got just 69 pieces. A small sample size is vulnerable to attenuation and probably other stat-talk weaknesses that I do not understand, but which can throw off results or make it hard to see a pattern.
In this case, though, the limited sample still sets the clock to right about where we would expect. It turns out that 2/3 (n=46) of the sample falls entirely within the date ranges established by the Chances and Roenke that correspond to the Mission occupation (AD 1848-1860).* If you include the chronologies' AD 1855-1885 period, which overlaps with but is not entirely within the Mission period, then over 85% of the window glass is in the expected age range. The mean, median, and modal thickness of the Mission Spit collection all fall in the 1840-1860 range of the chronologies.Three pieces are too thin to appear on the chronologies, and 4 are too thick; these may be flat pieces of bottles, rather than window glass.
So what? We already knew that the Mission was active from 1848-1860, and the glass merely agrees. But stand on the other side of the pane and take a look. First of all, there's the distribution. The chart above gives counts per period for the Chance & Chance chronology. Most of the glass is at the early end of the spread, as would be expected of a new Mission. But there is also a dropoff at the late end of the scale as well--there's nothing to suggest that occupation continued after the Missionaries left, and after their caretaker gave up interest a couple of years later.
The archaeological implication is that the salvaged Mission Spit deposit, scooped up with heavy machinery and virtually no vertical control over excavation, does not appear to be mixed with post-Mission cultural materials. Along with a few other observations, the window glass chronology for the site helps increase our confidence that we are dealing with a Mission period deposit, and not generations of subsequent settlement.
* Strictly speaking, some of the glass appears to be older than the Mission, but none of the tribal oral history or other written histories indicate that a building with windows would have been present anywhere nearby prior to the Mission. Maybe HBC was still selling off inventory from a shipment a few years prior?
