Marquette Mission Site - St. Ignace, Michigan

Interpretation

Architecture

Our analysis of site structure at Marquette Mission identified additional sections of the post-in-trench structures that had not previously been recognized. This type of architecture is a well-known type of French construction and the earliest buildings at Fort Michilimackinac include this type. On the compiled map above a large French structure that appears at the western extent of excavations and has a 77 foot long wall extending toward the east can be seen. At least one room can be identified at the southwestern edge. This French structure is too large to be the remains of an individual French house in the village or to belong to the mission. We argue that it is related to the military occupation of Michilimackinac. We also note that the French structure spatially overlaps with what we suggest is part of a Huron village.

The line of posts identified as the western wall of a Huron longhouse, extending approximately 110 feet, fit Cadillac’s description and are consistent with other archaeological examples of longhouses. The eastern wall of that structure has not yet been clearly defined. Archaeological patterns of Huron longhouses typically have hearths 10 to 12 feet apart. The limited excavations of the central area of the longhouse show four hearths each about 21 feet distant from the others. It is expected that, if excavations are completed, hearths will also be found in between those presently known.

An analysis comparing the mean width of postmolds between those found within the wall trenches and those found as single driven posts (such as those in the longhouse) demonstrates a statistically significant difference in the size of these two groups of posts. Those found within the trenches (i.e., French) are larger than the longhouse (i.e., Huron) and other single-post remains.

Based on the observed architectural patterns and the information gleaned from the historical record, we suggest that the area excavated at Marquette Mission holds the remains of both French and Huron occupations. Further, we argue that these occupations are not contemporary. The section of the Huron village recognized to date appears to have been abandoned and then re-occuppied by the French military. This explains why the French post-in-trench pattern is intrusive to the longhouse pattern. We can use the historical record to conjecture about when this may have happened and do so in the conclusion.

Artifacts and Features

In this analysis we examined 87 artifact types using descriptive statistics and also spatially using the GIS. At this point in our research, features and artifacts found within them are the primary data sets. There were several middens on the site that we treated as features and these often had an abundance of each artifact type. To arrive at a comparable intensity measure for the artifacts, we divided the number of artifacts per feature by the area of the feature or midden. Given our architectural analysis, we were particularly interested in spatial differences of distributions associated directly with the longhouse and those further to the west where the French trenches ran not only east-west, but north-south as well.

Artifact counts by category and feature were transferred from our Microsoft Excel database into SPSS. Artifact counts were examined for bimodal distributions, but none were found to be present. A One-Tailed Pearson’s Correlation Matrix was then done for all artifact categories. Additionally, certain artifact categories were lumped into more functional categories, and the correlation matrix was again performed (e.g., combining glass beads, bone beads, wooden beads and clay beads into one category). Based on the findings significant to 0.01, categories were selected for examination on the GIS, to look for spatial correlations. Many of these correlations were so abundant on the site that spatial patterns were not visible.

Native American Areas and French Areas

Using those classes that we assume to be more indicative of Native American presence, we examined the spatial distribution of these materials as single artifact classes and also looked at the distribution of the strongly correlated materials (see binder for examples). One interesting correlation was that of cores and chert projectile points. Features that contained both chert projectile points and cores were located only on the western portion of the site, the area associated with French trenches. Given our interpretation of this area as initially being Huron-related with the French structure of a relatively later time, we interpret this area as an outside activity area used by the Huron. Features classified as hearths with aboriginal ceramics also displayed an interesting spatial pattern and lend support to the longhouse interpretation. Several are found within the longhouse. Two hearths with aboriginal ceramics were found in the southern portion of the site, but Fitting found that they were below the French structure in this area. He argued that the hearths were Seventeenth century, but were used prior to the French building which was located there.

Bear mandibles are considered a uniquely Huron item at the Marquette Mission site. In 1973, Fitting found two bear mandibles in his northern excavation area (called Test Pit 3), which we have not been able to tie into our maps at this point. Additionally, five bear mandibles have been found in features. Features 154, 158, 211, 221, and 228 each have a bear mandible. Three of these features are found within the longhouse area. Two features are found to the west – one in a midden perhaps associated with a small building and the other in a feature to the west of that midden.

Several artifact types are relatively ubiquitous across the site. Chipped stone, aboriginal ceramics, catlinite and copper fragments are all found in a large portion of features across the length of the site. With the separation of some of these categories, such as worked bone, into more salient, functional categories (e.g. harpoons, combs, etc…), spatial patterning may become evident.