Unformatted text preview: Hawkes' "ladder of inference" = Binford's Pyramid data universe sample units l i Early broad critique of archaeology as NOT anthropology... anthropology technological, social and ideological systems nonnon-arbitrary arbitrary arbitrar "No one ever excavated a kinship system" archaeologists are forced to work with the physical remains of past behavior i f tb h i archaeologists cannot study past kinship systems True: T the archaeological record minimal units
artifacts features burials ecofacts sites i regions population non- b bili ti non-probabilistic sampling li probabilistic sampling False: composite units units not rigid categories g g scale of the record sample frame l f sample size/fraction 100% sampling simple random sampling systematic sampling y p g stratified sampling
1 2 levels are connected and mutually influencing; non-material aspects of p can be reflected in material remains p past Christopher Hawkes (1954) Lewis Binford (1965)
IDEATIONAL SOCIAL TECHNOLOGICAL 3 4 militaristic ideology kinship system Hawkes' fundamental realizations... archaeological record is incomplete!!!! h l i l di i l t !!!! cracked and broken mirror with missing pieces two issues... sampling p g (discuss today) ( y) site formation processes (discuss next class) belief in the afterlife trade network 5 6 What constitutes the archaeological record? the "archaeological record" archaeological record Six primary components of the archaeological record What constitutes the archaeological record? Wh t tit t th h l i l d? What are the minimal "bits" of data archaeologists h h i i l bi d h l i use? How do archaeologists organized and manipulate those minimal bits into larger d t sets with which th i i l bit i t l data t ith hi h to understand past human behavior? Minimal
Artifact Feature F t Composite
Site Region R i Burial Ecofact
7 8 Artifact = portable object whose form is difi d in h l modified i whole, or i part b h in by human activity Feature = non-portable object modified in nonwhole or in part by human activity; cannot be removed without being altered or destroyed 9 10 Burial = special category of feature containing human (or animal remains) and sometimes associated artifacts Ecofacts = (a)biotic remains relevant to explaining human behavior
may or may not be modified by human activity floral ecofacts = seeds pollen seeds, faunal ecofacts = animal bones soils and sediments (e.g., anthrosol) (e g anthrosol) closest thing to a "time capsule" there is time capsule is... Buckland Anglo-Saxon cemetery, Dover, Kent AD 475-750 11 12 Sites = Spatial clusters of artifacts, features and/or ecofacts (i all combinations) d/ f (in ll bi i ) Region = a spatially-temporally bounded area spatiallycontaining two or more archaeological sites
geographic = well-defined geographic features well ecological = well-defined ecological community well cultural region = well-defined cultural features well often classified by function habitation; cemeteries; kill sites; ports; trading centers; quarry, workshop or resource extraction site defining site boundaries can be a major issue 13 14 not "rigid" units today's first deep theoretical statement th ti l t t t minimal vs. composite p simple vs. complex 15 16 choice of unit has to be appropriate to question!!!
region site feature/burial Gathering Archaeological Data Data scale of the archaeological record Universe = ALL of the potential archaeological data contained within a bounded area of study artifact ecofact increasing spatial & temporal scale t l l (but exceptions exist!) 17 18 Gathering Archaeological Data MAXIMUM Professor Brantingham's Data Universe The Tibetan Plateau (1.5 million km2) d universe usually a region data i ll i investigation includes all of the known and k d i f unknown d within the b data i hi h boundaries of a region Note: Archaeologists usually describe their specialty "specialty" according to their data universe
19 too little
20 Gathering Archaeological Data Gathering Archaeological Data
N Nonarbitrary bi Note: Sample Unit
a subdivision of the data universe an actual unit of investigation units i arbitrary = determined by chance time and money available! rooms investigated in a large SW Pueblo Sites investigated along roads in Tibet Areas of investigation (shaded gray) determined by walls of the rooms in the pueblo
21 22 Qinghai Lake Basin, Tibet Gathering Archaeological Data A bi Arbitrary Note: nonnon-arbitrary sample unit defined by the boundaries of the basin units i arbitrary = determined by chance Excavation of a mammoth fossil using 2x2 m arbitrary units y Why are the units arbitrary? u s a b a y?
23 24 Gathering Archaeological Data
P Population l i the Are sample units representative of the data universe? combined set of all sampled units includes only known archaeological data but used to make inferences about known AND unknown data in the data universe Only 42 of 92 rooms y investigated. Can we use data from the 42 rooms, to infer what was going on in the other 50 rooms? How do we "know" that our data from the 42 rooms accurately represent what is happening in all rooms?
25 26 Making sure that archaeological data are representati e representative Total Non-Probabilistic Non Using (e.g., Systematic Sampling) Data Gathering (100% sample) a non-random criterion to determine which nonunits are investigated (e.g., equal spacing of samples, sampling along an observed feature) archaeological record clearly represented if you go out and gather it all... Aside from time and money, money why might we not want to gather 100% of the data?
27 28 Probabilistic
U i Using (e.g., Simple Random Sampling) Stratified Sampling
of Systematic and Random Sampling randomization procedures to determine which d i ti d t d t i hi h units are investigated; removes bias in sample nit Simple Random = each unit has eq al p of being equal selected combination 29 Sampling in action... What Readings for Week 1 & 2 g is the ratio of women to men in this class? Systematic TODAY: A&S Chapter 4 sampling in action... action what problems might arise with systematic sampling? human behavior too regular? Simple Next Class: Continue A&S Chapter 4; Stein on Website random sampling in action what problems might arise with random sampling? data universe too large? need to account for known g variation?
31 32 ...
View Full Document
- Spring '07
- Archaeology, Archaeological Record, Christopher Hawkes